Plant light supplementing self-adaptive stable light control method and device

Abstract

The application provides a plant light supplementing self-adaptive stable light control method and device, comprising constructing a distributed multi-spectrum LED light supplementing array; collecting crown layer mixed PAR, natural light PAR distribution and LED junction temperature data in real time through a multi-source photoelectric sensing network to generate a light intensity deviation and a disturbance state vector; based on a feedforward-feedback compound control law, calculating an initial PWM control instruction for inhibiting transient light fluctuation under the satisfaction of plant physiological constraints; based on a photoelectric attenuation dynamics model and a spectral ratio deviation, adaptively modifying the initial instruction from the light decay compensation and spectral reconstruction dimensions to generate multi-spectrum channel target duty cycles; issuing the instruction to execute light modulation, and using a full life cycle light efficiency residual feedback mechanism to perform online rolling correction on the control parameters. The application effectively solves the stable light control problem caused by natural light mutation, LED light decay and long-period operation parameter drift, and realizes systematic stable state control and full life cycle maintenance-free operation of plant light supplementing in three-dimensional space dimensions.

Description

Technical Field

[0001] This invention relates to the field of light environment control technology, specifically to a method and device for adaptive light stabilization control of plant supplemental lighting. Background Technology

[0002] In greenhouses or plant factories, natural light environments often experience drastic transient fluctuations due to factors such as cloudy weather and external shading, leading to uneven light exposure in the plant canopy and causing light stress. To ensure efficient photosynthesis, LED supplemental lighting systems are typically used for artificial intervention.

[0003] However, existing plant supplemental lighting control technologies have the following defects and shortcomings in practical applications: Slow transient response and lack of physiological constraints: Traditional supplemental lighting systems mostly use constant power or simple feedback dimming, which is difficult to cope with transient light stress caused by sudden drops in natural light. Moreover, during the adjustment process, there is a lack of consideration for physiological safety constraints such as plant light compensation point, light saturation point and light adaptation threshold, which can easily lead to excessive supplemental lighting or sudden changes in light, which can have an adverse effect on plant growth.

[0004] Ignoring output degradation caused by light decay and thermal drift: LED chips experience a decrease in luminous efficacy with increasing usage time and junction temperature fluctuations, and the inconsistent decay of different color chips (such as red and blue light) leads to spectral drift. Traditional systems lack dynamic identification and compensation mechanisms for photoelectric degradation, resulting in a significant deviation between the actual output light intensity and spectral ratio from the target reference after long-term operation.

[0005] Full life cycle parameter mismatch: With the aging of light source, changes in environmental temperature and humidity, and the dynamic growth of plant canopy morphology, the control model with fixed parameters will experience phased drift and time-dependent evolution, resulting in a mismatch between the control prediction model and the execution mapping relationship. The system is unable to maintain a high-precision stable light control effect over a long period of time, and its engineering applicability is poor. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a plant supplemental lighting adaptive light stabilization control method and device to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a plant supplemental lighting adaptive light stabilization control method, comprising the following steps: S100: Based on the spatial topology and non-uniform light characteristics of the plant canopy, a distributed multispectral LED supplementary lighting array is constructed. The LED supplementary lighting array consists of several LED supplementary lighting nodes with independent dimming capabilities of multispectral channels, and the initial light intensity and spectral dimming reference of each node are established. S200: Real-time data collection of plant canopy mixed photosynthetic effective radiation, external natural light PAR distribution, and LED node junction temperature status data are obtained through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, high-frequency light intensity deviation and disturbance state vectors are generated. S300: Based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics, a feedforward-feedback composite control law is introduced. Combined with the light intensity deviation state vector, under the condition of satisfying the physiological constraints of the plant light compensation point and light saturation point, the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations is calculated. S400: Based on the LED photoelectric attenuation dynamics model and the multi-spectral channel ratio deviation, the initial PWM control command is adaptively corrected online from two dimensions: total light intensity compensation and multi-channel spectrum reconstruction, generating the target PWM duty cycle for each LED supplementary light node corresponding to different spectral channels. S500: The target duty cycle is sent to drive the LED fill light array to perform dimming action. At the same time, the transient parameters of the feedforward-feedback composite control law and the time-varying parameters of the photoelectric attenuation model are corrected online using the full life cycle luminous efficacy residual feedback mechanism.

[0008] As a further preferred embodiment, in step S100: Based on the three-dimensional topological morphology and shading characteristics of the plant canopy in a greenhouse or plant factory, the LED supplementary lighting nodes are distributed along the vertical height and horizontal extension direction of the canopy to form a matrix-distributed multispectral lighting node group. The constant current drive modules of each node are connected into a collaborative logic whole through a multi-machine group control communication link; The distributed arrangement is used to address the differentiated supplemental lighting needs of different canopy heights, different foliage shading gradients, and different light intensity requirements; the multispectral channel configuration is used to address the spectral preferences of plants at different growth stages or under different light environments.

[0009] As a further preferred option, in step S200: The multi-source optoelectronic sensing network adopts a light environment monitoring system that integrates PAR sensing and spectral analysis, including distributed photosynthetically active radiation sensing units deployed in key parts of the canopy, spectral recognition units facing the natural light channel, and junction temperature feedback units integrated into the LED actuator. The collected data is preprocessed by the edge computing terminal for clock synchronization, noise suppression and feature fusion, and then forms a high-frequency light intensity deviation and disturbance state vector for control triggering.

[0010] As a further preferred option, the specific process of S300 includes: Step 301: In a unified supplemental lighting spatial coordinate system, the plant canopy is discretized into multiple controlled light spot regions, and the external natural light environment is discretized into multiple light source disturbance components, thus constructing a set of controlled canopy regions and a set of disturbance sources. Step 302: Based on the actual mixed photosynthetically active radiation received in the canopy controlled area and the target photosynthetic requirement benchmark, construct a light intensity deviation equation to quantify the difference between the current actual light environment and the target requirement; Step 303: Introduce external disturbance terms and noise terms for natural light fluctuations, calculate the transient rate of change of natural light, and construct a transient light environment state equation that includes mixed total illumination, natural illumination and its rate of change; Step 304: Establish a physiological safety constraint system: constrain the predicted mixed total light intensity to be no less than the plant light compensation point and no more than the light saturation point, while constraining the transient adjustment light intensity change rate to not exceed the plant light adaptation threshold; Step 305: Under physiological safety constraints, construct an objective function that includes light intensity error, disturbance compensation, energy consumption and dimming smoothness, and solve it based on the composite control law to output the initial PWM control command for the current sampling period.

[0011] As a further preferred embodiment, the specific execution flow of steps 302 to 305 includes: Obtain the target total photosynthetically active radiation set at the current plant growth stage. And the total illuminance at the current moment is expressed as ; Get the current natural illuminance Natural illuminance compared to the previous sampling time Calculate the transient rate of change of natural illuminance. ,in The sampling time interval; Based on plant light compensation point threshold and plant light saturation threshold Establish master constraints And based on the maximum allowed dimming step size threshold Establish secondary constraints ,in The rate of change of the instantaneous control quantity; Under the conditions of satisfying the primary and secondary constraints, the proportional, integral, and derivative gain coefficients of the light intensity feedback control loop are introduced. , , and feedforward gain coefficient The light intensity error With the transient rate of change Substituting the values ​​into the feedforward-feedback composite control law, the theoretical PWM duty cycle command is calculated, which is the initial PWM control command. .

[0012] As a further preferred option, the specific process of S400 includes: Step 401: Calculate the deviation between the actual luminous flux and the theoretical luminous flux of the distributed LED array, construct a photoelectric attenuation dynamics model, and generate a real-time luminous attenuation compensation factor; Step 402: Combine the light attenuation compensation factor with the initial PWM control command to calculate the total PWM control command, so that the overall output luminous flux converges to the target value; Step 403: Calculate the ratio deviation of different spectral channels such as red and blue light, construct an objective function for the overall stability of the spectrum, and map the ratio deviation into the channel correction direction through a multi-channel collaborative allocation algorithm to obtain the multi-channel control increment; Step 404: Combine the total PWM control command with the multi-channel control increment to obtain the final PWM target duty cycle, and apply physical constraints to the LED drive system.

[0013] As a further preferred embodiment, the specific execution flow of steps 401 to 404 includes: Get the cumulative lighting time of the LED array and real-time junction temperature Combined with the preset light decay coefficient per unit time Negative temperature coefficient of luminous efficacy at junction temperature Calculate the difference between the current junction temperature and the nominal junction temperature. Generate real-time light decay compensation factor The real-time optical attenuation compensation factor is then subjected to amplitude limiting to ensure it meets the requirements. ,in The maximum light decay compensation factor threshold is determined by the LED rated power redundancy; Obtain the target red-blue light ratio required by the current plant. The measured red and blue light ratio Calculate the spectral ratio deviation ; The initial PWM control command Light attenuation compensation factor after limiting Multiply and sum to calculate the total PWM control command. for: ; Introducing spectral adjustment integral gain coefficient Based on the total PWM control command and the spectral ratio deviation, the PWM command for the red light channel is calculated respectively. and Blu-ray channel PWM instructions ; The red light channel PWM command and the blue light channel PWM command are used as the final target PWM duty cycle, and physical constraints are applied. and .

[0014] As a further optimization, in the S500: The full life cycle luminous efficacy residual feedback mechanism establishes a predictive model of parameter evolution with operating conditions and time by encoding the cumulative operating time and junction temperature status. When the feedforward-feedback composite control law calculates the initial PWM command, it relies on the feedforward gain and feedback PID parameters output by this mechanism; when the optical decay identification and online learning compensation model performs optical decay compensation calculation, this mechanism performs evolution prediction of the optical decay coefficient and thermal drift coefficient and performs residual drive correction when necessary. The residual between the light intensity monitoring observation and the predicted observation formed after the execution of composite control and compensation is used as the update driver for parameter evolution prediction. When the residual evaluation index exceeds the set threshold, online parameter correction is triggered. When the quality of monitoring data is insufficient or residual fluctuations are abnormal, making it impossible to reliably update parameters, an abnormal degradation strategy is initiated: online calibration is paused and the parameters from the previous moment are maintained, or the parameters are switched to conservative baseline parameters.

[0015] As a further preferred embodiment, a plant supplemental lighting adaptive light stabilization control device is provided for implementing the above-mentioned plant supplemental lighting adaptive light stabilization control method, comprising: The sensing and acquisition module is used to collect data on the mixed photosynthetic effective radiation of the plant canopy, the distribution of external natural light PAR, and the junction temperature status of LED nodes in real time through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, it generates high-frequency light intensity deviation and disturbance state vectors. The initial control generation module is used to introduce a feedforward-feedback composite control law based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics. Combined with the light intensity deviation state vector, it calculates the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations under the physiological constraints of the plant light compensation point and light saturation point. The adaptive correction module is used to adaptively correct the initial PWM control command based on the photoelectric attenuation dynamics model and spectral ratio deviation from two dimensions: total light intensity compensation for light attenuation and multi-channel spectral reconstruction, and generate the target duty cycle of each LED supplementary light node for different spectral channels such as red and blue. The drive execution module is used to issue the target duty cycle to drive the LED fill light array to perform dimming actions; Among them, the supplementary lighting nodes in the LED supplementary lighting array are distributed along the vertical height and horizontal extension direction of the canopy, forming a matrix distribution; The full life cycle correction module is used to perform online rolling correction of the transient parameters of the feedforward-feedback composite control law in the initial control generation module and the time-varying parameters of the photoelectric attenuation model in the adaptive correction module by utilizing the full life cycle luminous efficacy residual feedback mechanism.

[0016] This invention provides a method and apparatus for adaptive light stabilization control of plants, which has the following beneficial effects: This invention achieves three-dimensional spatial coordination and precise multispectral allocation by constructing a distributed multispectral LED supplementary lighting array. It introduces a feedforward-feedback composite control law combined with plant physiological constraints to actively suppress transient light fluctuations and ensure plant physiological safety. At the same time, based on light decay identification and online learning compensation models, it overcomes the output attenuation and spectral drift caused by light decay and thermal drift, ensuring long-term stable luminous efficacy. In addition, it innovatively utilizes a full-lifecycle luminous efficacy residual feedback mechanism to perform triggered online correction of control parameters, effectively solving the problem of long-term parameter mismatch and significantly improving the system's engineering applicability and maintenance-free capability. Attached Figure Description

[0017] Figure 1 This is a flowchart of the plant supplemental lighting adaptive light stabilization control method of the present invention; Figure 2 This is a schematic diagram of the plant supplemental lighting adaptive light stabilization control method of the present invention; Figure 3 This is a block diagram of the plant supplemental lighting adaptive light stabilization control device of the present invention. Detailed Implementation

[0018] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0019] like Figure 1 and 2 As shown, this embodiment of the invention provides a plant supplemental lighting adaptive light stabilization control method, including the following steps: Step 100: Based on the spatial topology and non-uniform light characteristics of the plant canopy, a distributed multispectral LED supplementary lighting array is constructed. This array consists of several LED supplementary lighting nodes with independent dimming capabilities for multispectral channels, and the initial light intensity and spectral dimming reference of each node are established. Step 200: Real-time data on mixed photosynthetically active radiation (PAR) of plant canopy, distribution of external natural light PAR, and junction temperature of LED nodes are collected through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, high-frequency light intensity deviation and disturbance state vectors are generated. Step 300: Based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics, a feedforward-feedback composite control law is introduced. Combined with the light intensity deviation state vector, under the condition of satisfying the physiological constraints of the plant light compensation point and light saturation point, the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations is calculated. Step 400: Based on the LED photoelectric attenuation dynamics model and the multispectral channel ratio deviation, the initial PWM control command is adaptively corrected online from two dimensions: total light intensity compensation for light attenuation and multi-channel spectrum reconstruction, to generate the target PWM duty cycle for each LED supplementary lighting node corresponding to different spectral channels. Step 500: Send the target duty cycle to drive the LED filler array to perform dimming action. At the same time, use the full life cycle luminous efficacy residual feedback mechanism to perform online rolling correction on the transient parameters of the feedforward-feedback composite control law and the time-varying parameters of the photoelectric attenuation model.

[0020] Based on the three-dimensional topology and shading characteristics of plant canopies in greenhouses or plant factories, LED supplemental lighting nodes are distributed along the vertical height and horizontal extension of the canopy, forming a distributed multispectral LED supplemental lighting array that matches the canopy morphology. This array spatially manifests as a matrix-distributed group of multispectral lighting nodes, with the constant current drive modules of each node connected into a collaborative logical whole via a multi-machine group control communication link. Its function is to construct a spatially collaborative light environment control system. By independently or in conjunction with other LED supplemental lighting nodes in different spatial locations, it effectively suppresses localized light stress in the canopy and uneven spatial light distribution, thereby achieving systematic steady-state control of plant supplemental lighting in a three-dimensional spatial dimension.

[0021] Among them, the distributed arrangement is used to correspond to the differentiated supplemental lighting needs of different canopy heights, different foliage shading gradients and different light intensity demand sections, so as to improve the ability to identify and adjust the effective photosynthetic radiation gradient and light intensity mutation nodes in space; the multispectral channel configuration is used to correspond to the spectral preferences of plants at different growth stages or under different light environments, so as to achieve multi-point spectral linkage adjustment and precise allocation of photon flux in the same layer.

[0022] The multi-source optoelectronic sensing network employs a light environment monitoring system that integrates PAR sensing and spectral analysis. It includes distributed photosynthetically active radiation sensing units deployed at key locations in the canopy, spectral recognition units oriented towards natural light channels, and junction temperature feedback units integrated into the LED actuator. Data collected by the multi-source optoelectronic sensing network undergoes preprocessing at the edge computing terminal, including clock synchronization, noise suppression, and feature fusion, to form high-frequency light intensity deviation and disturbance state vectors for control triggering. This enhances the system's ability to rapidly identify localized light surges, canopy shading, and abnormal operating conditions.

[0023] This embodiment addresses the challenge of stable light control caused by fluctuations in natural light and the degradation of LED performance. While satisfying the physiological constraints of plant photosynthesis, it introduces a feedforward-feedback composite control law. This law outputs an initial PWM control command for the LED supplementary lighting node to meet the plant's light adaptation constraints within each sampling period, enabling adaptive decision-making and proactive intervention in the plant's light environment. Specifically, the steps include: Step 301: In a unified supplementary lighting spatial coordinate system, the plant canopy is discretized into multiple controlled light spot regions, and the external natural light environment is discretized into multiple light source disturbance components to construct a set of canopy controlled regions and a set of disturbance sources. Step 302: Based on the actual mixed photosynthetically active radiation received in the canopy controlled area and the target photosynthetic requirement benchmark, construct a light intensity deviation equation to quantify the difference between the current actual light environment and the target requirement. Step 303: Introduce external disturbance terms and noise terms for natural light fluctuations, calculate the transient rate of change of natural light, and construct a transient light environment state equation that includes total mixed illumination, natural illumination and its rate of change. Step 304, establish a physiological safety constraint system: constrain the predicted mixed total light intensity to be no less than the plant light compensation point and no more than the light saturation point, while constraining the transient adjustment light intensity change rate to not exceed the plant light adaptation threshold; Step 305: Under physiological safety constraints, construct an objective function that includes light intensity error, disturbance compensation, energy consumption and dimming smoothness, and solve it based on the composite control law to output the initial PWM control command for the current sampling period.

[0024] In the specific implementation process, before the system runs, a unified three-dimensional rectangular coordinate system is established based on the plant canopy morphology and the relative position of the supplemental lighting device. The plant canopy is discretized into m controlled segments. The total mixed illuminance of the i-th controlled segment at time t is expressed as... ; Let the target total photosynthetically active radiation for the current plant growth stage be... Then, the light intensity deviation equation is constructed as follows: ; in, Let t be the total illumination error parameter at time t; A constant target illumination reference; This is the measured total mixed illuminance of the canopy.

[0025] To address sudden changes in natural light caused by cloudy weather or shading, the transient rate of change of natural light intensity is calculated as follows: ; in, The transient rate of change of natural light intensity; The current natural illuminance; The natural illuminance at the previous sampling time; This represents the sampling time interval.

[0026] Based on the above light environment state equation, a physiological safety constraint system is constructed to ensure that the mixed total light intensity meets the physiological safety constraints of plants (i.e., the light intensity meets the compensation point and saturation point constraints): ; in, This is the threshold for plant light compensation. This is the light saturation threshold for plants.

[0027] After completing the primary constraint, a secondary constraint for smooth light intensity transition is applied. A sensitive index is constructed based on the rate of change, constraining the rate of change of illumination to not exceed an allowable threshold, i.e.: ; in, The rate of change of the instantaneous control quantity. The maximum allowed dimming step size threshold.

[0028] After completing the constraint system construction, a feedforward-feedback composite light intensity controller was designed, and the initial PWM control command for compensating for light intensity fluctuations was calculated as follows: ; in, This is the output of the light intensity controller (i.e., the theoretical PWM duty cycle command). , , These are the proportional, integral, and derivative gain coefficients of the light intensity feedback control loop, respectively. This is the first derivative of the total illumination error parameter; This is the feedforward gain coefficient.

[0029] Specifically, when ambient light suddenly drops, the feedforward component directly drives the LED to increase its output, achieving "zero-delay" compensation for sudden changes in ambient light; the feedback component is used to eliminate the steady-state error of the system. However, this initial instruction has not yet taken into account the luminous efficacy decay of the LED chip itself due to long-term operation and the problem of inconsistent light decay across multiple channels.

[0030] In step 3, although the initial PWM control command has been calculated based on the feedforward-feedback composite mechanism, and the target dimming value of each LED supplementary lighting node has been determined at the transient level, the implementation of this command in an actual LED array system still needs to consider factors such as output attenuation caused by LED light decay and thermal drift, and spectral drift caused by inconsistent attenuation of different color chips. Therefore, based on the output results of step 3, this embodiment constructs a light decay identification and online learning compensation model, transforming the theoretical control quantity into a stably implementable adaptive control form starting from light intensity correction and spectral reconstruction. Specifically, this includes the following steps: Step 401: Calculate the deviation between the actual luminous flux and the theoretical luminous flux of the distributed LED array, construct a photoelectric attenuation dynamics model, and generate a real-time luminous attenuation compensation factor; Step 402: Combine the light attenuation compensation factor with the initial PWM control command to calculate the total PWM control command, so that the overall output luminous flux converges to the target value; Step 403: Calculate the ratio deviation of different spectral channels such as red and blue light, construct an objective function for the overall stability of the spectrum, and map the ratio deviation into the channel correction direction through a multi-channel collaborative allocation algorithm to obtain the multi-channel control increment. Step 404: Combine the total PWM control command with the multi-channel control increment to obtain the final PWM target duty cycle, and apply physical constraints to the LED drive system.

[0031] Specifically, regarding light decay and thermal drift, the luminous efficacy of LEDs decreases with increasing usage time and junction temperature fluctuations. The cumulative lighting time of the LED array is obtained. and real-time junction temperature Define the light decay compensation factor The optical decay compensation factor is calculated as follows: ; in, The light decay coefficient per unit time; This refers to the cumulative lighting time; It has a negative temperature coefficient for junction-temperature luminous efficacy; This is the difference between the current junction temperature and the nominal junction temperature. This is the real-time optical attenuation compensation factor, indicating the need for additional power compensation.

[0032] To prevent LED overload damage due to overcompensation, a limiting procedure is added to the compensation factor in the measurement and control system. ; in, The maximum permissible light decay compensation factor threshold is determined by the LED rated power redundancy.

[0033] Let the target red-blue light ratio required by the current plant be... Obtain the measured red and blue light ratio The spectral ratio deviation equation is constructed as follows: ; in, This is a parameter for spectral ratio deviation. To achieve the target spectral ratio; The measured spectral ratio is used.

[0034] Combined with initial PWM control instructions Light decay compensation factor and spectral ratio deviation The overall adaptive stabilization control command and multi-channel allocation strategy were designed. Overall PWM control command. for: ; Red light channel PWM command With Blu-ray channel PWM command They are respectively: ; ; in, This is the total PWM control command after adaptive correction for optical attenuation; and These are the PWM duty cycle commands sent to the red and blue light drivers, respectively. This is the integral gain coefficient for spectral adjustment.

[0035] Based on the above corrections, and by imposing physical constraints on each LED fill light node, namely: ; ; The above constraints are used to ensure that the control output meets the capability boundaries and safety requirements of the LED driver system.

[0036] The light decay identification and online learning compensation model proposed in this embodiment, compared with the traditional control method driven by constant power or constant current, can realize automatic sensing and compensation of light decay during long-term operation, and eliminate spectral drift caused by light decay differences through multi-channel collaborative allocation; by suppressing adverse effects such as light intensity attenuation and ratio imbalance, it significantly enhances the continuity, stability and feasibility of control output, thereby effectively ensuring the continuous and efficient operation of plant photosynthesis.

[0037] To address the issue of mismatch between the control prediction model and execution mapping caused by the phased drift and time-dependent evolution of control parameters during the entire life cycle of plant supplemental lighting due to LED light source aging, changes in environmental temperature and humidity, and plant canopy morphology growth, this embodiment proposes a life-cycle luminous efficacy residual feedback mechanism. This method establishes a prediction model of parameter evolution over time and operating conditions by encoding the cumulative operating time and junction temperature status. It then combines this with the residuals from luminous intensity feedback data to achieve triggered online correction, thereby ensuring the effectiveness, consistency, and engineering applicability of the feedforward-feedback composite control law in step 3 and the luminous attenuation identification and online learning compensation model in step 4 under evolving operating conditions.

[0038] In this embodiment, the full lifecycle luminous efficacy residual feedback mechanism, the feedforward-feedback composite control law in step 3, and the luminous attenuation identification and online learning compensation model in step 4 form a closed-loop coupling relationship between parameters and feedback. When the feedforward-feedback composite control law calculates the initial PWM command, it relies on the feedforward gain and feedback PID parameters output by this mechanism to ensure effective characterization of the current lighting environment. When the luminous attenuation identification and online learning compensation model performs luminous attenuation compensation calculations, this mechanism adjusts the luminous attenuation coefficient... With thermal drift coefficient Evolutionary prediction is performed, and residual-driven correction is initiated when necessary. The residual between the light intensity monitoring observation and the predicted observation, formed after the execution of composite control and compensation, serves as the update driver for parameter evolution prediction. When the residual evaluation index (specifically, the root mean square error or absolute deviation between the actual hybrid PAR observation and the model predicted observation) exceeds a set threshold, online parameter correction is triggered to suppress parameter mismatch caused by factors such as light source aging and junction temperature fluctuations, thus forming a closed-loop link of parameter evolution prediction—transient control—cooperative reconstruction—residual feedback—parameter correction. Therefore, the various mechanisms constitute a closed-loop coupling relationship, ensuring the consistency of the control objectives and engineering applicability of this invention throughout its entire lifecycle.

[0039] In step 5, the target duty cycle is sent to drive the LED supplemental lighting array to perform dimming, while periodically executing parameter evolution prediction and correction processes. When the monitoring data quality is insufficient or residual fluctuations abnormally prevent reliable parameter updates, an abnormal degradation strategy is initiated: online correction is paused and the parameters from the previous moment are maintained, or the parameters are switched to a conservative baseline to ensure the stability of the control strategy and the safety of plant growth. This method enables continuous prediction and triggered online correction of key control parameters as the light source ages and deteriorates, thereby avoiding mismatches in fixed parameters throughout the entire cycle and improving the engineering applicability and maintenance-free capability of plant supplemental lighting control.

[0040] like Figure 3 As shown, this embodiment also provides a plant supplemental lighting adaptive light stabilization control device for implementing the above-mentioned plant supplemental lighting adaptive light stabilization control method, including: The sensing and acquisition module is used to collect data on the mixed photosynthetic effective radiation of the plant canopy, the distribution of external natural light PAR, and the junction temperature status of LED nodes in real time through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, it generates high-frequency light intensity deviation and disturbance state vectors. The initial control generation module is used to introduce a feedforward-feedback composite control law based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics. Combined with the light intensity deviation state vector, it calculates the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations under the physiological constraints of the plant light compensation point and light saturation point. The adaptive correction module is used to adaptively correct the initial PWM control command based on the photoelectric attenuation dynamics model and spectral ratio deviation from two dimensions: total light intensity compensation for light attenuation and multi-channel spectral reconstruction, and generate the target duty cycle of each LED supplementary light node for different spectral channels such as red and blue. The drive execution module is used to issue the target duty cycle to drive the LED fill light array to perform dimming actions; Among them, the supplementary lighting nodes in the LED supplementary lighting array are distributed along the vertical height and horizontal extension direction of the canopy, forming a matrix distribution; The full life cycle correction module is used to perform online rolling correction of the transient parameters of the feedforward-feedback composite control law in the initial control generation module and the time-varying parameters of the photoelectric attenuation model in the adaptive correction module by utilizing the full life cycle luminous efficacy residual feedback mechanism.

[0041] 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 method for adaptive light stabilization control of plants through supplemental lighting, characterized in that, Includes the following steps: S100: Based on the spatial topology and non-uniform light characteristics of the plant canopy, a distributed multispectral LED supplementary lighting array is constructed. The LED supplementary lighting array consists of several LED supplementary lighting nodes with independent dimming capabilities of multispectral channels, and the initial light intensity and spectral dimming reference of each node are established. S200: Real-time data collection of plant canopy mixed photosynthetic effective radiation, external natural light PAR distribution, and LED node junction temperature status data are obtained through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, high-frequency light intensity deviation and disturbance state vectors are generated. S300: Based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics, a feedforward-feedback composite control law is introduced. Combined with the light intensity deviation state vector, under the condition of satisfying the physiological constraints of the plant light compensation point and light saturation point, the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations is calculated. S400: Based on the LED photoelectric attenuation dynamics model and the multi-spectral channel ratio deviation, the initial PWM control command is adaptively corrected online from two dimensions: total light intensity compensation and multi-channel spectrum reconstruction, generating the target PWM duty cycle for each LED supplementary light node corresponding to different spectral channels. S500: The target duty cycle is sent to drive the LED fill light array to perform dimming action. At the same time, the transient parameters of the feedforward-feedback composite control law and the time-varying parameters of the photoelectric attenuation model are corrected online using the full life cycle luminous efficacy residual feedback mechanism.

2. The plant supplemental lighting adaptive light stabilization control method according to claim 1, characterized in that, In step S100: Based on the three-dimensional topological morphology and shading characteristics of the plant canopy in a greenhouse or plant factory, the LED supplementary lighting nodes are distributed along the vertical height and horizontal extension direction of the canopy to form a matrix-distributed multispectral lighting node group. The constant current drive modules of each node are connected into a collaborative logic whole through a multi-machine group control communication link; The distributed arrangement is used to address the differentiated supplemental lighting needs of different canopy heights, different foliage shading gradients, and different light intensity requirements; the multispectral channel configuration is used to address the spectral preferences of plants at different growth stages or under different light environments.

3. The plant supplemental lighting adaptive light stabilization control method according to claim 1, characterized in that, In step S200: The multi-source optoelectronic sensing network adopts a light environment monitoring system that integrates PAR sensing and spectral analysis, including distributed photosynthetically active radiation sensing units deployed in key parts of the canopy, spectral recognition units facing the natural light channel, and junction temperature feedback units integrated into the LED actuator. The collected data is preprocessed by the edge computing terminal for clock synchronization, noise suppression and feature fusion, and then forms a high-frequency light intensity deviation and disturbance state vector for control triggering.

4. The plant supplemental lighting adaptive light stabilization control method according to claim 1, characterized in that, The specific process of S300 includes: Step 301: In a unified supplemental lighting spatial coordinate system, the plant canopy is discretized into multiple controlled light spot regions, and the external natural light environment is discretized into multiple light source disturbance components, thus constructing a set of controlled canopy regions and a set of disturbance sources. Step 302: Based on the actual mixed photosynthetically active radiation received in the canopy controlled area and the target photosynthetic requirement benchmark, construct a light intensity deviation equation to quantify the difference between the current actual light environment and the target requirement; Step 303: Introduce external disturbance terms and noise terms for natural light fluctuations, calculate the transient rate of change of natural light, and construct a transient light environment state equation that includes mixed total illumination, natural illumination and its rate of change; Step 304: Establish a physiological safety constraint system: constrain the predicted mixed total light intensity to be no less than the plant light compensation point and no more than the light saturation point, while constraining the transient adjustment light intensity change rate to not exceed the plant light adaptation threshold; Step 305: Under physiological safety constraints, construct an objective function that includes light intensity error, disturbance compensation, energy consumption and dimming smoothness, and solve it based on the composite control law to output the initial PWM control command for the current sampling period.

5. The plant supplemental lighting adaptive light stabilization control method according to claim 1, characterized in that, The specific execution flow of steps 302 to 305 includes: Obtain the target total photosynthetically active radiation set at the current plant growth stage. And the total illuminance at the current moment is expressed as ; Get the current natural illuminance Natural illuminance compared to the previous sampling time Calculate the transient rate of change of natural illuminance. ,in The sampling time interval; Based on plant light compensation point threshold and plant light saturation threshold Establish master constraints And based on the maximum allowed dimming step size threshold Establish secondary constraints ,in The rate of change of the instantaneous control quantity; Under the conditions of satisfying the primary and secondary constraints, the proportional, integral, and derivative gain coefficients of the light intensity feedback control loop are introduced. , , and feedforward gain coefficient The light intensity error With the transient rate of change Substituting the values ​​into the feedforward-feedback composite control law, the theoretical PWM duty cycle command is calculated, which is the initial PWM control command. .

6. The plant supplemental lighting adaptive light stabilization control method according to claim 5, characterized in that, The specific process of S400 includes: Step 401: Calculate the deviation between the actual luminous flux and the theoretical luminous flux of the distributed LED array, construct a photoelectric attenuation dynamics model, and generate a real-time luminous attenuation compensation factor; Step 402: Combine the light attenuation compensation factor with the initial PWM control command to calculate the total PWM control command, so that the overall output luminous flux converges to the target value; Step 403: Calculate the ratio deviation of different spectral channels such as red and blue light, construct an objective function for the overall stability of the spectrum, and map the ratio deviation into the channel correction direction through a multi-channel collaborative allocation algorithm to obtain the multi-channel control increment; Step 404: Combine the total PWM control command with the multi-channel control increment to obtain the final PWM target duty cycle, and apply physical constraints to the LED drive system.

7. The plant supplemental lighting adaptive light stabilization control method according to claim 6, characterized in that, The specific execution flow of steps 401 to 404 includes: Get the cumulative lighting time of the LED array and real-time junction temperature Combined with the preset light decay coefficient per unit time Negative temperature coefficient of luminous efficacy at junction temperature Calculate the difference between the current junction temperature and the nominal junction temperature. Generate real-time light decay compensation factor The real-time optical attenuation compensation factor is then subjected to amplitude limiting to ensure it meets the requirements. ,in The maximum light decay compensation factor threshold is determined by the LED rated power redundancy; Obtain the target red-blue light ratio required by the current plant. The measured red and blue light ratio Calculate the spectral ratio deviation ; The initial PWM control command Light attenuation compensation factor after limiting Multiply and sum to calculate the total PWM control command. for: ; Introducing spectral adjustment integral gain coefficient Based on the total PWM control command and the spectral ratio deviation, the PWM command for the red light channel is calculated respectively. and Blu-ray channel PWM instructions ; The red light channel PWM command and the blue light channel PWM command are used as the final target PWM duty cycle, and physical constraints are applied. and .

8. The plant supplemental lighting adaptive light stabilization control method according to claim 1, characterized in that, In S500: The full life cycle luminous efficacy residual feedback mechanism establishes a predictive model of parameter evolution with operating conditions and time by encoding the cumulative operating time and junction temperature status. When the feedforward-feedback composite control law calculates the initial PWM command, it relies on the feedforward gain and feedback PID parameters output by this mechanism; when the optical decay identification and online learning compensation model performs optical decay compensation calculation, this mechanism performs evolution prediction of the optical decay coefficient and thermal drift coefficient and performs residual drive correction when necessary. The residual between the light intensity monitoring observation and the predicted observation formed after the execution of composite control and compensation is used as the update driver for parameter evolution prediction. When the residual evaluation index exceeds the set threshold, online parameter correction is triggered. When the quality of monitoring data is insufficient or residual fluctuations are abnormal, making it impossible to reliably update parameters, an abnormal degradation strategy is initiated: online calibration is paused and the parameters from the previous moment are maintained, or the parameters are switched to conservative baseline parameters.

9. A plant supplemental lighting adaptive light stabilization control device, used to implement the plant supplemental lighting adaptive light stabilization control method according to any one of claims 1-8, characterized in that, include: The sensing and acquisition module is used to collect data on the mixed photosynthetic effective radiation of the plant canopy, the distribution of external natural light PAR, and the junction temperature status of LED nodes in real time through a multi-source photoelectric sensing network. After spatiotemporal synchronization and filtering preprocessing, it generates high-frequency light intensity deviation and disturbance state vectors. The initial control generation module is used to introduce a feedforward-feedback composite control law based on the plant photoresponse dynamics benchmark and natural light disturbance characteristics. Combined with the light intensity deviation state vector, it calculates the initial PWM control command for each LED supplementary lighting node to suppress transient light fluctuations under the physiological constraints of the plant light compensation point and light saturation point. The adaptive correction module is used to adaptively correct the initial PWM control command based on the photoelectric attenuation dynamics model and spectral ratio deviation from two dimensions: total light intensity compensation for light attenuation and multi-channel spectral reconstruction, and generate the target duty cycle of each LED supplementary light node for different spectral channels such as red and blue. The drive execution module is used to issue the target duty cycle to drive the LED fill light array to perform dimming actions; Among them, the supplementary lighting nodes in the LED supplementary lighting array are distributed along the vertical height and horizontal extension direction of the canopy, forming a matrix distribution; The full life cycle correction module is used to perform online rolling correction of the transient parameters of the feedforward-feedback composite control law in the initial control generation module and the time-varying parameters of the photoelectric attenuation model in the adaptive correction module by utilizing the full life cycle luminous efficacy residual feedback mechanism.

Images