A daylight greenhouse structure, control system and method of growing control thereof

Abstract

This invention relates to the field of greenhouse structure technology, and discloses a solar greenhouse structure, control system, and planting control method. The solar greenhouse structure includes a greenhouse body, steel insulated doors, windproof walls, roller shutter support arms, roller shutter machine, outer insulation blanket, insulation blanket roller rod, support frame, hot water storage bags, inner insulation blanket, ventilation openings, and windproof roller shutters. The system includes a regionalized environmental perception module, an environmental stress dynamic adaptation module, a resource benefit priority configuration module, a regionalized control parameter generation module, and a regionalized control parameter adjustment module. This invention can achieve proactive predictive stress avoidance, precise regionalized regulation, and dynamic resource benefit matching in the solar greenhouse structure environment. It effectively solves the defects of existing technologies such as homogeneous regulation, passive stress response, and poor demand adaptability, significantly improving crop growth environment adaptability, stress resistance, and comprehensive planting benefits, providing reliable technical support for refined and efficient planting in facility agriculture.

Description

Technical Field

[0001] This invention relates to the field of greenhouse structure technology, and more specifically, to a solar greenhouse structure, control system, and planting control method thereof. Background Technology

[0002] As a core component of agricultural facilities, solar greenhouses can regulate environmental parameters through artificial intervention, breaking the limitations of natural climate and enabling off-season, high-quality crop cultivation. Currently, existing solar greenhouse structures have gradually achieved environmental parameter acquisition and automated control; however, the following defects or shortcomings still exist in practical applications: Lack of regionalized regulation capabilities: Traditional greenhouse structure control often adopts a unified control model for the entire greenhouse, ignoring the environmental heterogeneity caused by differences in lighting, ventilation, and heat retention within the greenhouse structure. When the same control parameters are used for the well-lit front area and the poorly lit rear area within the greenhouse structure, it is easy for some areas to have excessive parameters and others to have insufficient parameters, which can easily affect crop growth consistency and resource utilization efficiency.

[0003] The response to stress is passive and lacks coordination: Traditional greenhouse structural control is mostly a response after stress occurs, lacking the ability to predict stress risks in the future. At the same time, when facing multiple stresses such as high temperature, low light, and high humidity, setting control thresholds only for a single stress can easily lead to the problem that the relief of a single stress can cause the exacerbation of other stresses. It is impossible to achieve coordinated avoidance of multiple stresses, which can easily lead to insufficient protection of crop stress resistance.

[0004] Poor adaptability between resource benefits and user needs: The control objectives of traditional solar greenhouse structural control systems are mostly fixed to maximizing output, lacking a flexible priority configuration mechanism. This makes it difficult to meet users' differentiated needs for high quality and high returns, water and fertilizer conservation, etc., and to balance the relationship between crop output and resource consumption (water, fertilizer, electricity), thus limiting the overall planting benefits.

[0005] The generation of control parameters lacks multi-dimensional collaborative optimization: existing parameter generation only combines crop physiological needs or single environmental parameters, without systematically integrating multiple factors such as regional spatial characteristics, stress tolerance constraints, and resource benefit weights. This results in insufficient targeting of the generated control parameters, which cannot accurately adapt to the needs of crops in different regions and at different growth stages, leading to low regulation accuracy and adaptability.

[0006] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0007] In response to the problems in related technologies, this invention proposes a solar greenhouse structure, control system, and planting control method to overcome the aforementioned technical problems existing in the existing related technologies.

[0008] Therefore, the specific technical solution adopted by the present invention is as follows: According to a first aspect of the present invention, a solar greenhouse structure is provided, including a greenhouse body, a steel heat-insulating door provided at one end of the greenhouse body, a windproof wall provided at the other end of the greenhouse body, a roller shutter support arm provided on one side of the greenhouse body, and a roller shutter machine provided at one end of the roller shutter support arm and at the top of the greenhouse body. The top of the shed is equipped with an external insulation blanket, and the middle of the external insulation blanket is equipped with an insulation blanket roll rod, with one end of the insulation blanket roll rod connected to the curtain rolling machine. The greenhouse is equipped with a support frame inside, which divides the interior space into a zone with good lighting, a zone with balanced environment, and a zone with low light and heat preservation. A hot water storage bag is installed on one side of the interior of the greenhouse. The tops of the areas with good lighting, the areas with balanced environment, and the areas with low light and heat preservation are all equipped with inner insulation blankets. Ventilation openings are provided at the top of the shed, corresponding to the areas with good lighting, the areas with balanced environment, and the areas with low light and heat preservation, respectively. Windproof roller shutters are installed on one side of the ventilation openings.

[0009] According to a second aspect of the present invention, a control system for a solar greenhouse structure is provided, comprising: The regionalized environmental sensing module is used to divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. It collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data to provide differentiated data support for the generation of regionalized control parameters. The environmental stress dynamic adaptation module is used to predict the stress risk in the future preset period based on the real-time data collected by the regional environmental perception module, and to determine the stress tolerance constraints by combining the stress tolerance coefficients of crops at different growth stages, so as to achieve multi-stress collaborative avoidance. The resource efficiency priority configuration module provides an interface for selecting priority modes and setting custom weights to transform user needs into resource efficiency weights. The priority modes include high quality and high return, yield-efficiency balance, and water and fertilizer conservation priority. Different priority modes correspond to different resource efficiency weights for yield, quality, and resource consumption. The regionalized control parameter generation module is used to obtain physiological requirement data of crops at different growth stages, and combine the spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate the optimal control parameters based on the region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. The regional control parameter adjustment module is used to generate personalized parameter control commands based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat preservation zone according to the personalized parameter control commands. Among them, the regionalized environmental perception module, the environmental stress dynamic adaptation module, the resource benefit priority configuration module, the regionalized control parameter generation module, and the regionalized control parameter adjustment module work together to achieve proactive stress prediction and avoidance, dynamic resource benefit matching, and coordinated precise regulation, thereby improving crop growth environment adaptability, stress resistance, and overall planting benefits.

[0010] Furthermore, the regionalized environmental sensing module, when dividing the greenhouse structure into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone, collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data, includes: The structure of the solar greenhouse is divided into three equal parts from front to back: front end, middle part, and rear end. Based on the actual application scenario, the front end, middle part, and rear end are marked as the area with superior lighting, the area with balanced environment, and the area with weak light and heat preservation. Sensors pre-set in different areas are used to collect data on temperature and humidity, light intensity, CO concentration, soil moisture, soil temperature, water and fertilizer application, and power consumption in each area.

[0011] Furthermore, the environmental stress dynamic adaptation module, based on real-time data collected by the regional environmental perception module, predicts the stress risk for a preset period in the future, and determines the stress tolerance constraints by combining the stress tolerance coefficients of crops at different growth stages, including: The system acquires real-time environmental parameters, soil parameters, and resource consumption data collected by the regional environmental perception module. It then performs outlier removal, missing value completion, and data standardization processing in sequence, and verifies the consistency of data timestamps and the integrity of regional coverage to ensure the accuracy and continuity of input data. Based on the verified real-time environmental parameters, soil parameters and resource consumption data, a pre-trained regionally adapted stress prediction model is used to predict the temperature, humidity, light intensity and CO concentration data of each region in the future preset time period, and obtain the environmental parameter change trend curves of each region. The trend of environmental parameter changes in each region is compared with the preset stress judgment thresholds for each region. The stress type is determined based on the comparison results. The risk level of the stress type is assessed based on the magnitude and duration of the deviation of environmental parameters from the stress judgment thresholds, and the stress risk level of each region is determined. Obtain information on the current crop growth stage and the stress tolerance coefficient of each stress type under the corresponding growth stage. Combined with the stress risk level of each region, perform quantitative calculations on each stress type to determine the initial stress tolerance constraints for each region. For scenarios where multiple stress risks exist in the same area, a collaborative optimization algorithm is used to optimize and adjust the initial stress tolerance constraints, generate the final executable set of stress tolerance constraints for each area, and clarify the parameter control thresholds and execution priorities for each stress type.

[0012] Furthermore, the changing trends of environmental parameters in each region are compared with preset stress judgment thresholds for each region. Based on the comparison results, the stress type is determined, and the risk level of the stress type is assessed according to the magnitude and duration of the environmental parameters deviating from the stress judgment threshold. The stress risk level of each region is determined as follows: The parameter values ​​of each node in the environmental parameter change trend curve of each region are compared with the corresponding stress judgment threshold one by one. The time interval in which the parameter value is continuously higher than the upper limit of the stress judgment threshold or lower than the lower limit of the stress judgment threshold within the preset time period is recorded to obtain the time interval of parameter threshold deviation for each region. Based on the parameter type threshold deviation direction, the corresponding stress type is matched for the parameter threshold deviation time interval of each region, and the stress type set of each region is obtained. The stress type set includes the stress type, the corresponding parameter, and the threshold deviation time interval. Based on the environmental parameter change trend curves and stress determination thresholds corresponding to the stress types in each region, the magnitude of environmental parameters deviating from the stress determination thresholds is calculated; based on the time intervals of parameter threshold deviations and the set of stress types in each region, the duration of continuous deviation of parameters from the corresponding thresholds for each stress type within a preset time period is statistically analyzed. The stress risk level assessment matrix is ​​used to assess the degree and duration of environmental parameters deviating from the stress judgment threshold to obtain the stress risk level of each region. The stress risk level assessment matrix uses the degree of deviation and duration as two core dimensions to classify the risk level into three levels: mild, moderate and severe.

[0013] Furthermore, information on the current crop growth stage and the stress tolerance coefficients for each stress type at the corresponding growth stage are obtained. Combined with the stress risk level of each region, quantitative calculations are performed on each stress type to determine the initial stress tolerance constraints for each region, including: The crop type information of the solar greenhouse structure is obtained through the planting management database to identify the types and growth stages of the crops planted; based on the types and growth stages of the crops planted, the stress tolerance coefficients of each stress type under the current growth stage are determined, and the stress tolerance coefficients are mapped to the stress types. Using the stress risk level and stress tolerance coefficient of each region as input variables and the stress judgment threshold as output variable, an initial stress tolerance constraint quantification model is constructed. Using the initial stress tolerance constraint quantification model, combined with the parameter benchmark value, risk level correction coefficient and deviation range reference value of each region, the initial constraint threshold of stress type is calculated. The initial constraint thresholds are optimized and adjusted based on the environmental heterogeneity characteristics of each region to ensure that the constraints are adapted to the actual environment of the region. The optimized initial constraint thresholds are then classified and organized according to the region type to obtain the initial stress tolerance constraint set for each region.

[0014] Furthermore, the regionalized control parameter generation module, when acquiring physiological requirement data of crops at different growth stages and combining spatial characteristics, resource benefit weights, and stress tolerance constraints of different regions, constructs a multi-objective optimization model and automatically generates optimal control parameters based on region and growth stage to better adapt to the planting needs of crops in different regions and growth stages, includes: Acquire physiological requirements data of crops at different growth stages, spatial characteristic data of different regions, resource benefit weight data, and stress tolerance constraint data, and perform integrity verification and standardization processing on each type of data in turn; The optimization objective of the multi-objective optimization model is determined based on the resource benefit weight data. Based on the optimization objective and the constraint boundary, the multi-objective optimization model is constructed and the decision variables are defined. The decision variables include the temperature setpoint, light intensity adjustment value, CO concentration regulation value, soil moisture control threshold, and water and fertilizer supply rate for each region. The multi-objective optimization model is iteratively solved using optimization algorithms to obtain the Pareto optimal solution. The feasibility and adaptability of the Pareto optimal solution are then verified. The verified Pareto optimal solution is used as the optimal control parameter based on the region and growth stage.

[0015] Furthermore, based on resource benefit weight data, the optimization objectives of the multi-objective optimization model are determined to include: When the priority mode corresponding to the resource efficiency weight is high quality and high return, the optimization goal is to maximize the quality indicators, improve the output indicators, and control the resource consumption reasonably. When the priority mode corresponding to the resource benefit weight is output-benefit balance, the optimization objective is to balance output and quality and minimize resource consumption cost. When the priority mode corresponding to the resource benefit weight is water and fertilizer conservation priority, the optimization objective is to minimize water and fertilizer consumption and to ensure that the yield and quality are greater than or equal to the preset benchmark value. The constraint boundaries include the upper and lower thresholds of crop physiological requirements parameters, the control thresholds of stress tolerance constraints, and the limits of regulatory capacity determined by regional spatial characteristics.

[0016] Furthermore, iteratively solving the multi-objective optimization model using optimization algorithms includes: When the solution area is the light-advantage zone in the solar greenhouse structure, the genetic algorithm is used to iteratively solve the multi-objective optimization model. When the solution region is the environmental equilibrium zone in the solar greenhouse structure, the particle swarm optimization algorithm is used to iteratively solve the multi-objective optimization model. When the solution region is the low-light insulation zone in the solar greenhouse structure, the multi-objective optimization model is solved iteratively using a linear programming algorithm.

[0017] According to a third aspect of the present invention, a method for controlling planting in a solar greenhouse structure is provided, the method comprising the following steps: S1. Divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. Collect real-time environmental and soil parameters for different zones and simultaneously acquire resource consumption data. S2. Based on the collected real-time data, predict the stress risk in the future preset period, and combine the stress tolerance coefficient of crops at different growth stages to determine the stress tolerance constraints. S3. Using the priority mode selection and custom weight setting interface, user needs are transformed into resource benefit weights. The priority modes include high quality and high return, yield-benefit balance and water and fertilizer conservation priority. Different priority modes correspond to different resource benefit weights for yield, quality and resource consumption. S4. Obtain physiological requirements data of crops at different growth stages, and combine them with spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate optimal control parameters based on region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. S5. Generate personalized parameter control instructions based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat-insulating zone according to the personalized parameter control instructions.

[0018] Compared with the prior art, the present invention provides a solar greenhouse structure, a control system and a planting control method thereof, which have the following beneficial effects: This invention, through the coordinated operation of its various modules, enables proactive and predictive stress avoidance, precise regional regulation, and dynamic resource benefit matching in the structural environment of solar greenhouses. It effectively solves the defects of existing technologies, such as homogeneous regulation, passive stress response, and poor demand adaptability. It significantly improves crop growth environment adaptability, stress resistance, and overall planting benefits, providing reliable technical support for refined and efficient planting in facility agriculture.

[0019] This invention divides the structure of a solar greenhouse into three differentiated zones: a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. It also combines zoned sensors to collect environmental, soil, and resource consumption data, breaking the limitations of traditional unified data collection for the entire greenhouse. This provides precise regional data support for subsequent modules, ensuring that subsequent control strategies can adapt to the environmental heterogeneity of each zone. It can effectively avoid the problem of traditional one-size-fits-all control, thereby improving the targeting of control and the accuracy of resource utilization.

[0020] This invention, through data preprocessing, trend prediction, risk level assessment, quantification of stress tolerance constraints, and multi-stress synergistic optimization, can achieve a shift from passively responding to stress to actively predicting and avoiding it. At the same time, it can combine the stress tolerance coefficient of crop growth stages to determine the constraints, take into account the synergistic balance of multiple stresses, effectively reduce the risk of secondary stress caused by single stress regulation, improve the crop's resistance to complex environments, ensure crop growth stability, and reduce yield and quality losses caused by stress.

[0021] This invention provides three preset priority modes and a custom weight setting interface, which can transform users' differentiated planting needs into quantifiable resource benefit weights. This allows the control system to move beyond a single output target and flexibly adapt to the planting needs of different users, achieving a dynamic balance between yield, quality, and resource consumption. It maximizes the fit with actual planting scenarios and enhances the flexibility and relevance of planting benefits.

[0022] This invention integrates crop physiological needs, regional spatial characteristics, resource benefit weights, and stress tolerance constraints to construct a multi-objective optimization model. It also adapts specific solution algorithms for different regions, generating optimal control parameters that simultaneously meet the requirements of regional adaptation, growth stage adaptation, demand adaptation, and stress avoidance adaptation. Compared with traditional single-dimensional parameter generation methods, this invention offers higher control precision and better ensures the optimal growth state of crops at each growth stage and in each region. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of a solar greenhouse structure according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the internal structure of a solar greenhouse according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the control system of a solar greenhouse structure according to an embodiment of the present invention.

[0024] In the picture: 1. Shed structure; 2. Steel insulated door; 3. Windproof wall; 4. Roller shutter support arm; 5. Roller shutter machine; 6. External insulation blanket; 7. Insulation blanket roller; 8. Support frame; 9. Area with good lighting; 10. Area with balanced environment; 11. Low-light insulation area; 12. Hot water storage bag; 13. Internal insulation blanket; 14. Ventilation vent; 15. Windproof roller shutter. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Example 1 like Figures 1-2 As shown, a solar greenhouse structure includes a greenhouse body 1, with a steel insulated door 2 at one end and a windproof wall 3 at the other end. A roller shutter support arm 4 is located on one side of the greenhouse body 1, and a roller shutter machine 5 is installed at one end of the support arm 4 and at the top of the greenhouse body 1. An external insulation blanket 6 is installed at the top of the greenhouse body 1, with an insulation blanket roller rod 7 in the middle of the blanket, one end of which is connected to the roller shutter machine 5. A support frame 8 is installed inside the greenhouse body 1, dividing the interior space into a high-light-advantage zone 9, an environmentally balanced zone 10, and a low-light insulation zone 11. A hot water storage bag 12 is installed on one side of the interior of the greenhouse 1. Inner insulation blankets 13 are installed on the top of the light-advantage zone 9, the environmental balance zone 10, and the low-light heat-preservation zone 11. Each inner insulation blanket can be extended or retracted via a steel wire rope. Ventilation vents 14 are opened at the top of the greenhouse 1, corresponding to the light-advantage zone 9, the environmental balance zone 10, and the low-light heat-preservation zone 11. A windproof roller shutter 15 (in this embodiment, the windproof roller shutter can be an electric roller shutter door) is installed on one side of each ventilation vent 14. The windproof roller shutter in each area can be extended or retracted via a motor to open or close the ventilation vent. In practical applications, sensors (not shown in the figure) are also installed inside the greenhouse structure to collect environmental and soil data within the light-advantage zone 9, the environmental balance zone 10, and the low-light heat-preservation zone 11.

[0027] In a specific application, in this embodiment, the hot water storage bag can also be a nano-coated copper-aluminum composite solar collector, namely a nano-deposited coated copper-aluminum composite solar collector. The collector plate uses a nano-black chromium coated aluminum plate. The coating consists of three layers. The first layer is an anti-reflection layer, which reduces the reflection of sunlight on the surface. The second layer is an absorption layer, which efficiently captures and converts light (sunlight) of a specific wavelength into heat energy through the resonance and interference effect of nanoparticles. The third layer is an infrared reflective layer. The bottom layer is a metal layer with high infrared reflectivity (such as aluminum or copper). Its function is to reflect any long-wave infrared rays that penetrate the absorption layer back and reabsorb them, thereby reducing the emissivity of the coating in the infrared band.

[0028] The core parameters of this solar greenhouse structure are as follows: 1) Main specifications: Standard width 12 meters (customization available); Standard length 60-120 meters (customization available); Height 6.1 meters (adjustable according to latitude).

[0029] 2) Frame system: National standard hot-dip galvanized steel, sturdy and durable.

[0030] 3) Covering system: High-transmittance, drip-free PO film with a transmittance of ≥86%.

[0031] 4) Thermal Insulation System: ① The internal thermal insulation system includes a lightweight, waterproof, folding internal thermal insulation blanket, a roller, a motor, and a blanket rolling mechanism. This thermal insulation blanket is non-absorbent, waterproof, and lightweight (approximately 300g / m³). 2 It features a lightweight and thin profile (approximately 5mm thick), and is durable, with a heat transfer coefficient of 3.35 W / (m²). 2 ·K); ② The normal service life of the inner folding insulation blanket is more than 5 years; ③ The outer insulation blanket is a waterproof insulation blanket with a heat transfer coefficient of 2.71 W / (m). 2 ·K).

[0032] 5) Heat storage system: The hot water storage bag actively stores heat during the day and releases heat evenly at night.

[0033] 6) Ventilation system: It consists of three 3-meter-wide air vents (i.e., air outlets) running the length of the system, motors, rollers, etc., and has a good ventilation and cooling effect in summer.

[0034] The greenhouse structure may also include optional systems, including: Intelligent control system: Through sensors, the greenhouse environment is monitored in real time, and combined with crop needs, the temperature, light and other parameters of the crop throughout its entire life cycle are automatically and on demand. Intelligent fertilization system: Through sensors, the system monitors the crop's growth environment and root environment in real time, and combined with crop growth models, it realizes fully automatic and precise supply of water and fertilizer on demand throughout the entire life cycle of the crop. The smart cloud platform enables functions such as equipment management, planting management, operation monitoring, statistical analysis, and product traceability. Rail transport system: for transporting production materials and products within the greenhouse; Hanging system: A hanging system specifically designed for crop production; Emergency hot air furnace: This equipment is for emergency use and is designed to prevent damage to crops in greenhouses in the event of extreme cold weather.

[0035] Example 2 like Figure 3 As shown, a control system for a solar greenhouse structure includes: The regionalized environmental sensing module is used to divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. It collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data to provide differentiated data support for the generation of regionalized control parameters. Traditional greenhouse structural control often employs a uniform control model for the entire greenhouse, neglecting the environmental heterogeneity caused by differences in lighting, ventilation, and heat retention within the greenhouse structure. When the same control parameters are applied to the well-lit front area and the low-light rear area within the greenhouse structure, some areas may have excessive parameters while others have insufficient parameters, potentially affecting crop growth uniformity and resource utilization efficiency. Therefore, this approach introduces regionalization, dividing the greenhouse structure into three differentiated zones: a high-light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. Combined with the use of zoned sensors to collect environmental, soil, and resource consumption data, this method breaks through the limitations of traditional uniform data collection across the entire greenhouse. It provides precise regional data support for subsequent modules, ensuring that subsequent control strategies can adapt to the environmental heterogeneity of each zone. This effectively avoids the one-size-fits-all problem of traditional control methods, thereby improving the targeting of control and the accuracy of resource utilization.

[0036] Specifically, the regionalized environmental sensing module, when dividing the structure of a solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone, collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data, includes: 1) Divide the structure of the greenhouse into three equal parts from front to back: front end, middle part and rear end. Mark the front end, middle part and rear end as the light-advantage zone, the environmental balance zone and the low-light heat preservation zone according to the actual application scenario. Specifically, when the actual application scenario is a single application (i.e., only one greenhouse structure), the area with the best lighting is the front 1 / 3 of the greenhouse structure, which has the longest duration and highest intensity of sunlight, but also the largest temperature fluctuations; the area with balanced environment is the middle 1 / 3 of the greenhouse structure, where the lighting, temperature, and ventilation conditions are the most stable; and the area with weak light and heat preservation is the rear 1 / 3 of the greenhouse structure, which has the shortest duration and lowest intensity of sunlight, with lower temperatures but tending to be higher humidity. When the actual application scenario involves multiple greenhouses used together (i.e., multiple greenhouse structures exist), since the front greenhouse structure affects the lighting effect of the rear greenhouse structure, the area division is reversed compared to the single application scenario. That is, the area with the best lighting is the rear 1 / 3 of the greenhouse structure; the area with balanced environment is the middle 1 / 3 of the greenhouse structure; and the area with weak light and heat preservation is the front 1 / 3 of the greenhouse structure. It should be noted that the subsequent related processing in this embodiment is based on a single application scenario.

[0037] 2) Use sensors pre-set in different areas to collect data on temperature and humidity, light intensity, CO concentration, soil moisture, soil temperature, water and fertilizer application, and power consumption in each area; Specifically, each zone independently deploys a dual-dimensional sensor group for environment and soil, with the sensor spacing consistent with the fixed node spacing of the shed (500mm). The collection frequency is set to once every 5 minutes for environmental parameters, once every 30 minutes for soil parameters, and once every 10 minutes for resource consumption data, to ensure the uniformity, real-time nature, and continuity of data collection.

[0038] The environmental stress dynamic adaptation module is used to predict the stress risk in the future preset period based on the real-time data collected by the regional environmental perception module, and to determine the stress tolerance constraints by combining the stress tolerance coefficients of crops at different growth stages, so as to achieve multi-stress collaborative avoidance. Traditional greenhouse control systems focus solely on providing suitable environmental parameters. However, the structural environment of a solar greenhouse is inherently volatile, exhibiting characteristics such as sudden high temperatures in summer, low nighttime temperatures in winter, prolonged periods of overcast skies with low light, and high humidity during the rainy season. These fluctuations create environmental stress, directly hindering crop growth, such as heat scorching, excessive vegetative growth due to low light, and disease induction due to high humidity. Therefore, this embodiment introduces an environmental stress level to upgrade the parameter control logic from passively meeting suitable conditions to actively avoiding stress risks and maximizing the utilization of environmental resources. This precisely addresses the core pain point in actual planting where parameters meet standards but crops still grow poorly.

[0039] Specifically, the environmental stress dynamic adaptation module, based on real-time data collected by the regional environmental sensing module, predicts stress risks for a preset period in the future and, in conjunction with the stress tolerance coefficients of crops at different growth stages, determines stress tolerance constraints, including: 1) Obtain real-time environmental parameters, soil parameters, and resource consumption data collected by the regional environmental perception module, and sequentially perform outlier removal, missing value completion, and data standardization processing. Verify the consistency of data timestamps and the integrity of regional coverage to ensure the accuracy and continuity of input data. 2) Based on the real-time environmental parameters, soil parameters and resource consumption data after verification, the pre-trained regional adaptive stress prediction model is used to predict the temperature, humidity, light intensity and CO concentration data of each region within a preset time period (1-3 hours in this embodiment), and obtain the environmental parameter change trend curve of each region. Specifically, the regionally adapted stress prediction model is a regional time series prediction model based on long short-term memory network (LSTM). It is specifically designed for accurate prediction of the environmental parameter change trends in different regions of a solar greenhouse structure in the next 1-3 hours. Its core feature is regional-specific modeling, which ensures that the model prediction results are highly matched with the heterogeneous characteristics of the environment in each region.

[0040] The region-adaptive stress prediction model employs a deep neural network structure consisting of an input layer, hidden layers, and an output layer. Input layer: The dimension is N×M, where N is the time step (in this embodiment, 30 minutes of continuous sensing data are taken, corresponding to 6 time steps with a time interval of 5 minutes), and M is the input feature dimension, which includes real-time environmental parameters (temperature and humidity, light intensity, CO concentration), soil parameters (soil moisture, soil temperature) and resource consumption data (water and fertilizer application, power consumption) collected by the regional environmental sensing module, for a total of 7 types of feature parameters.

[0041] Hidden Layers: Two LSTM hidden layers are configured, with the number of neurons in each layer varying according to regional characteristics—64 neurons are used in areas with high light intensity and low light intensity with good insulation, where environmental parameters fluctuate significantly; and 32 neurons are used in areas with stable environmental parameters. The hidden layers use a gating mechanism to capture long-term temporal dependencies in the data, avoiding the vanishing gradient problem.

[0042] Output layer: The dimension is T×K, where T is the prediction time step (corresponding to the next 1-3 hours, divided into 5-minute intervals, for a total of 12-36 time steps), and K is the output feature dimension, which only includes the core environmental parameters (temperature and humidity, light intensity, CO concentration) that need to be used for stress prediction, for a total of 4 types of parameters.

[0043] The region-adaptive stress prediction model adopts a region-specific independent training mode to ensure that the model for each region is adapted to its environmental characteristics. The specific training steps are as follows: Dataset Construction: Historical sensing data of the target greenhouse structure for the past 12 months was collected and divided into three independent datasets: a light-advantageous area, an environmentally balanced area, and a low-light heat-preservation area. Each dataset was further divided into training, validation, and test sets in a 7:2:1 ratio. Data Preprocessing: The same preprocessing operations (outlier removal, missing value completion, and data standardization) as those in the environmental stress dynamic adaptation module were performed on each dataset to ensure consistency between the training data and the model input data. Model Training: Using the training set data as input, the Adam optimizer was used to minimize the mean squared error (MSE) loss function between the predicted and actual values. The model performance was monitored in real time using the validation set. Training was stopped when the validation set loss no longer decreased after 10 consecutive iterations to avoid overfitting. Model Evaluation and Optimization: The trained model was evaluated using the test set. The core evaluation metrics were mean absolute error (MAE) and root mean square error (RMSE). If the evaluation metrics did not meet the preset accuracy requirements (MAE ≤ 5%, RMSE ≤ 8%), the number of hidden layer neurons or the time step was adjusted, and retraining was performed. Model deployment and online updates: The trained regional models are deployed to the system model library; during system operation, the latest perception data is automatically collected daily as an incremental dataset to fine-tune the model online, ensuring that the model's prediction accuracy continues to improve over time.

[0044] 3) Compare the changing trends of environmental parameters in each region with the preset stress judgment thresholds for each region, determine the stress type based on the comparison results, and assess the risk level of the stress type based on the magnitude and duration of the deviation of environmental parameters from the stress judgment thresholds, and determine the stress risk level of each region. Specifically, the changing trends of environmental parameters in each region are compared with preset stress judgment thresholds for each region. Based on the comparison results, the stress type is determined, and the risk level of the stress type is assessed according to the magnitude and duration of the environmental parameters deviating from the stress judgment threshold. The stress risk level of each region is determined as follows: 31) Compare the parameter values ​​of each node in the environmental parameter change trend curve of each region with the corresponding stress judgment threshold one by one, and record the time interval in which the parameter values ​​are continuously higher than the upper limit of the stress judgment threshold or lower than the lower limit of the stress judgment threshold within the preset time period to obtain the time interval of parameter threshold deviation for each region. The stress determination threshold is obtained from a pre-set stress determination threshold table. This threshold table contains the characteristic parameter determination thresholds for each region for different stress types (high temperature, low temperature, weak light, high humidity, CO deficiency, etc.). Among them, high temperature, high humidity, and CO deficiency correspond to the upper limit threshold of the parameter, while low temperature and weak light correspond to the lower limit threshold of the parameter. Specifically, the stress determination threshold table is a regionalized, stress-type-specific threshold dataset pre-stored in the system database. It serves as the core basis for determining whether environmental parameters constitute stress and the type of stress. The table adopts a two-dimensional relational structure, with the region type as the row dimension and the stress type as the column dimension. Each cell contains three core pieces of information: the corresponding feature parameter, the threshold type (upper limit / lower limit), and the specific threshold value, achieving precise binding between region, stress type, and parameter threshold.

[0045] The basic structure of the stress determination threshold table is shown in Table 1 below, where the threshold values ​​are based on the structure of a solar greenhouse for common solanaceous crops: Table 1. Thresholds for Determining Stress in Solanaceous Crops

[0046] The rules governing the correspondence between stress types and thresholds are as follows: The association rules between threshold type and stress type are as follows: For excess stress, stress caused by excessively high parameters, such as high temperature and high humidity, the upper limit threshold is used. That is, when the parameter value is continuously higher than the threshold, it is determined to be the corresponding stress. For deficiency stress, stress caused by excessively low parameters, such as low temperature, weak light, and CO deficiency, the lower limit threshold is used. That is, when the parameter value is continuously lower than the threshold, it is determined to be the corresponding stress.

[0047] The correlation rules between regional heterogeneity and threshold values: The threshold values ​​for the same stress type differ in different regions, and the pattern of these differences closely matches the regional spatial characteristics. For example, in areas with abundant daylight, the lower threshold for low-light stress is much higher than in areas with low-light insulation; conversely, in areas with strong heat retention capacity, the upper threshold for high-temperature stress is lower than in areas with abundant daylight.

[0048] Dynamic adjustability of thresholds: The stress determination threshold table supports custom adjustments based on crop type and planting season. For example, when planting cold-resistant crops, the lower limit threshold for low-temperature stress in each region can be lowered; when planting in summer, the upper limit threshold for high-temperature stress in areas with good sunlight exposure can be raised. 32) Based on the parameter type threshold deviation direction, i.e., higher than the upper limit / lower than the lower limit, match the corresponding stress type for the parameter threshold deviation time interval of each region to obtain the stress type set of each region. The stress type set includes the stress type, the corresponding parameter, and the threshold deviation time interval. For example, if the temperature parameter in the area with good lighting conditions is consistently higher than the high temperature threshold, it is determined to be high temperature stress; if the light intensity parameter in the low light insulation area is consistently lower than the low light threshold, it is determined to be low light stress; if the same parameter deviates from the threshold in different directions in different time intervals, the corresponding stress type needs to be determined separately. Multiple stress types can exist in the same area at the same time, ultimately forming a set of stress types for each area. 33) Based on the environmental parameter change trend curves and stress judgment thresholds corresponding to the stress types in each region, calculate the magnitude of the environmental parameters deviating from the stress judgment thresholds; based on the time intervals of parameter threshold deviations and the set of stress types in each region, calculate the duration of continuous deviation of parameters from the corresponding thresholds for each stress type within a preset time period. Specifically, for stresses such as high temperature, high humidity, and CO deficiency, which are judged by the upper limit threshold, the deviation range = parameter peak value - upper limit threshold; for stresses such as low temperature and weak light, which are judged by the lower limit threshold, the deviation range = lower limit threshold - parameter valley value. When calculating the duration of parameter deviation from the corresponding threshold for each stress type within a preset time period, if there are multiple discontinuous threshold deviation time intervals for the same stress type, the duration of each interval needs to be summed to obtain the total duration. 34) The stress risk level assessment matrix is ​​used to assess the magnitude and duration of environmental parameters deviating from the stress judgment threshold to obtain the stress risk level of each region. The stress risk level assessment matrix uses the magnitude and duration of deviation as two core dimensions to classify the risk level into three levels: mild, moderate, and severe. The assessment matrix for different regions has been optimized according to their environmental heterogeneity characteristics. For example, the grading standards for the magnitude and duration of deviation in the high-temperature stress assessment matrix of the daylight-advantage area are different from those in the low-light stress assessment matrix of the low-light heat preservation area. This ensures that the applied assessment matrix is ​​accurately matched with the current treatment area and stress type. Specifically, the magnitude and duration of environmental parameters deviating from the stress determination threshold are substituted into the stress risk level assessment matrix of the corresponding region and stress type. Through cross-matching of matrix dimensions, the risk level of each stress type in each region is determined. For example, the deviation of high temperature stress in the daylight advantage area is 5°C and the duration is 1.5 hours. After being substituted into the corresponding matrix, it is determined to be of medium risk.

[0049] 4) Obtain the growth stage information of the current crop and the stress tolerance coefficient of each stress type under the corresponding growth stage, and combine the stress risk level of each region to perform quantitative calculations on each stress type and determine the initial stress tolerance constraints for each region. Specifically, information on the current crop growth stage and the stress tolerance coefficients for each stress type at the corresponding growth stage are obtained. Combined with the stress risk level of each region, the stress types are quantitatively calculated to determine the initial stress tolerance constraints for each region, including: 41) Obtain crop type information of the solar greenhouse structure through the planting management database, identify the types and growth stages of the crops to be planted, including the seedling stage, flowering stage, fruiting stage, and maturity stage, and record the start time and expected duration of each growth stage; determine the stress tolerance coefficient of each stress type under the current growth stage based on the types and growth stages of the crops to be planted, and match the stress tolerance coefficient with the stress type. The stress tolerance coefficient is a quantitative indicator characterizing a crop's tolerance to different stress types at a specific growth stage. Its value ranges from 0 to 1. A lower coefficient indicates a weaker tolerance and higher sensitivity to that stress type, requiring stricter stress tolerance constraints. The same crop may exhibit different tolerance coefficients to the same stress type at different growth stages, and the same growth stage may also show differences in tolerance coefficients to different stress types. The core correspondence rules are as follows: The correlation between growth stage and tolerance coefficient: the tolerance coefficient of crops to most stress types decreases significantly and their sensitivity increases during critical growth stages; the tolerance coefficient to some stress types is relatively high during the vegetative growth stage. The relationship between stress type and tolerance coefficient: The tolerance of the same crop to different stress types at the same growth stage varies. For example, the tolerance coefficient of solanaceous crops to high temperature stress is lower than that to weak light stress during the fruiting period, so high temperature stress should be controlled first. The correlation between crop type and tolerance coefficient: Different crops have significantly different stress tolerance characteristics. For example, leafy vegetables have a higher tolerance coefficient to low light stress than solanaceous crops, while their tolerance coefficient to high humidity stress is lower than that of solanaceous crops. The following table, Table 2, shows the stress tolerance coefficients of tomatoes, a common crop grown in greenhouse structures, corresponding to different growth stages and stress types. Table 2. Stress tolerance coefficients for different growth stages and stress types.

[0050] The stress tolerance coefficient is obtained and updated as follows: Basic coefficient acquisition: The system has a built-in three-dimensional table of stress tolerance coefficients for common crops (tomatoes, cucumbers, peppers, lettuce, etc.), which users can directly retrieve based on the type of crop they are growing.

[0051] Custom coefficient entry: For special crops or new varieties, users can enter stress tolerance coefficients measured in field experiments or laboratories through the system interface to build a custom coefficient table.

[0052] Dynamic updates of the coefficient: The system can dynamically fine-tune the tolerance coefficient based on crop growth status monitoring data and stress impact results. For example, if the yield decline of tomatoes exceeds expectations after a short period of high temperature stress during the fruiting stage, the system will automatically reduce the high temperature stress tolerance coefficient for that growth stage; 42) Using the stress risk level and stress tolerance coefficient of each region as input variables and the stress judgment threshold as output variable, construct an initial stress tolerance constraint condition quantitative model, and use the initial stress tolerance constraint condition quantitative model, combined with the parameter benchmark value, risk level correction coefficient and deviation range reference value of each region, to calculate the initial constraint threshold of stress type. Specifically, the model calculation logic is as follows: For stresses such as high temperature, high humidity, and CO deficiency, with the upper limit threshold as the constraint target, the constraint threshold = regional parameter baseline value - (stress tolerance coefficient × risk level correction coefficient × deviation range reference value); for stresses such as low temperature and weak light, with the lower limit threshold as the constraint target, the constraint threshold = regional parameter baseline value + (stress tolerance coefficient × risk level correction coefficient × deviation range reference value); where the risk level correction coefficient is set to 1.0, 1.5, and 2.0 for mild, moderate, and severe risks, respectively, and the regional parameter baseline value is taken from the average regional environmental parameters in the assessment results. For example: In the area with good light exposure, the current crop is in the fruiting stage, the high temperature stress tolerance coefficient is 0.6, corresponding to a moderate high temperature stress risk level (correction coefficient 1.5), the regional temperature baseline value is 28℃, and the deviation range reference value is 5℃, then the upper limit threshold for high temperature stress constraint = 28℃ - (0.6 × 1.5 × 5℃) = 23.5℃; 43) The initial constraint thresholds are optimized and adjusted according to the environmental heterogeneity characteristics of each region to ensure that the constraint conditions are adapted to the actual environment of the region. The optimized and adjusted initial constraint thresholds are then classified and organized according to the region type to obtain the initial stress tolerance constraint condition set for each region. Specifically, the optimization and adjustment principles are as follows: the constraint threshold must not exceed the adjustable range of the environmental parameters in the area, and must not logically conflict with the constraint thresholds of other stress types in the area; for example, the calculated lower limit threshold of the weak light stress constraint in the weak light insulation area is 800 lux, but if the maximum supplementary light intensity of the equipment in the area can only reach 1000 lux, then the threshold needs to be fine-tuned to 1000 lux; the fine-tuned constraint threshold is updated to the initial stress tolerance constraint condition set to ensure that the constraint conditions are adapted to the actual environment of the area.

[0053] 5) For scenarios where there are multiple stress risks in the same area, the initial stress tolerance constraints are optimized and adjusted using a collaborative optimization algorithm to generate the final, executable set of stress tolerance constraints for each area, and the parameter control thresholds and execution priorities for each stress type are defined. Specifically, for scenarios where multiple stress risks exist in the same region, a collaborative optimization algorithm is used to balance and adjust the initially determined stress tolerance constraints, preventing the execution of a single stress constraint from increasing the risk level of other stresses. After optimization, a final, executable set of stress tolerance constraints is generated for each region, clearly defining the parameter control thresholds and execution priorities for each stress type. The parameter control thresholds provide concrete operational standards for the regionalized control parameter adjustment module, ensuring clear numerical boundaries for equipment regulation, such as temperature adjustment, supplemental lighting, and ventilation, avoiding over- or under-regulation. The execution priorities clarify the order of handling core stresses, prioritizing stresses with greater impact on crops and lower tolerance coefficients, effectively resolving regulatory conflicts when multiple stresses coexist. This set of constraints simultaneously provides implementable constraint details for the regionalized control parameter generation module, enabling efficient collaboration with subsequent modules while considering regional heterogeneity and differences in crop growth stages, ensuring the accuracy and adaptability of stress avoidance strategies.

[0054] The resource efficiency priority configuration module provides an interface for selecting priority modes and setting custom weights to transform user needs into resource efficiency weights. The priority modes include high quality and high return, yield-efficiency balance, and water and fertilizer conservation priority. Different priority modes correspond to different resource efficiency weights for yield, quality, and resource consumption. Traditional greenhouse structural control systems often have a fixed control objective of maximizing yield, lacking a flexible priority configuration mechanism. This makes it difficult to meet users' differentiated needs for high quality and high returns, as well as water and fertilizer conservation. It also makes it difficult to balance crop output with resource consumption (water, fertilizer, electricity), thus limiting overall planting benefits. Therefore, this embodiment introduces resource benefit priority to transform users' differentiated planting needs into quantifiable resource benefit weights. This allows the control system to move beyond a single output objective and flexibly adapt to the planting demands of different users, achieving a dynamic balance between yield, quality, and resource consumption. This maximizes alignment with actual planting scenarios and enhances the flexibility and relevance of planting benefits.

[0055] Specifically, the priority weights of resource benefits in this embodiment are shown in Table 3 below: Table 3 Priority Weights of Resource Benefits

[0056] The regionalized control parameter generation module is used to obtain physiological requirement data of crops at different growth stages, and combine the spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate the optimal control parameters based on the region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. Traditional system parameter generation often only considers crop physiological needs or single environmental parameters, failing to systematically integrate multiple factors such as regional spatial characteristics, stress tolerance constraints, and resource benefit weights. This results in insufficient targeting of the generated control parameters, making it impossible to accurately adapt to the needs of crops in different regions and growth stages, leading to low regulation accuracy and adaptability. Therefore, this embodiment introduces regionalized control parameters to integrate crop physiological needs, regional spatial characteristics, resource benefit weights, and stress tolerance constraints to construct a multi-objective optimization model. Specific solution algorithms are adapted for different regions, and the generated optimal control parameters simultaneously meet the requirements of regional adaptation, growth stage adaptation, demand adaptation, and stress avoidance adaptation. Compared to traditional single-dimensional parameter generation methods, this results in higher regulation accuracy and better ensures the optimal growth state of crops at each growth stage and in each region.

[0057] Specifically, the regionalized control parameter generation module acquires physiological requirement data of crops at different growth stages, and, in conjunction with the spatial characteristics of different regions, resource benefit weights, and stress tolerance constraints, constructs a multi-objective optimization model to automatically generate optimal control parameters based on region and growth stage, so as to better adapt to the planting needs of crops in different regions and growth stages. This includes: 1) Obtain physiological requirements data of crops at different growth stages, spatial characteristic data of different regions, resource benefit weight data, and stress tolerance constraint data, and perform integrity verification and standardization processing on each type of data in turn; Specifically, the system retrieves physiological requirement data for crops at different growth stages from the crop physiological characteristic database, including the suitable range and optimal thresholds for parameters such as temperature and humidity, light intensity, CO2 concentration, soil moisture, and EC value at each growth stage; it retrieves spatial characteristic data for different regions from the regionalized environmental perception module, covering region-specific characteristics such as light potential, ventilation efficiency, heat retention capacity, and water and fertilizer transport limits in areas with advantageous lighting, balanced environment, and low-light heat preservation; and it retrieves resource benefit weight data from the resource benefit priority configuration module to clarify the priority mode selected by the current user and its corresponding... The weighting of yield, quality, and resource consumption is determined; stress tolerance constraint data is retrieved from the environmental stress dynamic adaptation module, including parameter control thresholds and execution priorities for different stress types in each region; after data retrieval, the regional coverage integrity, growth stage matching, and parameter dimension consistency of each type of data are verified to ensure no data is missing or mismatched, providing a reliable data foundation for subsequent model construction; the four types of core data are standardized, and parameters with different dimensions and value ranges are uniformly converted into standardized values ​​of 0-1 to eliminate the impact of data dimension differences on model calculation; 2) Determine the optimization objective of the multi-objective optimization model based on the resource benefit weight data. Based on the optimization objective and the constraint boundary, construct the multi-objective optimization model and define the decision variables, which include the temperature setpoint, light intensity adjustment value, CO concentration regulation value, soil moisture control threshold, and water and fertilizer supply rate for each region. Specifically, when the priority mode corresponding to the resource benefit weight is high quality and high return, the optimization goal is to maximize the quality index, improve the output index, and control the resource consumption reasonably; when the priority mode corresponding to the resource benefit weight is output-benefit balance, the optimization goal is to optimize the balance between output and quality and minimize the resource consumption cost; when the priority mode corresponding to the resource benefit weight is water and fertilizer conservation, the optimization goal is to minimize water and fertilizer consumption and ensure that the output and quality are greater than or equal to the preset benchmark value. The constraint boundaries include the upper and lower thresholds of crop physiological requirements parameters, the control thresholds of stress tolerance constraints, and the limits of regulatory capacity determined by regional spatial characteristics. The objective function of the multi-objective optimization model is quantified based on the optimization objectives. For example, the core quantitative indicators are soluble solids content Q(X) (quality), number of fruits per plant Y(X) (yield), water and fertilizer application and power consumption R(X) (resource consumption). The comprehensive objective function value is calculated by combining resource benefit weights. The constraint function of the model is transformed into a set of inequalities based on hard constraints and soft constraints. For example, the temperature during the fruiting period in the light-advantage area is ≤ the upper limit of high temperature stress tolerance, the light intensity during the seedling stage in the weak light insulation area is ≥ the lower limit of crop physiological needs, and the water and fertilizer supply rate in the environmental equilibrium area is ≤ the regional transport limit. The expressions for the comprehensive objective function values ​​under different priority modes are as follows: The objective function for the priority mode, which prioritizes high quality and high returns, is as follows: the optimization objective is to maximize quality indicators while also improving output indicators and rationally controlling resource consumption; therefore, the quality weight w is... Q Significantly higher than the output weight w Y With resource consumption weight w R In this embodiment, the values ​​are taken as 0.6, 0.3, and 0.1 respectively, and the objective function is to maximize the comprehensive objective value, expressed as: maxF1(X)=w Q Q(X)+w Y Y(X)+w R R(X); The priority mode is the objective function for output-efficiency balance: the optimization objective is to achieve a balance between output and quality and to minimize resource consumption costs; therefore, the quality weight w is... Q With production weight w Y Approximate, resource consumption weight w RSlightly higher than the high-quality, high-yield model, in this embodiment the values ​​are taken as 0.4, 0.4, and 0.2 respectively. The objective function is to maximize the comprehensive objective value, and the expression is: maxF2(X)=w Q Q(X)+w Y Y(X)+w R R(X); The objective function for the priority mode, which prioritizes water and fertilizer conservation, is to minimize water and fertilizer consumption and ensure that yield and quality are greater than or equal to preset benchmark values. Therefore, the sub-weight α for water and fertilizer consumption is significantly higher than the sub-weight β for electricity consumption. In this embodiment, α and β are set to 0.8 and 0.2, respectively. The overall objective function focuses on minimizing resource consumption, achieved by maximizing the inversely standardized R(X). The benchmark requirements for yield and quality are reflected through constraints (rather than the objective function). The objective function expression is as follows: maxF3(X)=w R R(X); 3) The multi-objective optimization model is solved iteratively using optimization algorithms. During the solution process, the algorithm will continuously adjust the values ​​of decision variables, calculate the corresponding objective function values ​​and constraint satisfaction, until the preset convergence conditions are met, such as the objective function value change being less than a threshold for N consecutive iterations. One or more Pareto optimal solutions are output, and the feasibility and adaptability of the Pareto optimal solutions are verified. The verified Pareto optimal solutions are used as the optimal control parameters based on the region and growth stage. Specifically, when the solution area is the light-advantage zone within the solar greenhouse structure, a genetic algorithm is used to iteratively solve the multi-objective optimization model. This is because the genetic algorithm, with its greater robustness, is chosen to address the large fluctuations in environmental parameters within the light-advantage zone. When the solution area is the environmental equilibrium zone within the solar greenhouse structure, a particle swarm optimization algorithm is used to iteratively solve the multi-objective optimization model. This is because the environmental equilibrium zone requires high parameter stability, and the particle swarm optimization algorithm, with its faster convergence speed, is chosen. When the solution area is the low-light insulation zone within the solar greenhouse structure, a linear programming algorithm is used to iteratively solve the multi-objective optimization model. This is because the low-light insulation zone has fewer controllable variables, and the simple and efficient linear programming algorithm is chosen. Simultaneously, based on the priority of the current optimization objective, the core parameters of the algorithm, such as the number of iterations, convergence accuracy, crossover probability, and mutation probability, are configured to ensure that the algorithm's solution process is highly matched with the regional characteristics, growth stage requirements, and resource efficiency objectives.

[0058] The regional control parameter adjustment module is used to generate personalized parameter control commands based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat preservation zone according to the personalized parameter control commands. Among them, the regionalized environmental perception module, the environmental stress dynamic adaptation module, the resource benefit priority configuration module, the regionalized control parameter generation module, and the regionalized control parameter adjustment module work together to achieve proactive stress prediction and avoidance, dynamic resource benefit matching, and coordinated precise regulation, thereby improving crop growth environment adaptability, stress resistance, and overall planting benefits.

[0059] Example 3 A method for controlling planting in a solar greenhouse structure, the method comprising the following steps: S1. Divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. Collect real-time environmental and soil parameters for different zones and simultaneously acquire resource consumption data. S2. Based on the collected real-time data, predict the stress risk in the future preset period, and combine the stress tolerance coefficient of crops at different growth stages to determine the stress tolerance constraints. S3. Using the priority mode selection and custom weight setting interface, user needs are transformed into resource benefit weights. The priority modes include high quality and high return, yield-benefit balance and water and fertilizer conservation priority. Different priority modes correspond to different resource benefit weights for yield, quality and resource consumption. S4. Obtain physiological requirements data of crops at different growth stages, and combine them with spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate optimal control parameters based on region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. S5. Generate personalized parameter control instructions based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat-insulating zone according to the personalized parameter control instructions.

[0060] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium. When the program is executed, it includes the steps described in the above methods. The storage medium may be, for example, ROM / RAM, magnetic disk, optical disk, etc.

[0061] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A solar greenhouse structure, characterized in that, The shed includes a shed body, one end of which is equipped with a steel insulated door, and the other end of which is equipped with a windproof wall. A roller shutter support arm is provided on one side of the shed body, and a roller shutter machine is provided at one end of the roller shutter support arm and at the top of the shed body. The top of the shed is provided with an external insulation blanket, and the middle of the external insulation blanket is provided with an insulation blanket roll rod, and one end of the insulation blanket roll rod is connected to the curtain rolling machine. The shed is equipped with a support frame inside, which divides the interior space into a light-advantage zone, an environmentally balanced zone, and a low-light heat-insulating zone. A hot water storage bag is installed on one side of the interior of the shed. The tops of the light-advantage zone, the environmental balance zone, and the low-light heat-insulating zone are all equipped with inner insulation blankets. Ventilation openings are provided at the top of the shed and at positions corresponding to the light-advantage zone, the environmental balance zone, and the low-light heat-insulating zone, respectively. A windproof roller shutter is provided on one side of each ventilation opening.

2. A control system for a solar greenhouse structure, used to control the solar greenhouse structure as described in claim 1, characterized in that, include: The regionalized environmental sensing module is used to divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. It collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data to provide differentiated data support for the generation of regionalized control parameters. The environmental stress dynamic adaptation module is used to predict the stress risk in the future preset period based on the real-time data collected by the regional environmental perception module, and to determine the stress tolerance constraints by combining the stress tolerance coefficients of crops at different growth stages, so as to achieve multi-stress collaborative avoidance. The resource efficiency priority configuration module provides an interface for selecting priority modes and setting custom weights to transform user needs into resource efficiency weights. The priority modes include high quality and high return, yield-efficiency balance, and water and fertilizer conservation priority. Different priority modes correspond to different resource efficiency weights for yield, quality, and resource consumption. The regionalized control parameter generation module is used to obtain physiological requirement data of crops at different growth stages, and combine the spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate the optimal control parameters based on the region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. The regional control parameter adjustment module is used to generate personalized parameter control commands based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat preservation zone according to the personalized parameter control commands. Among them, the regionalized environmental perception module, the environmental stress dynamic adaptation module, the resource benefit priority configuration module, the regionalized control parameter generation module, and the regionalized control parameter adjustment module work together to achieve proactive stress prediction and avoidance, dynamic resource benefit matching, and coordinated precise regulation, thereby improving crop growth environment adaptability, stress resistance, and overall planting benefits.

3. The control system for a solar greenhouse structure according to claim 2, characterized in that, The regionalized environmental sensing module, when dividing the solar greenhouse structure into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone, collects real-time environmental and soil parameters for different zones and simultaneously acquires resource consumption data, includes: The structure of the solar greenhouse is divided into three equal parts from front to back: front end, middle part, and rear end. Based on the actual application scenario, the front end, middle part, and rear end are marked as the area with superior lighting, the area with balanced environment, and the area with weak light and heat preservation. Sensors pre-set in different areas are used to collect data on temperature and humidity, light intensity, CO2 concentration, soil moisture, soil temperature, water and fertilizer application, and power consumption in each area.

4. The control system for a solar greenhouse structure according to claim 2, characterized in that, The environmental stress dynamic adaptation module, based on real-time data collected by the regional environmental sensing module, predicts stress risks for a preset period in the future and, in conjunction with the stress tolerance coefficients of crops at different growth stages, determines stress tolerance constraints, including: The system acquires real-time environmental parameters, soil parameters, and resource consumption data collected by the regional environmental perception module. It then performs outlier removal, missing value completion, and data standardization processing in sequence, and verifies the consistency of data timestamps and the integrity of regional coverage to ensure the accuracy and continuity of input data. Based on the verified real-time environmental parameters, soil parameters and resource consumption data, a pre-trained regional adaptive stress prediction model is used to predict the temperature, humidity, light intensity and CO2 concentration data of each region in the future preset time period, and obtain the environmental parameter change trend curves of each region. The trend of environmental parameter changes in each region is compared with the preset stress judgment thresholds for each region. The stress type is determined based on the comparison results. The risk level of the stress type is assessed based on the magnitude and duration of the deviation of environmental parameters from the stress judgment thresholds, and the stress risk level of each region is determined. Obtain information on the current crop growth stage and the stress tolerance coefficient of each stress type under the corresponding growth stage. Combined with the stress risk level of each region, perform quantitative calculations on each stress type to determine the initial stress tolerance constraints for each region. For scenarios where multiple stress risks exist in the same area, a collaborative optimization algorithm is used to optimize and adjust the initial stress tolerance constraints, generate the final executable set of stress tolerance constraints for each area, and clarify the parameter control thresholds and execution priorities for each stress type.

5. The control system for a solar greenhouse structure according to claim 4, characterized in that, The process involves comparing the changing trends of environmental parameters in each region with preset stress thresholds for each region, determining the stress type based on the comparison results, and assessing the risk level of the stress type based on the magnitude and duration of the environmental parameters deviating from the stress thresholds. The determination of the stress risk level for each region includes: The parameter values ​​of each node in the environmental parameter change trend curve of each region are compared with the corresponding stress judgment threshold one by one. The time interval in which the parameter value is continuously higher than the upper limit of the stress judgment threshold or lower than the lower limit of the stress judgment threshold within the preset time period is recorded to obtain the time interval of parameter threshold deviation for each region. Based on the parameter type threshold deviation direction, the corresponding stress type is matched for the parameter threshold deviation time interval of each region, and the stress type set of each region is obtained. The stress type set includes the stress type, the corresponding parameter, and the threshold deviation time interval. Based on the environmental parameter change trend curves and stress determination thresholds corresponding to the stress types in each region, the magnitude of environmental parameters deviating from the stress determination thresholds is calculated; based on the time intervals of parameter threshold deviations and the set of stress types in each region, the duration of continuous deviation of parameters from the corresponding thresholds for each stress type within a preset time period is statistically analyzed. The stress risk level assessment matrix is ​​used to assess the degree and duration of environmental parameters deviating from the stress judgment threshold to obtain the stress risk level of each region. The stress risk level assessment matrix uses the degree of deviation and duration as two core dimensions to classify the risk level into three levels: mild, moderate and severe.

6. The control system for a solar greenhouse structure according to claim 4, characterized in that, The process of obtaining the growth stage information of the currently planted crop and the stress tolerance coefficient of each stress type under the corresponding growth stage, and combining it with the stress risk level of each region, to quantitatively calculate each stress type and determine the initial stress tolerance constraints for each region includes: The crop type information of the solar greenhouse structure is obtained through the planting management database to identify the types and growth stages of the crops planted; based on the types and growth stages of the crops planted, the stress tolerance coefficients of each stress type under the current growth stage are determined, and the stress tolerance coefficients are mapped to the stress types. Using the stress risk level and stress tolerance coefficient of each region as input variables and the stress judgment threshold as output variable, an initial stress tolerance constraint quantification model is constructed. Using the initial stress tolerance constraint quantification model, combined with the parameter benchmark value, risk level correction coefficient and deviation range reference value of each region, the initial constraint threshold of stress type is calculated. The initial constraint thresholds are optimized and adjusted based on the environmental heterogeneity characteristics of each region to ensure that the constraints are adapted to the actual environment of the region. The optimized initial constraint thresholds are then classified and organized according to the region type to obtain the initial stress tolerance constraint set for each region.

7. The control system for a solar greenhouse structure according to claim 2, characterized in that, The regionalized control parameter generation module acquires physiological requirement data of crops at different growth stages, and, in conjunction with the spatial characteristics, resource benefit weights, and stress tolerance constraints of different regions, constructs a multi-objective optimization model to automatically generate optimal control parameters based on region and growth stage, so as to better adapt to the planting needs of crops at different regions and growth stages. This includes: Acquire physiological requirements data of crops at different growth stages, spatial characteristic data of different regions, resource benefit weight data, and stress tolerance constraint data, and perform integrity verification and standardization processing on each type of data in turn; The optimization objective of the multi-objective optimization model is determined based on the resource benefit weight data. Based on the optimization objective and the constraint boundary, the multi-objective optimization model is constructed and the decision variables are defined. The decision variables include the temperature setpoint, light intensity adjustment value, CO2 concentration regulation value, soil moisture control threshold, and water and fertilizer supply rate for each region. The multi-objective optimization model is iteratively solved using optimization algorithms to obtain the Pareto optimal solution. The feasibility and adaptability of the Pareto optimal solution are then verified. The verified Pareto optimal solution is used as the optimal control parameter based on the region and growth stage.

8. The control system for a solar greenhouse structure according to claim 7, characterized in that, The optimization objectives of the multi-objective optimization model determined based on resource benefit weight data include: When the priority mode corresponding to the resource efficiency weight is high quality and high return, the optimization goal is to maximize the quality indicators, improve the output indicators, and control the resource consumption reasonably. When the priority mode corresponding to the resource benefit weight is output-benefit balance, the optimization objective is to balance output and quality and minimize resource consumption cost. When the priority mode corresponding to the resource benefit weight is water and fertilizer conservation priority, the optimization objective is to minimize water and fertilizer consumption and to ensure that the yield and quality are greater than or equal to the preset benchmark value. The constraint boundaries include the upper and lower thresholds of crop physiological requirement parameters, the control threshold of stress tolerance constraints, and the limit of regulatory capacity determined by regional spatial characteristics.

9. The control system for a solar greenhouse structure according to claim 7, characterized in that, The iterative solution of the multi-objective optimization model using optimization algorithms includes: When the solution area is the light-advantage zone in the solar greenhouse structure, the genetic algorithm is used to iteratively solve the multi-objective optimization model. When the solution region is the environmental equilibrium zone in the solar greenhouse structure, the particle swarm optimization algorithm is used to iteratively solve the multi-objective optimization model. When the solution area is the low-light insulation zone in the solar greenhouse structure, the multi-objective optimization model is solved iteratively using a linear programming algorithm.

10. A planting control method for a solar greenhouse structure, implemented based on a control system for a solar greenhouse structure as described in any one of claims 2-9, characterized in that, The method includes the following steps: S1. Divide the structure of the solar greenhouse into a light-advantage zone, an environmentally balanced zone, and a low-light heat-preservation zone. Collect real-time environmental and soil parameters for different zones and simultaneously acquire resource consumption data. S2. Based on the collected real-time data, predict the stress risk in the future preset period, and combine the stress tolerance coefficient of crops at different growth stages to determine the stress tolerance constraints. S3. Using the priority mode selection and custom weight setting interface, user needs are transformed into resource benefit weights. The priority modes include high quality and high return, yield-benefit balance and water and fertilizer conservation priority. Different priority modes correspond to different resource benefit weights for yield, quality and resource consumption. S4. Obtain physiological requirements data of crops at different growth stages, and combine them with spatial characteristics, resource benefit weights and stress tolerance constraints of different regions to construct a multi-objective optimization model and automatically generate optimal control parameters based on region and growth stage to better adapt to the planting needs of crops in different regions and growth stages. S5. Generate personalized parameter control instructions based on the optimal control parameters, and adjust the environmental and soil parameters of the light-advantage zone, the environmental balance zone, and the low-light heat-insulating zone according to the personalized parameter control instructions.

Images