Intelligent illumination control system for greenhouse plant growth

Abstract

The invention relates to the technical field of facility agriculture and environment control, and discloses an intelligent illumination control system for greenhouse plant growth. In order to solve the problems that an existing light supplementing strategy lacks physiological feedback, light damage is easily caused, and shutdown is caused by sensor faults, the system comprises a multi-mode sensing assembly, a central processing unit and a dynamic spectrum execution assembly. The system collects crop non-photochemical quenching coefficients (NPQ) and environmental factors in real time, and dynamically calculates light saturation points by using a multi-factor coupling model; when the NPQ exceeds a threshold value, the spectrum ratio is automatically adjusted to inhibit light stress; three-dimensional canopy scanning is combined to realize gridding differentiation light supplement; when the sensor fails, virtual parameters are reconstructed based on the temperature trend to maintain fault-tolerant operation; in addition, the system estimates stomatal conductance according to air humidity and matrix conductivity, and dynamically corrects target light intensity to avoid aggravation of water deficit; accurate light supply according to needs is achieved, the photosynthetic efficiency and the system reliability are remarkably improved, and the system is suitable for plant factories and multi-span greenhouses.

Description

Technical Field

[0001] This invention relates to the field of facility agriculture and environmental control technology, specifically to an intelligent lighting control system for greenhouse plant growth. Background Technology

[0002] With the rapid development of facility agriculture, artificial lighting has become a key means to regulate crop growth cycles and improve yield and quality. However, existing greenhouse plant lighting control systems are mostly based on open-loop control of environmental factors or static adjustment of fixed spectral ratios. These systems mainly rely on light intensity sensors or preset schedules to work, lacking coupled calculations of multiple factors such as carbon dioxide and temperature, and are unable to perceive the physiological state of crops in real time. This "blind supplemental lighting" mode causes the system to continue to output high intensity even when the stomata of crops close due to stress such as high temperature and water shortage. This not only wastes electricity but also exacerbates photoinhibition and may even cause leaf burn.

[0003] Furthermore, existing technologies generally lack dynamic spectral adjustment mechanisms based on real-time physiological feedback from crops (such as the non-photochemical quenching coefficient NPQ), making it impossible to dissipate excess light energy in the early stages of light stress by adjusting the spectral ratio (such as increasing far-red light), resulting in crops being in a sub-healthy state for a long time. At the same time, traditional LED arrays are usually controlled uniformly as a whole, ignoring the problem of uneven light reception caused by differences in canopy height, and the system is highly dependent on the accuracy of sensor data. Once a key detection unit fails, it often shuts down directly or outputs erroneous commands, lacking the fault tolerance capability to reconstruct parameters based on historical data, which seriously affects the operational reliability and energy efficiency of the system.

[0004] In summary, there is an urgent need for a smart lighting control system for greenhouse plant growth that can collect real-time physiological state parameters and environmental factor data of crop canopy, dynamically calculate the light saturation point through a multi-factor coupling model, and intelligently adjust the spectral ratio and spatial distribution based on physiological state markers, while also having the ability to reconstruct data anomalies. This system would solve the technical problems of current supplemental lighting strategies being outdated, having low energy efficiency, and being prone to light damage. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an intelligent lighting control system for greenhouse plant growth. It features real-time feedback adjustment based on crop physiological state (such as NPQ), precise calculation of dynamic light saturation point coupled with multiple environmental factors, adaptive optimization of spectral ratio and spatial distribution, and fault-tolerant reconstruction under sensor failure. It solves the problems of "blind supplemental lighting" and light suppression damage caused by the lack of physiological perception in traditional supplemental lighting strategies, the inability of fixed spectra to effectively alleviate light stress, uneven light reception in the canopy, and low operational reliability due to system reliance on a single sensor.

[0006] (II) Technical Solution To address the shortcomings of traditional supplemental lighting strategies, such as lag, low energy efficiency, and susceptibility to light damage, and to improve system reliability and crop photosynthetic efficiency, this invention provides the following technical solution: an intelligent lighting control system for greenhouse plant growth, comprising a multimodal sensing component, a central processing unit, and a dynamic spectrum execution component, wherein; A multimodal sensing component is used to collect physiological state parameters and environmental factor data of crop canopy in real time. The multimodal sensing component includes a physiological state detection unit for detecting the physiological state of crops. The central processing unit is connected to the multimodal sensing components via signals and has a pre-installed stress determination module and performance calculation module. The dynamic spectral execution component, which is signal-connected to the central processing unit, includes a multi-channel independently addressable LED array, which contains light sources of various different wavelengths; The stress determination module generates physiological state markers based on physiological state parameters; The performance calculation module calls a multi-factor coupling model based on environmental factor data to calculate the dynamic light saturation point; The central processing unit generates control commands containing the driving parameters of each channel light source based on the physiological state markers and dynamic light saturation points, and sends them to the dynamic spectrum execution component.

[0007] Preferably, the physiological state parameters include at least the non-photochemical quenching coefficient (NPQ); the environmental factor data include at least carbon dioxide concentration and leaf temperature; The execution logic of the central processing unit includes: When the non-photochemical quenching coefficient (NPQ) exceeds a preset threshold, it is determined to be a photo-stressed state, and a stress suppression strategy is executed first. The stress suppression strategy includes reducing the output ratio of the blue light band light source and increasing the output ratio of the far-infrared light band light source. When the non-photochemical quenching coefficient (NPQ) does not exceed the preset threshold, it is determined to be in a normal state, and a dynamic optimization strategy is executed: based on the current carbon dioxide concentration and leaf temperature, the target photosynthetically active radiation threshold is calculated using a multi-factor coupling model, and the output intensity of the red and blue light bands in the LED array is adjusted according to the difference between the target photosynthetically active radiation threshold and the natural light intensity.

[0008] Preferably, the multimodal sensing component further includes an air humidity sensor and a matrix conductivity sensor; The central processing unit is also equipped with a collaborative correction module for estimating crop stomatal conductance based on air humidity and matrix conductivity. When the estimated stomatal conductance is lower than the preset lower limit, the collaborative correction module reduces the target photosynthetically effective radiation threshold in the dynamic optimization strategy to reduce the light source output intensity.

[0009] Preferably, the multimodal sensing component further includes a three-dimensional canopy scanning unit for acquiring three-dimensional height distribution data of the crop canopy; The LED array in the dynamic spectral execution component is divided into multiple independently controlled spatial grids; The central processing unit identifies the differences in canopy height within each spatial grid based on the three-dimensional height distribution data, and adjusts the light source output intensity in different spatial grids accordingly, thereby making the distribution of photon flux density on the canopy surface more uniform.

[0010] Preferably, the central processing unit further includes a fault-tolerant processing module; When the multimodal sensing component experiences data anomalies, the fault-tolerant processing module reconstructs the virtual physiological state parameters based on the canopy temperature change trend and historical operating data, and switches to a preset safe spectrum mode for operation.

[0011] Preferably, the peak wavelength of the red light band in the LED array is 655nm-665nm, the peak wavelength of the blue light band is 445nm-455nm, and the peak wavelength of the far-infrared light band is 725nm-735nm. The physiological state detection unit is a chlorophyll fluorescence imager or a hyperspectral camera.

[0012] (III) Beneficial Effects Compared with the prior art, the present invention provides an intelligent lighting control system for greenhouse plant growth, which has the following beneficial effects: 1. This invention can directly sense whether crops are under light stress by real-time monitoring of key physiological parameters such as the non-photochemical quenching coefficient (NPQ). When the NPQ exceeds the threshold, the system automatically triggers a stress suppression strategy, dynamically reducing the proportion of blue light and increasing the proportion of far-red light, promoting the dissipation of excess light energy, and preventing leaf burn and decreased photosynthetic efficiency from the source. Compared with the traditional "blind supplemental lighting" that relies solely on illuminance sensors, this system ensures that the supplemental lighting intensity is always within the physiological range that crops can tolerate, significantly improving the crop's stress resistance and health level, while avoiding the waste of electricity caused by ineffective lighting.

[0013] 2. This invention overcomes the limitations of single-factor control by incorporating multiple environmental factors, such as carbon dioxide concentration, leaf temperature, air humidity, and substrate conductivity, into a coupled model. The system can dynamically correct the target photosynthetically effective radiation threshold based on real-time stomatal conductance estimates: it automatically reduces supplemental light intensity when stomata are closed or carbon assimilation is limited, and maximizes photosynthetic potential when the environment is suitable. This "on-demand light supply" strategy ensures that every unit of electrical energy is converted into effective photosynthetic products, significantly improving the energy efficiency ratio (kg / kWh) of greenhouse cultivation.

[0014] 3. This invention addresses the problem of overexposure in the upper layer and underexposure in the lower layer caused by the overall control of traditional LED arrays. This invention combines three-dimensional canopy scanning technology with a multi-channel independently addressable LED array, which can identify the differences in canopy height within different spatial grids and implement differentiated brightness and spectral control, making the distribution of photon flux density on the canopy surface more uniform. At the same time, the system can dynamically adjust the ratio of red, blue, and far-red light according to the crop growth stage and real-time stress status, which not only optimizes the photosynthesis of the canopy but also promotes the uniformity of crop morphogenesis, thereby improving the overall yield and quality.

[0015] 4. To address the issue of sensors in agricultural fields being susceptible to contamination or malfunction, this invention incorporates a fault-tolerant processing module. When data from critical physiological sensors (such as fluorescence detection units) is abnormal, the system will not shut down or output incorrect commands. Instead, it will reconstruct virtual physiological state parameters based on canopy temperature change trends and historical operating data, and automatically switch to a preset safe spectrum mode. This mechanism greatly reduces the system's dependence on a single sensor, ensuring that the lighting control system can maintain basic functions even in the event of partial equipment failure or extreme weather conditions, significantly improving the operational reliability and maintenance convenience of facility agriculture equipment. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structural framework of the intelligent lighting control system for greenhouse plant growth of the present invention; Figure 2 This is a flowchart of the control logic of the central intelligent decision-making unit of the present invention. Detailed Implementation

[0017] 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.

[0018] like Figure 1As shown, a smart lighting control system for greenhouse plant growth includes a multimodal physiological sensing module, an environmental factor monitoring module, a central intelligent decision-making unit, a distributed adjustable spectrum LED array, and a fault-tolerant control execution module. The multimodal physiological sensing module is used to collect chlorophyll fluorescence parameters (including non-photochemical quenching coefficient NPQ), canopy temperature and hyperspectral reflectance data of crop canopy in real time, so as to quantify the real-time photosynthetic activity and light stress status of crops. The environmental factor monitoring module is used to simultaneously acquire data on light intensity (PAR), carbon dioxide concentration, air temperature and humidity, and soil matrix moisture content in the greenhouse. The central intelligent decision-making unit has a built-in multi-factor coupled photosynthesis model. It receives the above-mentioned sensing and monitoring data, dynamically calculates the crop light saturation point and optimal spectral ratio under the current environment, and generates control instructions that include brightness, spectral composition and spatial distribution strategies. The distributed adjustable spectrum LED array consists of multiple independently addressable lamp groups. According to the control command, the light intensity of the lamp groups at different heights or in different areas is adjusted in a differentiated manner (red / blue / far-red ratio) to achieve uniform light reception in the canopy. The fault-tolerant control execution module is used to monitor the validity of sensor data in real time. When abnormal or missing data of key physiological sensors is detected, it automatically switches to the fault-tolerant mode based on historical data trends and environmental factor substitution parameters to maintain the continuous and stable operation of the system.

[0019] Example 1 System hardware architecture and basic control logic This embodiment provides an intelligent lighting control system for greenhouse plant growth, which is mainly applied in plant factories or multi-span greenhouses for high-value leafy vegetables (such as lettuce).

[0020] The system hardware includes: Multimodal sensing components: including a chlorophyll fluorescence imager (for acquiring NPQ data) and an infrared thermal imager (for acquiring leaf temperature). Sensors, air humidity sensors, matrix conductivity (EC) sensors, and lidar (LiDAR, as a three-dimensional canopy scanning unit).

[0021] Central processing unit: Adopts industrial-grade edge computing gateway, internally deployed with a stress determination module, performance calculation module, collaborative correction module and fault tolerance processing module.

[0022] Dynamic Spectrum Execution Component: Composed of several independently addressable LED panels arranged in a grid array; specifically, the array is divided into 8×8 or 10×10 spatial control grids, with each grid corresponding to one set of LED panels, capable of independently receiving control commands; each set of LED panels contains three independent drive channels, whose wavelength parameters are strictly matched to the crop's photosynthetic requirements: red light peak wavelength 655-665nm, blue light peak wavelength 445-455nm, and far-red light peak wavelength 725-735nm; the drive current adjustment accuracy reaches ±1mA, and the response time is ≤50ms, ensuring millisecond-level response to changes in physiological state.

[0023] The specific implementation of the core algorithm model: (1) Construction of multi-factor coupling model In the "Efficiency Calculation Module", an improved photosynthetic response model is used to dynamically calculate the target photosynthetically effective radiation threshold. The specific mathematical formula is as follows:

[0024] in: This is the maximum photosynthetic rate constant at the current crop growth stage; The intercellular CO2 concentration (estimated from the environmental CO2 concentration); This is the CO2 compensation point; This is an empirical coefficient; Let be the blade temperature response function, when the blade temperature... When deviating from the optimal temperature range (e.g., 20–25°C), the function value decreases parabolically; This is the porosity correction factor (see below).

[0025] To ensure the feasibility of the plan, this example provides specific parameter values ​​for a typical crop (calibrated through 3 sets of parallel experiments): Maximum photosynthetic rate Take 18-22g of lettuce. mol·m -2 ·s -1 Take 25-30 grams of tomatoes. mol·m -2 ·s -1 ; CO2 compensation points Take 50-60g of lettuce. mol·mol -1 Take 60-70g of tomatoes. mol·mol-1 ; Empirical coefficient: , ; Blade temperature response function : Lettuce (optimal temperature 20-25°C): ; Tomatoes (optimal temperature 23-29°C): .

[0026] (2) Porous Conductivity Estimation Method In the "Collaborative Correction Module", due to the direct measurement of porosity ( The cost is high and the time lag is significant. This system adopts an air humidity-based approach. ) and matrix conductivity ( The indirect estimation method is used; the corresponding relationships are constructed using the following lookup table or empirical formula:

[0027] in: This represents the maximum stomatal conductance of this crop variety; This is a humidity sensitivity coefficient, reflecting the rate at which stomata close due to dry air; The optimal substrate conductivity for crops; This represents the salt stress coefficient.

[0028] When calculated When the value falls below a preset lower limit (e.g., 40% of the normal value), it is determined to be a state of stomatal limitation, and the system will forcibly adjust the stomatal structure. Reduce the amount of water by multiplying it by a factor of 0.6 to prevent strong light from exacerbating water loss.

[0029] The specific values ​​of each parameter and the judgment criteria in this embodiment are as follows: Maximum porosity Take 0.3-0.5 mol·m -2 ·s -1 Take 0.4-0.6 mol / m of tomato. -2 ·s -1

[0030] Humidity sensitivity Take a uniform value of 0.015; Optimal matrix conductivity For lettuce, use 1.2-1.8 mS / cm; for tomatoes, use 1.8-2.5 mS / cm. Salt stress coefficient Take a uniform value of 0.12; Preset lower limit judgment criteria: when the estimated value ≤0.12mol·m -2 ·s -1 (Lettuce) or ≤0.16mol·m -2 ·s -1 (Tomatoes) at (i.e., approximately each) If 40% of the pores are confined, the system determines that the pores are severely restricted and forcibly restricts their movement. Adjustments were made by reducing the amount of material removed.

[0031] (3) Algorithm for reconstructing virtual physiological state parameters In the "fault tolerance module," when the chlorophyll fluorescence imager malfunctions or data is abnormal (such as data jumps exceeding 3 standard deviations), the system initiates a reconstruction mechanism; it employs the extended Kalman filter (EKF) algorithm, using the canopy temperature change rate ( Using ) as the observed variable, reconstruct the virtual NPQ value ( ): Equations of state: , where state variables Includes true NPQ and heat dissipation rate.

[0032] Observation equation: Among them, the observed values The rate of canopy warming is measured by an infrared thermal imager.

[0033] Logical correlation: It is known that an increase in NPQ (photoprotection activation) is usually accompanied by an increase in non-radiative energy dissipation, leading to a slowdown in the blade warming rate; the system uses historical normal data to train the matrix. and Real-time calculation ;like If the threshold is exceeded, the system automatically switches to "safe spectrum mode" (reduces total light intensity and increases the proportion of far-red light) until the sensor recovers.

[0034] Example 2 Regulation process of leafy vegetables (lettuce) in an environment with sufficient CO2 Scenario description: A plant factory grows lettuce. The ambient CO2 concentration is set at 1000ppm (sufficient), and the natural light is weak, so artificial lighting is required.

[0035] Regulation process: Data acquisition: The multimodal sensing component measured the current NPQ=0.8 (normal range), leaf temperature 22℃, air humidity 65%, and matrix EC normal.

[0036] Status determination: The stress determination module determines that NPQ has not exceeded the threshold and enters the "dynamic optimization strategy".

[0037] Model calculation: Substituting the multi-factor coupling model into the performance calculation module, and considering the sufficient CO2 and suitable temperature, the calculation yields... .

[0038] At this time, the natural light level is 100, and the system calculates that the required supplemental light level is 350.

[0039] The co-correction module estimates the porosity conductance to be normal, and no reduction is required.

[0040] Execution Instruction: The central processing unit sends an instruction to the dynamic spectral execution component to open the red (660nm) and blue (450nm) channels, set the ratio to 4:1, and adjust the intensity to 350 units.

[0041] Sudden stress response: Half an hour later, the temperature of some blades rose to 32°C due to ventilation failure.

[0042] The NPQ was detected to rise rapidly to 2.5 (exceeding the threshold of 1.5).

[0043] The stress determination module immediately triggers the "stress suppression strategy".

[0044] Action: Instantly reduces the proportion of blue light (from 20% to 5%), significantly increases the output of far-red light (730nm), induces non-photochemical quenching to dissipate excess energy, and reduces the total light intensity by 30%.

[0045] Result: After 5 minutes, the NPQ dropped to 1.2, and the system gradually returned to normal illumination.

[0046] Example 3 Regulation process of fruits and vegetables (tomatoes) under conditions of insufficient CO2 and sensor failure Scenario description: Tomatoes are grown in a multi-span greenhouse. The afternoon sunlight is strong, but poor ventilation causes the indoor CO2 concentration to drop to 200ppm (less than 200ppm). In addition, the lens of the chlorophyll fluorescence camera is blocked by water mist, resulting in data loss.

[0047] Regulation process: Fault detection: The central processing unit detects that the fluorescence data is invalid for three consecutive cycles, triggering the fault tolerance module.

[0048] Parameter reconstruction: The system reads data from the infrared thermal imager and finds that the canopy temperature is rising rapidly at a rate of 0.5℃ / min (indicating that the stomata are closing, transpiration cooling is weakening, and light energy is being converted into heat energy).

[0049] Reconstructing the virtual using the EKF algorithm (High-risk status).

[0050] Stomatal limitation judgment: Meanwhile, the CO2 sensor reading was 200 ppm, indicating low air humidity.

[0051] The collaborative correction module estimates porosity based on the formula. Extremely low (close to off).

[0052] Comprehensive decision-making: Despite the sunny weather, based on the reconstruction and low The system determined that the system was under severe light stress and carbon starvation.

[0053] Actions: Completely turn off artificial lighting (to avoid exacerbating the problem) and control the deployment of the shade net; at the same time, adjust the output of different spatial grids in the LED array: keep the high-density area at the top of the canopy, as identified by the 3D scan, off; for the lower areas with less light, only turn on a small amount of far-red light to maintain morphogenesis, without supplementing photosynthesis.

[0054] Result: The system avoided severe light suppression and leaf burn caused by forced supplemental lighting under low CO2 conditions. After ventilation improved, CO2 concentration recovered, and sensor data was restored, the system automatically deactivated the safety mode and resumed normal production.

[0055] In summary, this intelligent lighting control system for greenhouse plant growth achieves a qualitative leap from "experience-based supplemental lighting" to "on-demand lighting" through deep synergy between multimodal physiological perception and multi-environmental factor coupling models. The system can not only identify and suppress light stress in real time based on the non-photochemical quenching coefficient (NPQ) and optimize light energy utilization efficiency by dynamically adjusting the red / blue / far-red light ratio, but also solve the problem of uneven light distribution among plant populations by combining three-dimensional canopy scanning technology. Especially under extreme conditions such as sensor failure, its parameter reconstruction mechanism based on extended Kalman filtering ensures the continuity and robustness of the control strategy. Practical application scenarios show that while significantly improving crop photosynthetic efficiency and stress resistance, the system effectively avoids energy waste and light damage caused by blind supplemental lighting, providing reliable technical support for high-yield, high-quality, and low-carbon production in facility agriculture, and has significant value for widespread application.

[0056] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart lighting control system for greenhouse plant growth, characterized in that: It includes a multimodal sensing component, a central processing unit, and a dynamic spectral execution component, wherein; A multimodal sensing component is used to collect physiological state parameters and environmental factor data of crop canopy in real time. The multimodal sensing component includes a physiological state detection unit for detecting the physiological state of crops. The central processing unit is connected to the multimodal sensing components via signals and has a pre-installed stress determination module and performance calculation module. The dynamic spectral execution component, which is signal-connected to the central processing unit, includes a multi-channel independently addressable LED array, which contains light sources of various different wavelengths; The stress determination module generates physiological state markers based on physiological state parameters; The performance calculation module calls a multi-factor coupling model based on environmental factor data to calculate the dynamic light saturation point; The central processing unit generates control commands containing the driving parameters of each channel light source based on the physiological state markers and dynamic light saturation points, and sends them to the dynamic spectrum execution component.

2. The intelligent lighting control system for greenhouse plant growth according to claim 1, characterized in that: The physiological state parameters include at least the non-photochemical quenching coefficient (NPQ); the environmental factor data include at least carbon dioxide concentration and leaf temperature. The execution logic of the central processing unit includes: When the non-photochemical quenching coefficient (NPQ) exceeds a preset threshold, it is determined to be a photo-stressed state, and a stress suppression strategy is executed first. The stress suppression strategy includes reducing the output ratio of the blue light band light source and increasing the output ratio of the far-infrared light band light source. When the non-photochemical quenching coefficient (NPQ) does not exceed the preset threshold, it is determined to be in a normal state, and a dynamic optimization strategy is executed: based on the current carbon dioxide concentration and leaf temperature, the target photosynthetically active radiation threshold is calculated using a multi-factor coupling model, and the output intensity of the red and blue light bands in the LED array is adjusted according to the difference between the target photosynthetically active radiation threshold and the natural light intensity.

3. The intelligent lighting control system for greenhouse plant growth according to claim 2, characterized in that: The multimodal sensing component also includes an air humidity sensor and a matrix conductivity sensor; The central processing unit is also equipped with a collaborative correction module for estimating crop stomatal conductance based on air humidity and matrix conductivity. When the estimated stomatal conductance is lower than the preset lower limit, the collaborative correction module reduces the target photosynthetically effective radiation threshold in the dynamic optimization strategy to reduce the light source output intensity.

4. The intelligent lighting control system for greenhouse plant growth according to claim 1, characterized in that: The multimodal sensing component also includes a three-dimensional canopy scanning unit for acquiring three-dimensional height distribution data of the crop canopy; The LED array in the dynamic spectral execution component is divided into multiple independently controlled spatial grids; The central processing unit identifies the differences in canopy height within each spatial grid based on the three-dimensional height distribution data, and adjusts the light source output intensity in different spatial grids accordingly, thereby making the distribution of photon flux density on the canopy surface more uniform.

5. The intelligent lighting control system for greenhouse plant growth according to claim 1, characterized in that: The central processing unit also includes a fault-tolerant processing module; When the multimodal sensing component experiences data anomalies, the fault-tolerant processing module reconstructs the virtual physiological state parameters based on the canopy temperature change trend and historical operating data, and switches to a preset safe spectrum mode for operation.

6. The intelligent lighting control system for greenhouse plant growth according to claim 1, characterized in that: The peak wavelength of the red light band in the LED array is 655nm-665nm, the peak wavelength of the blue light band is 445nm-455nm, and the peak wavelength of the far-red light band is 725nm-735nm. The physiological state detection unit is a chlorophyll fluorescence imager or a hyperspectral camera.

Images