LED light source-based and spectrum-based and carbon dioxide compensation-based intelligent management method and system for agricultural greenhouse

Abstract

The application relates to an intelligent management method and system for an agricultural greenhouse based on an LED light source and carbon dioxide compensation, which comprises the following steps: acquiring indoor environment information and plant physiological information of the greenhouse and inputting the information into a photosynthesis-environment coupling model; matching a compensation strategy and determining compensation parameters based on simulation information output by the photosynthesis-environment coupling model; acquiring plant state information and adjusting the compensation parameters; the application realizes integrated linkage regulation and control of key environment parameters such as light and CO2 concentration by collecting environment parameters and plant physiological indexes, dynamically simulating and analyzing in combination with a photosynthesis simulation model, intelligently matching a compensation strategy and compensation parameters, effectively solves the problem of low resource utilization efficiency caused by traditional single-factor independent control, and can realize dynamic feedback and response according to the actual physiological state of crops, and has the effect of significantly improving the timeliness and accuracy of environment regulation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural greenhouse management, and in particular to an agricultural greenhouse intelligent management method and system based on LED light source and spectrum and carbon dioxide compensation. BACKGROUND

[0002] In modern facility agriculture, the environmental regulation technology of greenhouse is the key to improve crop yield and quality. The current greenhouse environment control system mainly adopts single factor independent regulation mode, that is, the temperature, humidity, light, CO2 concentration and other environmental parameters are controlled respectively, and this regulation mode has obvious technical limitations.

[0003] In actual production process, plant photosynthesis is a complex physiological process influenced by multiple factors, and there is significant interaction between environmental factors. For example, the increase of light intensity needs to be matched with appropriate CO2 concentration to fully exert the light energy utilization efficiency, and temperature change will directly affect plant stomatal conductance, and then affect the CO2 absorption efficiency.

[0004] However, the existing greenhouse environment control technology often treats these related factors separately, resulting in poor regulation effect. The specific performance is: when the light is enhanced, the CO2 concentration cannot be adjusted in time, causing waste of light energy; when the temperature rises, the light intensity is not reduced accordingly, aggravating the heat stress of plants; when the humidity changes, the ventilation strategy cannot be adjusted, affecting the transpiration of crops. In addition, the existing greenhouse environment control technology mostly adopts fixed threshold switch control mode, lacks the sensing ability of plant real-time physiological state, and is difficult to dynamically adjust the environmental parameters according to the actual photosynthetic efficiency of crops. This passive regulation mode not only causes energy waste, but also may cause the growth of crops to be hindered. SUMMARY

[0005] In order to solve the above-mentioned defects and realize dynamic optimization management based on the actual physiological needs of plants, the present application provides an agricultural greenhouse intelligent management method and system based on LED light source and spectrum and carbon dioxide compensation.

[0006] The above-mentioned invention purpose of the present application is realized by the following technical scheme:

[0007] An agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation, comprising the steps of:

[0008] acquiring indoor environmental information and plant physiological information of the greenhouse through a collection terminal;

[0009] inputting the indoor environmental information and plant physiological information into a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response;

[0010] The simulation information output by the photosynthesis-environment coupling model is matched with a preset corresponding compensation strategy, and a compensation parameter is determined, the compensation strategy including LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation.

[0011] The plant state information is obtained through monitoring, and the corresponding compensation parameter in the matched compensation strategy is adjusted based on the plant state information.

[0012] By adopting the technical scheme, the application constructs an intelligent management method of an agricultural greenhouse based on LED light sources and spectrum and carbon dioxide compensation, and realizes multi-factor collaborative regulation of a greenhouse environment. First, indoor environment information and plant physiological information of the greenhouse are collected by a collection terminal, and the obtained indoor environment information and plant physiological information are input into a pre-trained photosynthesis-environment coupling model for photosynthesis simulation based on environmental response. The photosynthesis-environment coupling model can simulate photosynthesis response characteristics under different environmental combinations and output simulation information. Based on the simulation information output by the model, a compensation strategy is matched, including dynamically adjusting the spectrum ratio of the LED light source (such as the ratio of red light to blue light), controlling the gradient compensation rate of CO2, and optimizing the frequency and duty cycle of pulse light compensation. Finally, through a closed-loop feedback mechanism, plant state changes are monitored and compensation parameters are dynamically corrected to ensure that crops are in an optimal growth state for a long time. The application collects environmental parameters and plant physiological indicators, dynamically simulates and analyzes them in combination with a pre-trained photosynthesis simulation model, intelligently matches the optimal LED spectrum ratio, CO2 compensation strategy, and pulse light compensation parameters, realizes integrated linkage regulation of key environmental parameters such as light and CO2 concentration, effectively solves the problem of low resource utilization efficiency caused by traditional single-factor independent control, and has the effect of significantly improving the timeliness and accuracy of environmental regulation compared with the traditional fixed threshold control method.

[0013] In a preferred example, the application can be further configured such that the photosynthesis-environment coupling model includes a data processing layer, a correlation analysis layer, and an information output layer, and the step of inputting the indoor environment information and the plant physiological information into the pre-trained photosynthesis-environment coupling model for photosynthesis simulation based on environmental response includes the following steps:

[0014] The data processing layer calculates a photosynthetic efficiency index and a water stress coefficient based on the received indoor environment information and plant physiological information;

[0015] The correlation analysis layer performs correlation analysis on the photosynthetic efficiency index and the indoor environment information, identifies the limiting factor of photosynthesis, and corrects the weight of the limiting factor in combination with the water stress coefficient to generate a weight vector of the limiting factor;

[0016] The information output layer identifies a dominant limiting factor based on a weight vector of the limiting factors, and generates simulation information for matching a compensation strategy based on the dominant limiting factor and the weight vector thereof.

[0017] By adopting the technical solution, the present application realizes intelligent diagnosis and optimal regulation and control of a plant growth environment by constructing a photosynthesis-environment coupling model comprising a data processing layer, a correlation analysis layer and an information output layer. The data processing layer first performs standardized processing on collected indoor environment information and plant physiological information, calculates a photosynthetic efficiency index reflecting photosynthetic efficiency of plants and a water stress coefficient representing water conditions, and provides a quantitative basis for subsequent analysis. The correlation analysis layer performs correlation analysis on the photosynthetic efficiency index and the indoor environment information, thereby identifying photosynthesis limiting factors, and dynamically revising the influence weight of each limiting factor in combination with the water stress degree, to generate a weight vector of the limiting factors. The information output layer identifies a dominant limiting factor based on the weight vector of the limiting factors, and generates a simulation information package comprising compensation strategy suggestions. The present application realizes analysis of the relationship between the environment and plant physiology by constructing a layered and progressive photosynthesis-environment coupling model processing architecture, accurately identifies key limiting factors of plant photosynthesis, and provides a basis for subsequent compensation strategy formulation, thereby reducing the subjectivity and hysteresis of traditional experience-based judgment, and improving the precision and timeliness of environmental regulation.

[0018] In a preferred example, the present application can be further configured as follows: the correlation analysis layer comprises a matrix construction sub-layer, a factor identification sub-layer and a dynamic revision sub-layer. The correlation analysis layer performs correlation analysis on the photosynthetic efficiency index and the indoor environment information, identifies limiting factors of photosynthesis, and revises the weight of the limiting factors in combination with the water stress coefficient, to generate a weight vector of the limiting factors. The steps include:

[0019] The matrix construction sub-layer constructs a correlation degree matrix based on the indoor environment information and the photosynthetic efficiency index.

[0020] The factor identification sub-layer identifies limiting factors affecting photosynthesis in the correlation degree matrix based on the environment, and calculates the initial weight of each limiting factor based on the correlation degree matrix.

[0021] The dynamic revision sub-layer revises the initial weight of the limiting factors in combination with the water stress coefficient, to generate a weight vector of the limiting factors.

[0022] By adopting the technical scheme, the present application realizes accurate identification and dynamic correction of photosynthesis limiting factors by refining the correlation analysis layer into a three-layer structure of a matrix construction sub-layer, a factor identification sub-layer and a dynamic correction sub-layer; the matrix construction sub-layer constructs a correlation degree matrix of indoor environment information and photosynthetic efficiency indexes, quantifies the influence degree of each environmental factor on photosynthesis, and provides a data basis for subsequent analysis; the factor identification sub-layer extracts limiting factors based on environmental influence on photosynthesis from the correlation degree matrix, and calculates initial weights of each limiting factor based on the correlation degree matrix; the dynamic correction sub-layer introduces a water stress coefficient as a regulating variable to adaptively correct the initial weights, and generates a limiting factor weight vector reflecting the actual environmental stress condition; the present application realizes multi-dimensional analysis and evaluation of photosynthetic limiting factors through the three-layer architecture analysis process, can effectively identify photosynthetic limiting factors under different environmental combinations, provides a basis for environmental regulation, overcomes the one-sidedness of traditional single index judgment and the limitation of static weight distribution, and has the effect of improving the authenticity and reliability of photosynthesis simulation.

[0023] In a preferred example, the present application can be further configured as: the information output layer includes a model construction sub-layer, a weight comparison sub-layer and an information generation sub-layer, and the steps of identifying a dominant limiting factor based on the weight vector of the limiting factor and generating simulation information for matching a compensation strategy based on the dominant limiting factor and its weight vector by the information output layer, include steps of:

[0024] The model construction sub-layer acquires plant variety information and its corresponding initial weight threshold value, and constructs a threshold adjustment model to dynamically correct the initial weight threshold value;

[0025] The weight comparison sub-layer compares the weight vector of the limiting factor with the dynamic threshold value dynamically corrected by the threshold adjustment model corresponding to the limiting factor, so as to identify the dominant limiting factor;

[0026] The information generation sub-layer generates simulation information for matching a compensation strategy based on the dominant limiting factor.

[0027] By adopting the technical solution, the information output layer is refined into a three-layer structure of a model construction sub-layer, a weight comparison sub-layer and an information generation sub-layer, intelligent identification of a dominant limiting factor and accurate generation of simulation information of a compensation strategy are realized, the model construction sub-layer constructs a threshold adjustment model with self-adaptive ability by obtaining plant variety information and its corresponding initial weight threshold, the threshold adjustment model can dynamically correct the initial weight threshold of each limiting factor, the weight vector of the limiting factor is compared with the dynamic threshold adjusted by the threshold adjustment model, the current dominant limiting factor is accurately identified, and the information generation sub-layer generates a simulation information package containing compensation strategy suggestions and compensation control parameters based on the dominant limiting factor. The application realizes accurate diagnosis and control of plant growth and photosynthesis limiting factors by constructing a progressive three-layer information processing structure, can provide optimal compensation scheme simulation suggestion information for different plant crop varieties and growth environments, and has the effects of improving the accuracy and applicability of environmental control.

[0028] In a preferred example, the application can be further configured to: the simulation information output based on the photosynthesis-environment coupling model is matched with a preset corresponding compensation strategy, and a compensation parameter is determined, the compensation strategy including the steps of LED spectrum adjustment, CO2 gradient compensation and pulsed light compensation.

[0029] A pre-constructed compensation strategy knowledge base is matched with a compensation strategy for simulation information;

[0030] A parameter optimization algorithm corresponding to the preset compensation strategy is matched based on the compensation strategy, and an optimization coefficient corresponding to the parameter optimization algorithm is obtained;

[0031] The compensation parameter is determined based on the parameter optimization algorithm and the optimization coefficient, and the compensation parameter includes a spectrum compensation amount, a CO2 compensation flux and a pulse parameter.

[0032] By adopting the above technical solution, a collaborative mechanism of the compensation strategy knowledge base and the parameter optimization algorithm is established, and intelligent optimization of photosynthesis environmental compensation parameters is realized. First, the optimal or most suitable compensation strategy is matched for simulation information based on the pre-constructed compensation strategy knowledge base. The parameter optimization algorithm corresponding to the preset compensation strategy is matched based on the compensation strategy, and the optimization coefficient corresponding to the parameter optimization algorithm is obtained. Finally, the compensation parameter is determined based on the parameter optimization algorithm and the optimization coefficient, and the compensation parameter includes a spectrum compensation amount, a CO2 compensation flux and a pulse parameter. The application realizes accurate matching of the compensation strategy and optimization calculation of the compensation parameter by combining the knowledge base and the optimization algorithm, can dynamically adjust the compensation scheme according to environmental changes and plant needs, effectively solves the problem of insufficient adaptability of traditional experience-based control methods, and has the effects of improving the accuracy and reliability of environmental compensation.

[0033] The application can be further configured in a preferred example as follows: the step of matching the compensation strategy based on the compensation strategy to the preset corresponding parameter optimization algorithm and obtaining the optimization coefficient corresponding to the parameter optimization algorithm comprises the steps of:

[0034] If several compensation strategies are matched at the same time, the intensity coefficient of different compensation strategies is determined based on the weight vector of the dominant limiting factor;

[0035] The current stress level of the dominant limiting factor is determined, and the multi-objective collaborative strategy is determined based on the intensity coefficient of the compensation strategy and the current stress level of the dominant limiting factor.

[0036] By adopting the above technical solution, a multi-compensation strategy collaborative optimization mechanism is established to realize intelligent regulation and control decision-making under complex environmental conditions. When multiple environmental factors that need to be compensated are identified at the same time, the intensity coefficient of each compensation strategy is calculated according to the weight vector of the dominant limiting factor, and the stress level of the dominant limiting factor caused by the current environmental factor or plant physiological factor is evaluated. Finally, the intensity coefficient of each compensation strategy and the stress level are balanced based on a multi-objective optimization algorithm to generate a multi-objective collaborative strategy that comprehensively considers the improvement of photosynthetic efficiency, the optimization of resource utilization, and the control of energy consumption. The application realizes the formulation of multi-environmental factor collaborative compensation decision-making through dynamic weight distribution and multi-objective optimization method, can intelligently adjust the coordination mode and implementation intensity of each compensation measure according to the actual stress degree, and effectively solves the resource conflict problem when multiple compensation strategies are implemented in parallel, thereby improving the overall regulation and control effect under complex environmental conditions.

[0037] The application can be further configured in a preferred example as follows: the step of obtaining plant state information through the monitoring terminal and adjusting the corresponding compensation parameter in the matched compensation strategy based on the plant state information comprises the steps of:

[0038] Determining the stress state of the plant based on the plant state information;

[0039] Adjusting the corresponding compensation parameter in the matched compensation strategy based on the stress state of the plant.

[0040] By adopting the above technical solution, a dynamic feedback regulation mechanism based on the physiological state of the plant is established to realize adaptive optimization of environmental regulation parameters. The plant state information is obtained in real time through the monitoring terminal, and the stress state type and degree of the plant are determined. Then, the corresponding compensation parameter in the matched compensation strategy is adjusted according to the stress state of the plant. The application realizes the mapping from environmental parameter control to actual plant response through the closed-loop regulation mode driven by the physiological state, can dynamically optimize the regulation parameters according to the real physiological needs of the plant, effectively avoids the problem of disconnection between environmental parameters and plant response in open-loop control, and has the effect of improving the biological effectiveness and resource utilization efficiency of environmental regulation.

[0041] The second application object of the present application is achieved by the following technical scheme.

[0042] An agricultural greenhouse intelligent management system based on LED light source and spectrum and carbon dioxide compensation, comprising:

[0043] An information acquisition module is configured to acquire indoor environment information and plant physiological information of the greenhouse through a collection terminal.

[0044] An input simulation module is configured to input the indoor environment information and the plant physiological information into a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response.

[0045] A strategy matching module is configured to match a preset corresponding compensation strategy based on simulation information output by the photosynthesis-environment coupling model and determine compensation parameters, wherein the compensation strategy includes LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation.

[0046] A parameter adjustment module is configured to acquire plant state information through a monitoring terminal and adjust corresponding compensation parameters in the matched compensation strategy based on the plant state information.

[0047] By adopting the above technical scheme, the information acquisition module is configured to acquire indoor environment information and plant physiological information of the greenhouse through a collection terminal. The input simulation module is configured to input the indoor environment information and the plant physiological information into a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response. The strategy matching module is configured to match a preset corresponding compensation strategy based on simulation information output by the photosynthesis-environment coupling model and determine compensation parameters, wherein the compensation strategy includes LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation. The parameter adjustment module is configured to acquire plant state information through a monitoring terminal and adjust corresponding compensation parameters in the matched compensation strategy based on the plant state information.

[0048] In summary, the present application has at least one of the following beneficial technical effects:

[0049] 1. The present application acquires environmental parameters and plant physiological indicators, dynamically simulates and analyzes them in combination with a pre-trained photosynthesis simulation model, intelligently matches optimal LED spectrum ratio, CO2 compensation strategy, and pulse light compensation parameters, realizes integrated linkage regulation and control of key environmental parameters such as light and CO2 concentration, effectively solves the problem of low resource utilization efficiency caused by traditional single-factor independent control, and, compared with the traditional fixed threshold control method, can realize dynamic feedback and response according to the actual physiological state of crops, thereby significantly improving the timeliness and accuracy of environmental regulation. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1is a flowchart of step S20 in an embodiment of the agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation according to the application;

[0051] Figure 2 is a flowchart of step S20 in an embodiment of the agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation according to the application;

[0052] Figure 3 is a flowchart of step S22 in an embodiment of the agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation according to the application;

[0053] Figure 4 is a flowchart of step S23 in an embodiment of the agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation according to the application;

[0054] Figure 5 is a flowchart of step S30 in an embodiment of the agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation according to the application. DETAILED DESCRIPTION

[0055] The following will be described in detail below with reference to the accompanying drawings. Figures 1-5 The application will be further described in detail.

[0056] In an embodiment, as shown in the accompanying drawings, the application discloses an agricultural greenhouse intelligent management method based on LED light source and spectrum and carbon dioxide compensation, which specifically comprises the following steps: Figure 1

[0057] S10: Obtain indoor environment information and plant physiological information of the greenhouse through the acquisition terminal;

[0058] In this embodiment, the acquisition terminal is a sensor network deployed in the greenhouse, including environment sensors (such as temperature sensors, humidity sensors, light sensors, CO2 concentration sensors) and plant physiological monitoring devices (such as chlorophyll fluorescence instruments, stem flow meters, multispectral cameras, etc.), which are used to collect environmental and plant physiological data in real time; the indoor environment information is the environmental parameter data inside the agricultural greenhouse, including temperature, humidity, light intensity, CO2 concentration, and other real-time monitored physical environment indicators; the indoor environment information is collected through the environment sensor network, and is used to evaluate the environmental state in the greenhouse and its influence on plant growth; the plant physiological information is data reflecting the plant growth state and physiological activity, including stomatal conductance, stem flow rate, chlorophyll fluorescence parameters, leaf temperature, etc.; the plant physiological information is obtained through the plant physiological monitoring device, and is used to evaluate the health status of the plant and its response to environmental changes.

[0059] ​Specifically, as described in step S10 above, the environmental parameters (including temperature, humidity, light intensity, CO2 concentration, etc.) and plant physiological indicators (such as chlorophyll fluorescence parameters, stomatal conductance, stem flow rate, etc.) are collected in real time by the multi-source sensor network deployed in the greenhouse. The collection terminal can use a hybrid networking method of wired and wireless to ensure the real-time and reliability of data transmission.

[0060] S20: inputting the indoor environmental information and plant physiological information into a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response;

[0061] In this embodiment, the photosynthesis-environment coupling model is a pre-trained mathematical model based on machine learning. After inputting the indoor environmental information and plant physiological information, the photosynthesis under different environmental conditions is simulated based on the response, and the simulation information is output.

[0062] Specifically, as described in step S20 above, the collected indoor environmental information and plant physiological information are input into the pre-trained photosynthesis-environment coupling model for analysis and processing. The photosynthesis-environment coupling model can be constructed using a deep learning algorithm to analyze the complex nonlinear relationship between environmental parameters and photosynthetic efficiency based on historical data. The simulation operation process of the photosynthesis-environment coupling model includes environmental parameter standardization processing, feature extraction, and photosynthetic efficiency prediction. Through the photosynthesis-environment coupling model, the photosynthetic response of plants under different environmental combinations can be simulated, and the optimal environmental parameter combination can be predicted. This data-driven modeling method can accurately reflect the actual impact of environmental factors on plant growth.

[0063] S30: matching the simulation information output by the photosynthesis-environment coupling model with a pre-set corresponding compensation strategy, and determining the compensation parameters, the compensation strategy including LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation;

[0064] In this embodiment, the simulation information is the prediction result output by the photosynthesis-environment coupling model, including photosynthetic efficiency prediction and optimal compensation strategy suggestion, etc. The simulation information is used to guide environmental regulation decisions. The compensation strategy includes LED spectrum adjustment (red / blue light ratio optimization, etc.), CO2 gradient compensation (concentration and rate control, etc.), and pulse light compensation (frequency, duty cycle adjustment, etc.), which are used to optimize the plant growth environment. The compensation parameters are specific control values that need to be set when executing the compensation strategy, including quantifiable operation parameters such as LED spectrum ratio (e.g. red:blue ratio), CO2 compensation rate, pulse light compensation frequency and duty cycle, etc.

[0065] Further, LED spectrum adjustment refers to the technical means of optimizing plant light morphogenesis and photosynthetic efficiency by adjusting the output ratio of different wavelengths of light in the LED light source (such as the light quantum flux ratio of red light 630-660 nm to blue light 450-470 nm), and the core control parameters include the light intensity ratio of each waveband (such as R:B=7:3), the light quantum flux density, and the photoperiod; the LED spectrum ratio is a quantitative representation of the energy distribution relationship of different waveband light, and typical parameters include the ratio of red light to blue light, the proportion of far-red light, and the absolute light intensity of each waveband; CO2 gradient compensation refers to the technology of dynamically regulating the CO2 concentration in the greenhouse according to the carbon assimilation requirements of photosynthesis, including compensation reference value (such as increasing from atmospheric concentration 400 ppm to 800 ppm), rising gradient (ppm / h), and maintenance time (h); CO2 compensation rate is the adjustment amplitude of CO2 concentration per unit time (ppm / min), including initial compensation rate (such as 20 ppm / min), gradient turning point (such as reducing speed when reaching 80% of the target concentration), and steady-state fluctuation range (±30 ppm); pulsed light refers to a technical solution that replaces continuous light with intermittent light (millisecond to second level period), and the core parameters include pulse frequency, duty cycle, peak light intensity, and phase synchronization accuracy with natural light; the pulse light frequency is the number of light pulses per minute, and its setting needs to consider the relaxation time of the light system, high frequency (such as >5Hz) is suitable for fast response plant crops, and low frequency (such as 1-2Hz) is suitable for light inhibition sensitive crops.

[0066] Specifically, as described in step S30 above, the optimal compensation strategy scheme is matched based on the simulation results of the model output, and the compensation strategy matching process considers multiple factors such as the current environmental state, plant growth stage, and variety characteristics, and determines the compensation parameters, wherein the compensation strategy and its compensation parameters include: the spectrum ratio of the LED light source (such as the ratio of red light to blue light), the rate and gradient of CO2 compensation, the frequency and duty cycle of pulsed light, and other control parameters. This step realizes the transformation from environmental diagnosis to control decision, ensuring that each control measure has data support.

[0067] S40: Obtain plant state information through the monitoring terminal, and adjust the corresponding compensation parameters in the matched compensation strategy based on the plant state information.

[0068] In this embodiment, the monitoring terminal is a device for real-time acquisition of plant physiological state, providing feedback data for closed-loop control; the plant state information is the plant physiological feedback data obtained in real time by the monitoring terminal, including photosynthetic activity change, transpiration rate adjustment, chlorophyll fluorescence dynamics, etc., and the plant state information is used to verify the effect of the compensation strategy and further optimize the control parameters.

[0069] Specifically, as described in step S40 above, the real-time physiological state data of the plant, including photosynthetic efficiency, transpiration rate, etc., is continuously acquired by monitoring the terminal, the actual effect of the implemented compensation measures is evaluated based on the plant state information, and the compensation parameters are dynamically adjusted according to the evaluation results: if the compensation is insufficient, the control intensity is increased; if the compensation is excessive, the control intensity is reduced; this closed-loop feedback mechanism realizes the control cycle of "monitoring, decision-making, execution, and evaluation", and through continuous optimization of the compensation parameters, it ensures that the plant is always in the best control state, while avoiding resource waste.

[0070] In an embodiment, the photosynthesis-environment coupling model includes a data processing layer, a correlation analysis layer, and an information output layer, as shown in Figure 2 Step S20 includes the following steps:

[0071] S21: The data processing layer calculates the photosynthetic efficiency index and the water stress coefficient based on the received indoor environment information and the plant physiological information;

[0072] S22: The correlation analysis layer performs correlation analysis on the photosynthetic efficiency index and the indoor environment information, identifies the limiting factor of photosynthesis, and corrects the weight of the limiting factor in combination with the water stress coefficient to generate a weight vector of the limiting factor;

[0073] S23: The information output layer identifies the dominant limiting factor based on the weight vector of the limiting factor, and generates simulation information for matching the compensation strategy based on the dominant limiting factor and its weight vector.

[0074] In the embodiment, the data processing layer is a front-end processing module of the photosynthesis-environment coupling model, used for pre-processing the input original indoor environment information and plant physiological information and calculating the photosynthetic efficiency index and the water stress coefficient; the correlation analysis layer is an analysis and calculation module of the photosynthesis-environment coupling model, used for identifying the main limiting factor by constructing a correlation matrix of the environmental parameters and the photosynthetic efficiency and correcting the weight of the limiting factor in combination with the water stress coefficient; the information output layer is an output result generation module of the photosynthesis-environment coupling model, used for identifying the dominant limiting factor based on the weight vector of the limiting factor and generating simulation information for matching the compensation strategy based on the dominant limiting factor and the weight vector of the dominant limiting factor, including the compensation strategy type and specific parameter suggestions; the photosynthetic efficiency index is an index for quantifying the photosynthesis efficiency of the plant, usually represented by the CO2 assimilation rate per unit leaf area, reflecting the light energy conversion efficiency; the water stress coefficient is a parameter for representing the water status of the plant, calculated based on the stomatal conductance, transpiration rate and other data, and the greater the calculated value, the more serious the water stress; the limiting factor is an element that has the greatest limiting effect on the photosynthesis in the current environment, including the light intensity, CO2 concentration and temperature; the weight vector is a three-dimensional vector, representing the weight values of the light limitation, carbon limitation and temperature limitation; the dominant limiting factor is the limiting factor with the greatest value in the weight vector, which is identified as the dominant limiting factor when the weight vector of the limiting factor exceeds a preset value, and the dominant limiting factor determines the compensation strategy to be taken first.

[0075] Specifically, as described in steps S21-S23, the data processing layer receives the original indoor environment information (light, CO2 concentration, temperature and humidity, etc.) and plant physiological information (chlorophyll fluorescence, stem flow, etc.) collected by the sensor, calculates the photosynthetic efficiency index and the water stress coefficient after pre-processing; the correlation analysis layer constructs a correlation matrix of the photosynthetic efficiency index and the indoor environment information, extracts the limiting factor of photosynthesis by principal component analysis, and dynamically corrects the initial weight of the limiting factor according to the water stress coefficient to generate a standardized weight vector; the information output layer identifies the dominant limiting factor with a value exceeding a threshold based on the weight vector of the limiting factor, and generates a simulation information package containing the compensation type, intensity parameter and expected effect in combination with the weight vector of the dominant limiting factor.

[0076] In an embodiment, the correlation analysis layer includes a matrix construction sub-layer, a factor identification sub-layer and a dynamic correction sub-layer, as shown in Figure 3 The step S22 includes the following steps:

[0077] S221: The matrix construction sub-layer constructs a correlation matrix based on the indoor environment information and the photosynthetic efficiency index;

[0078] S222: The factor identification sub-layer identifies the limiting factor that affects photosynthesis based on the environment in the correlation matrix, and calculates the initial weight of each limiting factor based on the correlation matrix;

[0079] S223: dynamically correct the initial weight of the limiting factor combined with the water stress coefficient to generate the weight vector of the limiting factor.

[0080] In the present embodiment, the matrix construction sublayer is a functional sub-module in the correlation analysis layer for correlation analysis of environmental parameters and photosynthetic efficiency. The matrix construction sublayer constructs a correlation matrix of indoor environmental information and photosynthetic efficiency index. The correlation matrix is a matrix constructed by the matrix construction sublayer using a grey correlation analysis method or the like. The correlation matrix can include 6 environmental factors: light intensity, CO2 concentration, temperature, humidity, air flow velocity, soil moisture, and 1 photosynthetic efficiency index. The greater the value range in the matrix, the stronger the correlation between the environmental factor and the photosynthetic efficiency. The factor identification sublayer is a functional sub-module in the correlation analysis layer for extracting limiting factors. The factor identification sublayer includes functions such as eigenvalue calculation, principal component screening, and weight conversion. The factor identification sublayer identifies the limiting factors in the correlation matrix and calculates the initial weight of each limiting factor based on the correlation matrix. The initial weight is the original influence coefficient extracted from the correlation matrix by the factor identification sublayer through principal component analysis or the like. The initial weight includes components such as light limitation weight, carbon limitation weight, and temperature limitation weight. The initial weight is calculated using an eigenvalue decomposition method. The initial weight does not take into account the adjustment effect of water stress. The dynamic correction sublayer is a functional sub-module in the correlation analysis layer for optimizing the weight. The dynamic correction sublayer corrects the initial weight of the limiting factor combined with the water stress coefficient to generate the weight vector of the limiting factor. The dynamic correction sublayer can be provided with processing units such as linear correction, compensation correction, and normalization. The water stress coefficient is a dynamic parameter quantifying the degree of water deficit of the plant. The water stress coefficient is usually calculated by fusing multiple sources of data such as stomatal conductance (percentage deviation of measured value from optimal value of variety), stem flow rate (decay ratio of current value to historical average value), and leaf water content (relative water content estimated based on multispectral imaging). In the present embodiment, the water stress coefficient can be calculated by a water stress calculation model based on the above multiple sources of data.

[0081] Specifically, as described above in steps S221-S223, the matrix construction sublayer receives indoor environmental information and photosynthetic efficiency index and constructs a correlation matrix. The factor identification sublayer identifies the correlation matrix and extracts corresponding principal components as limiting factors. The factor identification sublayer calculates the initial weight vector of each limiting factor based on the correlation matrix. The dynamic correction sublayer obtains the water stress coefficient and corrects the initial weight of the limiting factor combined with the water stress coefficient to generate the weight vector of the limiting factor.

[0082] In an embodiment, the information output layer includes a model construction sublayer, a weight comparison sublayer, and an information generation sublayer, as shown in FIG. 8. Figure 4 Step S23 includes the following steps:

[0083] S231: The model construction sub-layer obtains plant variety information and its corresponding initial weight threshold, and constructs a threshold adjustment model to dynamically correct the initial weight threshold;

[0084] S232: The weight comparison sublayer compares the weight vector of the limiting factor with its corresponding dynamic threshold, which has been dynamically corrected by the threshold adjustment model, in order to identify the dominant limiting factor.

[0085] S233: The information generation sublayer generates simulation information for matching compensation strategies based on the dominant constraint factor.

[0086] In this embodiment, the model construction sublayer is the dynamic threshold management function submodule of the information output layer. It can optionally integrate a plant variety database and a growth stage identification algorithm. By acquiring plant variety information and its corresponding initial weight thresholds, and constructing a threshold adjustment model, it dynamically corrects and optimizes the initial weight thresholds of the limiting factors. The weight comparison sublayer is the dominant factor determination function submodule of the information output layer. It can include a real-time weight vector parser, a dynamic threshold matcher, and a confidence evaluation unit. By comparing the weight vector of the limiting factor with its corresponding dynamic threshold, which has been dynamically corrected by the threshold adjustment model, it identifies the dominant limiting factor. The information generation sublayer is the compensation strategy compilation function submodule of the information output layer. It can integrate a strategy template library and configure a corresponding strategy template library matching algorithm to match the corresponding strategy template based on the dominant limiting factor. Finally, it outputs a simulation information package containing compensation type and intensity parameters. The simulation information package can also include time series planning and expected effects.

[0087] Specifically, as described in steps S231-S233 above, the model construction sublayer obtains the variety information of the currently cultivated crop and its corresponding initial weight threshold parameters. The initial weight thresholds include preset values ​​such as light limitation, carbon limitation, and temperature limitation. Subsequently, the model construction sublayer constructs a threshold adjustment model to predict the threshold adjustment direction of each limiting factor, thereby dynamically correcting the initial weight thresholds. For example, when a continuous warming trend is detected, the temperature limitation threshold will be increased. The weight comparison sublayer analyzes the dynamic thresholds after dynamic correction by the threshold adjustment model and compares the weight vectors of the limiting factors with the corresponding dynamic thresholds to identify the dominant limiting factors. The information generation sublayer retrieves the matching basic compensation scheme from the preset strategy template library based on the determined dominant limiting factor type and outputs a simulation information package containing information such as compensation type and intensity parameters.

[0088] In one embodiment, such as Figure 5 As shown, step S30 includes the following steps:

[0089] S31: match a compensation strategy for the simulation information based on a pre-constructed compensation strategy knowledge base;

[0090] S32: match a preset corresponding parameter optimization algorithm based on the compensation strategy, and obtain an optimization coefficient corresponding to the parameter optimization algorithm;

[0091] S33: determine compensation parameters based on the parameter optimization algorithm and the optimization coefficient, the compensation parameters including a spectral compensation amount, a CO2 compensation flux, and a pulse parameter.

[0092] In the embodiment, the compensation strategy knowledge base is a structured database storing various environmental compensation schemes, including spectral adjustment, CO2 compensation, and pulse light compensation. Each strategy records metadata such as applicable environmental condition range, target crop type, and execution parameter interval. The parameter optimization algorithm is a special algorithm designed for various compensation strategies, which can include a particle swarm algorithm for spectral optimization, a model predictive control for CO2 compensation, and a genetic algorithm for pulse parameter. Each algorithm is configured with a corresponding optimization coefficient matrix, including crop variety coefficient, environmental correction coefficient, and device characteristic coefficient. The optimization coefficient is an adjustment parameter of the parameter optimization algorithm, which is stored in the form of a multi-dimensional vector, including basic coefficient, dynamic correction term, and device constraint term. The optimization coefficient can be obtained from the variety database, environmental sensors, and device controllers. The spectral compensation amount is a quantitative parameter of LED light source adjustment, including waveband ratio, light intensity, and photoperiod, which is calculated by the spectral optimization algorithm and needs to meet the daily cumulative amount requirement of photosynthetic photon flux density. The CO2 compensation flux is the execution parameter of carbon compensation, including target concentration, rising rate, and maintenance time length. The optimal compensation gradient is calculated by the first derivative of the carbon response curve, and is corrected considering the gas exchange efficiency of the ventilation system. The pulse parameter is the control variable of intermittent light compensation, including frequency, duty cycle, and peak light intensity.

[0093] Specifically, as described in steps S31-S33 above, the dominant limiting factor type and environmental state parameters in the simulation information package are analyzed, and multi-condition matching is performed based on the compensation strategy knowledge base. The query conditions can include photosynthetic characteristics of the current crop variety, deviation degree of environmental parameters, and working state of the device, etc. The matching process can select the nearest neighbor algorithm to calculate the feature vector distance of the candidate strategy and the current scene, select several strategies with high similarity to form a candidate set, and then determine the compensation strategy to be finally used through economic benefit evaluation and other methods. For example, when the dominant factor is light limitation and the environmental temperature is in the range of 25-28℃, the "enhanced red light supplement" strategy template is preferentially matched. According to the selected compensation strategy type, the corresponding parameter optimization algorithm is called and the optimization coefficient required by the parameter optimization algorithm is obtained. Compensation parameters are determined based on the parameter optimization algorithm and the optimization coefficient, including spectral compensation amount, CO2 compensation flux, and pulse parameter.

[0094] Wherein, the parameter optimization algorithm of spectral compensation amount is:

[0095] Δλ i is the spectral compensation amount of the i-th waveband, K i is the waveband weight coefficient, calibrated through plant crop spectrum response experiment, and the waveband weight coefficient adopts the existing experimental calibration value, is the partial derivative of photosynthetic efficiency Φ to specific wavelength light intensity λ i and the partial derivative of photosynthetic efficiency Φ to specific wavelength light intensity λ j , measured through multi-spectral LED gradient test, and the partial derivative adopts the existing test measurement value;

[0096] The parameter optimization algorithm of CO2 compensation flux is:

[0097] is the CO2 compensation flux, β is the weight vector coefficient of carbon limitation factor, that is, the carbon limitation dimension component extracted from the weight vector output by the photosynthesis-environment coupling model, V cmax (T) is the temperature-corrected maximum carboxylation rate, which is often approximated by an exponential value in agricultural applications, and the temperature-corrected maximum carboxylation rate adopts the typical value for different types of plants recorded in existing literature, γ is the temperature response attenuation coefficient, obtained by enzyme kinetics test fitting, and the temperature response attenuation coefficient adopts the existing experimental test value, ΔT is the temperature deviation from the optimum value, calculated by the difference between leaf temperature and photosynthetic optimum temperature, wherein the leaf temperature is measured by an infrared thermal imager or a thermocouple, and the photosynthetic optimum temperature adopts the existing literature calibration value, [CO2] is the current environmental (i.e. in the greenhouse) CO2 concentration, obtained by a collection terminal, Γ * is the CO2 compensation point, obtained by low-oxygen photosynthesis measurement, K m is the Michaelis constant, obtained by in vitro enzyme activity test, and the Michaelis constant adopts the existing experimental calibration value;

[0098] The parameter optimization algorithm of pulse parameters includes:

[0099] Frequency optimization: f opt is the frequency optimization parameter of pulse parameters, τQA is the reduction time of QA, obtained by chlorophyll fluorescence transient measurement, τPQ is the plastoquinone diffusion time, obtained by fluorescence relaxation experiment, and the plastoquinone diffusion time adopts the existing experimental calibration value, ΔT is the temperature deviation from the optimum value, calculated by the difference between leaf temperature and photosynthetic optimum temperature, wherein the leaf temperature is measured by an infrared thermal imager or a thermocouple, and the photosynthetic optimum temperature adopts the existing literature calibration value;

[0100] Duty cycle optimization: D opt D is a duty cycle optimization parameter for pulse parameters base K is a base duty cycle calibrated by light saturation experiment, which adopts the existing experimental calibration value D α is a duty cycle adjustment gain coefficient calibrated by light and utilization efficiency optimization experiment, and which adopts the existing experimental calibration value, and α is a weight vector coefficient of light limiting factor, that is, a light limiting dimension component extracted from a weight vector output by a photosynthesis-environment coupling model comp α is a light compensation point corresponding weight obtained by low light photosynthesis measurement sat α is a light saturation point corresponding weight obtained by light response curve fitting.

[0101] In an embodiment, step S32 comprises the following steps:

[0102] S321: If several compensation strategies are matched at the same time, the intensity coefficients of different compensation strategies are determined based on the weight vector of the dominant limiting factor;

[0103] S322: The current stress level of the dominant limiting factor is judged, and a multi-objective collaborative strategy is determined based on the intensity coefficients of the compensation strategies and the current stress level of the dominant limiting factor.

[0104] In the embodiment, the intensity coefficients of the compensation strategies are parameters for quantifying the priority and execution intensity of different compensation strategies under the current environmental conditions, which are calculated by normalization of the weight vector of the dominant limiting factor; the stress level is a physiological state level of the plant divided according to the weight value of the dominant limiting factor and the environmental deviation, which can be divided into: mild stress, moderate stress and severe stress; the multi-objective collaborative strategy is an optimized control scheme considering the photosynthetic efficiency improvement, energy consumption control and equipment life balance, which is generated by a multi-objective optimization algorithm, and outputs an instruction set containing compensation type, parameters and execution timing;

[0105] For example, the plant performance under mild stress can be selected as: slight photosynthetic rate decline, no visible morphological damage, such as leaf curling or chlorosis; the plant performance under moderate stress can be selected as: obvious photosynthetic inhibition, slight leaf curling or local chlorosis, which needs to be intervened within a limited time to avoid irreversible damage; the plant performance under severe stress can be selected as: photosynthesis close to stagnation, severe wilting or burning of leaves, which needs to start emergency compensation immediately;

[0106] For example, taking the collaborative regulation of light limitation and carbon limitation as an example, the scenario is that greenhouse tomatoes encounter continuous overcast days (low light and low CO2 concentration), the light limitation weight is 0.7 (judged as moderate stress), and the carbon limitation weight is 0.3 (judged as mild stress), at this time, the multi-objective collaborative strategy optimization logic is:

[0107] photosynthetic efficiency priority: light limitation weight is higher, so the spectral adjustment intensity is greater than CO2 compensation;

[0108] energy consumption control: reduce the proportion of blue light (higher energy consumption) and limit the CO2 rising rate;

[0109] equipment protection: stagger the LED spectrum regulation and the full load operation period of the CO2 generator.

[0110] Specifically, as described in steps S321-S322 above, the number of currently matched compensation strategies is checked, and if there are multiple strategies (such as the need for spectral adjustment and CO2 compensation at the same time), the weight components corresponding to each strategy are extracted from the weight vector, the intensity coefficient is calculated through normalization processing, and is used for subsequent priority sorting and resource allocation of the strategy; determine the current stress level according to the weight value of the dominant limiting factor and the real-time environmental parameters, and combine the intensity coefficient of the compensation strategy and the current stress level to call a multi-objective optimization algorithm to determine the optimal multi-objective collaborative strategy.

[0111] In an embodiment, step S40 includes the following steps:

[0112] S41: determining the plant stress state based on plant state information;

[0113] S42: adjusting the corresponding compensation parameters in the matched compensation strategy based on the plant stress state.

[0114] In this embodiment, the plant stress state is the degree of environmental stress quantified by plant physiological indicators (such as photosynthetic efficiency, stomatal conductance, and chlorophyll fluorescence parameters), which is divided into three levels of mild, moderate, and severe, and its determination needs to be combined with real-time monitoring data and variety-specific threshold values; the compensation parameter adjustment is to dynamically correct the compensation parameters (such as spectral ratio, CO2 flux, and pulse duty cycle) of the executed strategy according to the stress state, and the adjustment amplitude follows a linear relationship of "stress level x baseline adjustment amount", while being constrained by the maximum working parameters of the equipment.

[0115] Specifically, as described in steps S41-S42 above, plant state information is collected at a fixed frequency, and the plant stress state is determined based on the plant state information, for example, a pre-trained stress classification model (based on the random forest algorithm) is input, so that the stress classification model standardizes the original data of the plant state information and calculates the deviation percentage of each indicator relative to the optimal growth state of the variety, thereby determining the plant stress state; the currently executed compensation strategy parameters are called, and the compensation parameters are adjusted based on the plant stress state.

[0116] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0117] In an embodiment, an LED light source and spectrum and carbon dioxide compensation based intelligent management system for agricultural greenhouse is provided, which corresponds to the LED light source and spectrum and carbon dioxide compensation based intelligent management method for agricultural greenhouse in the above embodiment. The LED light source and spectrum and carbon dioxide compensation based intelligent management system for agricultural greenhouse comprises:

[0118] An information acquisition module is configured to acquire indoor environment information and plant physiological information of the greenhouse through a terminal;

[0119] An input simulation module is configured to input the indoor environment information and the plant physiological information to a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response;

[0120] A strategy matching module is configured to match a preset corresponding compensation strategy based on simulation information output by the photosynthesis-environment coupling model and determine compensation parameters, wherein the compensation strategy comprises LED spectrum adjustment, CO2 gradient compensation, and pulsed light compensation;

[0121] A parameter adjustment module is configured to acquire plant state information through a monitoring terminal and adjust corresponding compensation parameters in the matched compensation strategy based on the plant state information.

[0122] For specific limitations of the LED light source and spectrum and carbon dioxide compensation based intelligent management system for agricultural greenhouse, refer to the limitations of the LED light source and spectrum and carbon dioxide compensation based intelligent management method for agricultural greenhouse in the above, which will not be repeated here. Each module in the above LED light source and spectrum and carbon dioxide compensation based intelligent management system for agricultural greenhouse can be realized by software, hardware, and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0123] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. An intelligent management method for an agricultural greenhouse based on an LED light source and carbon dioxide compensation, characterized in that: The method comprises the steps of: obtaining indoor environment information and plant physiological information of the greenhouse through a collection terminal; inputting the indoor environment information and the plant physiological information into a pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response; matching a preset corresponding compensation strategy based on simulation information output by the photosynthesis-environment coupling model, and determining compensation parameters, wherein the compensation strategy comprises LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation; obtaining plant state information through a monitoring terminal, and adjusting corresponding compensation parameters in the matched compensation strategy based on the plant state information; ​ The photosynthesis-environment coupling model comprises a data processing layer, an association analysis layer, and an information output layer. The step of inputting the indoor environment information and the plant physiological information into the pre-trained photosynthesis-environment coupling model to simulate photosynthesis based on environmental response comprises the steps of: calculating a photosynthetic efficiency index and a water stress coefficient based on the received indoor environment information and the plant physiological information by the data processing layer; The association analysis layer performs association analysis on the photosynthetic efficiency index and the indoor environment information, identifies a limiting factor of photosynthesis, and corrects the weight of the limiting factor in combination with the water stress coefficient to generate a weight vector of the limiting factor. The information output layer identifies a dominant limiting factor based on the weight vector of the limiting factor, and generates simulation information for matching the compensation strategy based on the dominant limiting factor and the weight vector thereof. The information output layer comprises a model construction sublayer, a weight comparison sublayer, and an information generation sublayer. The step of identifying a dominant limiting factor based on the weight vector of the limiting factor, and generating simulation information for matching the compensation strategy based on the dominant limiting factor and the weight vector thereof by the information output layer comprises the steps of: obtaining plant variety information and corresponding initial weight threshold values thereof by the model construction sublayer, and constructing a threshold adjustment model to dynamically correct the initial weight threshold values; comparing the weight vector of the limiting factor with the dynamic threshold values thereof that are dynamically corrected by the threshold adjustment model by the weight comparison sublayer, so as to identify the dominant limiting factor; and generating simulation information for matching the compensation strategy based on the dominant limiting factor by the information generation sublayer.

2. The LED light source and spectrum and carbon dioxide compensation-based intelligent management method for agricultural greenhouse according to claim 1, characterized in that: The association analysis layer comprises a matrix construction sublayer, a factor identification sublayer, and a dynamic correction sublayer. The step of performing association analysis on the photosynthetic efficiency index and the indoor environment information by the association analysis layer, identifying a limiting factor of photosynthesis, and correcting the weight of the limiting factor in combination with the water stress coefficient to generate a weight vector of the limiting factor comprises the steps of: constructing an association degree matrix based on the indoor environment information and the photosynthetic efficiency index by the matrix construction sublayer; The factor identification sublayer identifies a limiting factor that affects photosynthesis based on the association degree matrix in the association degree matrix, and calculates initial weights of each limiting factor based on the association degree matrix. The dynamic correction sublayer corrects the initial weights of the limiting factors in combination with the water stress coefficient to generate a weight vector of the limiting factors. 3.The LED light source and spectrum and carbon dioxide compensation based intelligent management method of agricultural greenhouse according to claim 1, characterized in that: The simulation information output by the photosynthesis-environment coupling model is matched with a preset corresponding compensation strategy, and a compensation parameter is determined, the compensation strategy including steps of LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation, including the step of matching the simulation information with a compensation strategy based on a pre-constructed compensation strategy knowledge base; A parameter optimization algorithm corresponding to the compensation strategy is matched based on a preset corresponding parameter optimization algorithm, and an optimization coefficient corresponding to the parameter optimization algorithm is obtained; The compensation parameter is determined based on the parameter optimization algorithm and the optimization coefficient, the compensation parameter including a spectrum compensation amount, a CO2 compensation flux, and a pulse parameter. 4.The LED light source and spectrum and carbon dioxide compensation based intelligent management method of agricultural greenhouse according to claim 3, characterized in that: The step of matching the compensation strategy with a preset corresponding parameter optimization algorithm and obtaining an optimization coefficient corresponding to the parameter optimization algorithm includes the steps of: if a plurality of compensation strategies are matched at the same time, determining an intensity coefficient of different compensation strategies based on a weight vector of a dominant limiting factor; judging a current stress level of the dominant limiting factor, and determining a multi-objective collaborative strategy based on the intensity coefficient of the compensation strategy and the current stress level of the dominant limiting factor. 5.The LED light source and spectrum and carbon dioxide compensation based intelligent management method of agricultural greenhouse according to claim 1, characterized in that: The step of obtaining plant state information through a monitoring terminal and adjusting the corresponding compensation parameter in the matched compensation strategy based on the plant state information includes the steps of: judging a plant stress state based on the plant state information; and adjusting the corresponding compensation parameter in the matched compensation strategy based on the plant stress state.

6. An intelligent management system for LED light source and spectrum and carbon dioxide compensation based agricultural greenhouse, used for the method of any one of claims 1-5, characterized in that: It includes: An information acquisition module for acquiring indoor environment information and plant physiological information of a greenhouse through an acquisition terminal; An input simulation module for inputting the indoor environment information and the plant physiological information to a pre-trained photosynthesis-environment coupling model for photosynthesis simulation based on environmental response; A strategy matching module for matching a preset corresponding compensation strategy based on simulation information output by the photosynthesis-environment coupling model, and determining a compensation parameter, the compensation strategy including LED spectrum adjustment, CO2 gradient compensation, and pulse light compensation; A parameter adjustment module for obtaining plant state information through a monitoring terminal and adjusting the corresponding compensation parameter in the matched compensation strategy based on the plant state information.

Images