Greenhouse light quality dynamic regulation and control system and method based on photonic crystal chip

Abstract

The invention provides a greenhouse light quality dynamic regulation and control system and method based on a photonic crystal chip. The method comprises the following steps: acquiring a parallel spectrum task sequence for an integrated layout; simulating a spectrum output process of the multi-layer photonic crystal structure according to the time sequence control protocol; if the deviation between the simulated spectrum output and the target spectrum distribution data exceeds a preset threshold value, adjusting the wave band modulation intensity of the independent subtask; whether the adjusted deviation meets the requirement or not is judged, and a final regulation and control parameter set is obtained; extracting a real-time switching rule from the final regulation and control parameter set; determining whether the light quality change triggers the real-time switching rule or not; if the light quality change triggers the real-time switching rule, obtaining a dynamic regulation and control instruction; driving the multi-layer photonic crystal structure to execute the parallel spectrum task through the dynamic regulation and control instruction; and acquiring light quality feedback data after the parallel spectrum task is executed.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a dynamic control system and method for greenhouse light quality based on photonic crystal chips. Background Technology

[0002] Greenhouse agriculture, as a crucial component of modern agriculture, plays an indispensable role in improving crop yield and quality. Especially given the critical role of light conditions in plant growth, precise control of light quality has become a core issue for promoting efficient agricultural development. Light quality regulation not only affects crop growth rate but also directly relates to nutrient accumulation and disease resistance. Therefore, research on dynamic light quality control systems for greenhouses based on photonic crystal chips is particularly important. This research aims to meet the diverse light environment requirements of crops at different growth stages through technological innovation, providing more intelligent and flexible solutions for agricultural production. Current methods for controlling light quality largely rely on traditional lighting equipment or simple combinations of light sources, which often fall short when dealing with complex environmental demands. This is especially true in greenhouses, where different crops or different growth stages of the same crop have vastly different spectral requirements, and existing technologies struggle to achieve flexible switching and precise matching of multiple spectra on a single device. This limitation leads to low efficiency in light resource utilization and fails to truly meet the urgent needs of modern agriculture for refined management.

[0003] Against this backdrop, photonic crystal chip-based control systems face significant technical challenges. As a device capable of precisely manipulating light wavelengths, the core difficulty of photonic crystal chips lies in achieving independent modulation of multiple spectra within a limited chip space. This challenge is first manifested in the complexity of the chip structure design, as multiple layers need to be integrated within a tiny area, each responsible for controlling light in different wavelengths. Interference between different layers can easily lead to spectral output distortion. With increasing structural complexity, how to achieve parallel processing of different spectral tasks on the same chip becomes another pressing issue. For example, in greenhouses, enhanced blue light is needed during the seedling stage to promote leaf growth, while enriched red light is needed during the flowering stage to stimulate flower bud differentiation. If the chip cannot flexibly switch spectral output modes at different times or in different spaces, the light effect will mismatch with the crop's needs, thus affecting growth efficiency.

[0004] Therefore, how to achieve efficient integration of multi-layer structures on photonic crystal chips and ensure parallel processing and precise control of multiple spectral tasks has become a key problem that urgently needs to be solved in this research. The solution to this problem will directly determine whether the greenhouse light quality control system can truly adapt to the diversified needs of modern agriculture and provide optimal light environment support for crop growth. Summary of the Invention

[0005] This invention provides a dynamic control system and method for greenhouse light quality based on photonic crystal chips, mainly comprising: Obtain a parallel spectral task sequence for the integrated layout; Simulate the spectral output process of a multilayer photonic crystal structure according to the timing control protocol; If the deviation between the simulated spectral output and the target spectral distribution data exceeds a preset threshold, the band modulation intensity of the independent subtask is adjusted. Determine whether the adjusted deviation meets the requirements to obtain the final set of control parameters; Extract real-time switching rules from the final set of control parameters; Determine whether the change in light quality triggers the real-time switching rule; If the change in light quality triggers the real-time switching rule, then a dynamic control command is obtained; The multilayer photonic crystal structure is driven to perform the parallel spectral task by the dynamic control commands; Obtain the optical quality feedback data after executing the parallel spectral task.

[0006] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a method for optimizing and controlling the greenhouse light environment based on a multilayer photonic crystal structure. Addressing the problem of mismatch between dynamic changes in light quality and the needs of crop growth stages in greenhouses, it proposes a solution integrating parallel spectral task sequence generation, spectral output simulation, and real-time control. This invention acquires parallel spectral task sequences, simulates the spectral output of the multilayer photonic crystal structure, and adjusts the band modulation intensity based on deviations in the target spectral distribution data to generate a final control parameter set and real-time switching rules. When changes in light quality trigger the switching rules, dynamic control commands drive the structure to execute tasks, and the greenhouse light environment configuration is optimized through matching analysis between light quality feedback data and the target spectral distribution. This invention achieves precise control of spectral output and dynamic adaptation to crop growth needs, significantly improving the support effect of the greenhouse light environment on crop growth and providing an efficient and intelligent means of light environment management for modern agriculture. Attached Figure Description

[0007] Figure 1 This is a flowchart of a greenhouse light quality dynamic control system and method based on a photonic crystal chip according to the present invention.

[0008] Figure 2 This is a schematic diagram of a greenhouse light quality dynamic control system and method based on a photonic crystal chip according to the present invention.

[0009] Figure 3This is another schematic diagram of a greenhouse light quality dynamic control system and method based on a photonic crystal chip according to the present invention. Detailed Implementation

[0010] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0011] like Figures 1-3 This embodiment of a dynamic control system and method for greenhouse light quality based on photonic crystal chips may specifically include: S101. Obtain the parallel spectral task sequence for the integrated layout.

[0012] Based on the business content and extracted relevant attributes, the following business solution is generated, revolving around the generation of integrated layout and parallel spectral task sequences, with each step forming a tightly logical chain: For the integrated layout, an initial layout scheme is obtained from a pre-established layout template library. The layout templates are initially screened to determine layout structures that meet the requirements of parallel spectral processing. Based on the screened layout structures, a corresponding set of spectral analysis tasks is obtained. The task set is categorized and organized using a data acquisition module, resulting in categorized task groups. For each categorized task group, resource scheduling is performed using a parallel processing mechanism. If the workload of a group exceeds a preset threshold, the tasks are split into sub-task units. Based on the sub-task units, a preliminary task sequence is generated. The task units are sorted by time using a sequence generation tool to determine the ordered task execution order. For the ordered task execution order, the dependencies between tasks are analyzed using an information integration module. If dependency conflicts are found between tasks, the execution order is adjusted to obtain an optimized task sequence. Based on the optimized task sequence and combined with layout optimization strategies, the integrated layout is dynamically adjusted. Resource allocation is balanced using a layout adjustment tool to determine the final layout configuration. For the final layout configuration, a parallel spectral task sequence is generated. The sequence is then bound to the layout configuration via the task allocation module to obtain an executable task sequence scheme.

[0013] For example, in the processing of integrated layouts, an initial scheme can be obtained from a pre-established layout template library. Assuming the template library contains multiple layout structures, a grid layout supporting multi-task parallelism is initially selected based on the requirements of parallel spectral processing. This layout can divide the processing unit into multiple independent regions, each region corresponding to a set of spectral analysis tasks. Through selection, it is determined that this layout supports the processing capacity of at least 10 parallel tasks.

[0014] Specifically, after acquiring the task set, the data acquisition module categorizes and organizes the tasks. Assuming there are 100 spectral analysis tasks, they are divided into a high-priority group (30 tasks) and a low-priority group (70 tasks). The categorization can be based on the urgency of the tasks or the size of the data. This categorization helps ensure rational resource scheduling and improves processing efficiency.

[0015] In one embodiment, a parallel processing mechanism is used for resource scheduling for the categorized task groups. If the number of tasks in the high-priority group exceeds a preset threshold of 20 tasks, it is split into two sub-units of 30 tasks, each containing 15 tasks. These split task units are easily allocated to different processing nodes, avoiding resource overload and ensuring system stability.

[0016] For example, when generating an initial task sequence, the sequence generation tool sorts the task units by time. Suppose two sub-units in a high-priority group require 2 hours and 3 hours to complete, respectively; the tool will prioritize the 2-hour task unit to optimize the overall time. This sorting method effectively reduces waiting time and improves task workflow efficiency.

[0017] Specifically, the information integration module analyzes task dependencies based on the sorted task execution order. If it finds a data dependency between two tasks, where the output of the former is the input of the latter, it adjusts the execution order to ensure the former completes first. This optimization avoids conflicts and ensures the logical consistency of the task sequence.

[0018] In one embodiment, the integrated layout is dynamically adjusted based on the optimized task sequence and a layout optimization strategy. Assuming uneven resource allocation in the initial mesh layout, the layout adjustment tool rebalances, allocating more computing resources to areas with weaker processing power, ultimately determining that the load balance for each area is between 80% and 90%. This adjustment improves resource utilization.

[0019] For example, after the final layout configuration is generated, the parallel spectral task sequences are bound to the layout through the task allocation module. Assuming there are 10 task sequences corresponding to 10 layout regions, after binding, each region has a clearly defined execution task and time schedule, forming an executable plan. This binding ensures precise matching of tasks and resources, significantly improving execution efficiency and system stability.

[0020] S102. Simulate the spectral output process of the multilayer photonic crystal structure according to the timing control protocol.

[0021] Initial incident light information is obtained through a pre-established wavelength sequence. The transmission matrix method is used to calculate the first-layer transmittance and reflectance data for the first-layer material parameters. Using the first-layer transmittance and reflectance data as input, the second-layer transmittance and reflectance data are calculated in conjunction with the second-layer material parameters. Based on the current cumulative transmittance and reflectance data, the next-layer material parameters are sequentially input under time control to obtain the next-layer transmittance and reflectance data. If the current layer sequence has not reached the last layer, the cumulative transmittance and reflectance data are continuously updated using the transmission matrix method. When the layer sequence reaches the last layer, the transmission spectrum curve of the complete multilayer photonic crystal is determined using the final cumulative transmittance data. Based on the transmission spectrum curve of the complete multilayer photonic crystal and the phase matching results, the final spectral simulation results are output.

[0022] Obtaining initial incident light information through a pre-established wavelength sequence is the starting point for multilayer photonic crystal simulation.

[0023] For example, in the visible light band, equally spaced wavelengths from 400 nm to 700 nm can be preset as the incident spectrum sequence, which can cover most practical optical application needs.

[0024] In one embodiment, for infrared detector design, the wavelength sequence can be extended to the 800 nm to 2500 nm range to match the characteristic spectral lines of a specific target. The core computational step is to calculate the first-layer transmittance and reflectance data based on the first-layer material parameters using the transfer matrix method.

[0025] Specifically, the first layer is usually made of a high refractive index material such as titanium dioxide, and its thickness is set to 60 nanometers. When the incident angle is zero degrees, the transmittance and reflectance values ​​of the layer at each wavelength can be obtained by matrix multiplication.

[0026] It should be noted that this step lays the foundation for subsequent interlayer light field transmission, ensuring that the energy conservation relationship is strictly satisfied in each layer. By using the transmittance and reflectance data of the first layer as input, and combining them with the material parameters of the second layer to perform calculations for the second layer, continuous propagation of light at the heterogeneous interface can be achieved.

[0027] For example, the second layer uses silicon dioxide, a material with a low refractive index, with a thickness of 90 nanometers. The field strength output from the previous layer is used as the incident field, and the transmission matrix operation is performed to obtain new transmission and reflection data.

[0028] In one possible implementation, increasing the thickness of the second layer to 120 nanometers could result in stronger interference peaks at specific wavelengths, contributing to a narrower bandgap. The key to this layer-by-layer recursion lies in sequentially inputting the material parameters of the next layer under time-controlled conditions, based on the current cumulative transmittance and reflectance data.

[0029] Preferably, the cumulative result is updated immediately after each layer of calculation is completed, and it is determined whether the preset total number of layers has been reached.

[0030] For example, when the fifth layer is accumulated, the overall transmittance curve begins to show obvious periodic oscillation characteristics, indicating that the bandgap effect of the multilayer structure is gradually emerging. When the layer sequence reaches the last layer, the transmission spectrum curve of the complete multilayer photonic crystal is determined by the final accumulated transmittance data.

[0031] For example, in a typical design with a ten-layer alternating structure, the final transmission spectrum forms a reflection band approximately 40 nm wide near 550 nm, while maintaining high transmittance in the 450 nm and 650 nm bands, demonstrating good spectral selectivity. Based on the transmission spectrum curves and phase-matching results of the complete multilayer photonic crystal, the final spectral simulation results are output.

[0032] In one embodiment, if the phase-matching condition points to laser reflection with a center wavelength of 532 nm, the reflection peak can be precisely aligned with this wavelength by fine-tuning the thickness of the last layer. This adjustment method can significantly improve the device's reflection efficiency at the target wavelength, resulting in better filtering or mirror effects, while providing a reliable theoretical basis for subsequent device fabrication and performance verification.

[0033] S103. If the deviation between the simulated spectral output and the target spectral distribution data exceeds a preset threshold, then adjust the band modulation intensity of the independent subtask.

[0034] Obtain the point-by-point difference sequence between the current simulated spectral output and the target spectral distribution data in each band. Calculate the overall deviation based on the point-by-point difference sequence and compare it with a preset threshold. If the overall deviation exceeds the preset threshold, extract the positive deviation band set and the negative deviation band set from the point-by-point difference sequence. Determine the current band modulation intensity for the corresponding independent subtask for the positive deviation band set. Determine the current band modulation intensity for the corresponding independent subtask for the negative deviation band set. Based on the deviation direction of the positive and negative deviation band sets, increase or decrease the band modulation intensity for the corresponding independent subtask, respectively. Use a k-nearest neighbor regression model to obtain the adjusted band modulation intensity values ​​for each independent subtask based on historical modulation intensity and corresponding deviation records.

[0035] For example, in the iterative adjustment process of optimizing the transmission spectrum of a multilayer photonic crystal, the point-by-point difference sequence between the current simulated spectrum and the target spectrum at each wavelength in the range of 400 to 700 nanometers is first obtained. Assuming that the difference is positive 0.12 near 450 nanometers, negative 0.09 at 620 nanometers, and smaller differences in other bands, a complete deviation sequence is formed.

[0036] Specifically, by summing the absolute values ​​of all wavelength differences and dividing by the total number of wavelength points, the overall deviation is calculated to be 0.037. When this value exceeds a preset threshold of 0.025, it is determined that targeted modulation is required.

[0037] It should be noted that this overall deviation index can effectively reflect the overall consistency of the spectrum and avoid misjudgments caused by focusing only on local peaks and valleys.

[0038] In one possible implementation, a set of positive deviation bands, mainly concentrated in the blue region of 430 to 480 nm, and a set of negative deviation bands, mainly distributed in the red region of 580 to 650 nm, are separated from the point-by-point difference sequence.

[0039] Preferably, for independent subtasks corresponding to the positive deviation band set, the current band modulation intensity is determined to be moderately high, for example, initially set to 0.18; while for independent subtasks corresponding to the negative deviation band set, the current modulation intensity is set to moderately low, for example, 0.11.

[0040] In one embodiment, based on the deviation direction characteristics, the modulation intensity of the positive deviation subtask in the blue light region is increased by approximately 25% to 0.225 to lower the transmittance in that region; while the modulation intensity of the negative deviation subtask in the red light region is decreased by approximately 20% to 0.088 to appropriately increase the transmittance in that region. This type of directional adjustment can achieve reverse compensation for the deviation and is highly targeted.

[0041] Understandably, to make the adjustment range more reasonable, a k-nearest neighbor regression model is used to learn from historical records. Assume that the historical database contains several sets of records that have completed iterations, each set containing the past modulation intensity values ​​of a subtask and the resulting deviation. When the deviation for the current blue light subtask is 0.12, the model retrieves the five closest historical samples and finds that when the modulation intensity is between 0.20 and 0.24, the deviation is usually effectively converged to below 0.03. Therefore, the recommended adjusted intensity for this iteration is 0.222. Similarly, the intensity of the red light subtask, predicted by the model, is adjusted to 0.085.

[0042] For example, after the above intensity update, the local deviation of the new round of simulated spectra in the key bands is significantly reduced, and the overall deviation decreases from 0.037 to 0.019, which is close to the convergence requirement. This adjustment strategy based on deviation direction separation and combined with historical experience regression not only improves the optimization efficiency, but also significantly reduces the risk of getting trapped in local optima, demonstrating good practical value in the design of actual multilayer film systems.

[0043] S104. Determine whether the deviation after adjustment meets the requirements, and obtain the final set of control parameters.

[0044] The process involves acquiring raw deviation data, filtering it based on a preset threshold to obtain a deviation sequence that meets initial criteria, calculating missing values ​​using linear interpolation, resulting in a continuous deviation sequence, and removing outliers based on the difference between adjacent points to obtain a smoothed deviation sequence. This smoothed deviation sequence is then scaled proportionally to obtain the adjusted deviation value. The adjusted deviation value is then checked against a pre-defined target range. If it is within the target range, the current control parameter set is directly output; otherwise, reverse compensation calculation is performed based on the deviation direction to obtain a corrected control parameter set. A control command sequence is then generated based on the final control parameter set. In one embodiment, after obtaining the original deviation data, the original deviation data is first filtered by a preset threshold.

[0045] For example, band points with an absolute deviation greater than 0.08 are retained as a preliminary anomaly set, while points with an absolute deviation less than or equal to 0.08 are considered acceptable fluctuations, thus obtaining a preliminary deviation sequence that meets the criteria. This screening effectively eliminates the interference of minor noise on subsequent processing.

[0046] For example, a linear interpolation method is used to fill in individual missing positions in the initial deviation sequence. If a band has a missing deviation at 450nm, while the deviations at 440nm and 460nm are 0.12 and 0.15 respectively, then a continuous value of approximately 0.135 can be interpolated at the 450nm position. This method yields a continuous deviation sequence, ensuring that subsequent analysis is free of discontinuities that could affect the overall trend assessment.

[0047] It should be noted that, based on the continuous deviation sequence, outliers are further removed according to the magnitude of the difference between adjacent points.

[0048] Specifically, when the difference between two adjacent points exceeds the preset mutation threshold of 0.06, the latter point is determined to be an abnormal point and is removed.

[0049] For example, if a certain segment of the deviation in the sequence is 0.09, 0.11, 0.18, and 0.12, and it is found that the jump from 0.11 to 0.18 is too large, then the 0.18 point is removed and the points before and after are connected by a smooth trend, thus obtaining a smoother deviation sequence.

[0050] Preferably, the smoothed deviation sequence is subjected to overall scaling.

[0051] For example, when the average deviation of the sequence is about 18% too high, it can be uniformly multiplied by a scaling factor of 0.82 to bring the overall deviation closer to the target level. The scaled deviation value is closer to the actual controllable range.

[0052] Specifically, it determines whether the adjusted deviation value is within the pre-set target range. If the target range is set to -0.03 to 0.03, and most band deviations fall within this range after scaling, then the current control parameter set is considered to meet the requirements and can be directly output for the next spectrum generation. If some band deviations still exceed the range, for example, the blue light region is still 0.045 higher, then a reverse compensation calculation is performed based on the deviation direction, i.e., reducing the driving intensity of the corresponding channel by approximately 12%, thus obtaining the corrected control parameter set.

[0053] Understandably, after generating the control command sequence based on the final set of control parameters, each independent emitting unit will recombine and output the spectrum according to the new parameter values. This closed-loop adjustment method, based on layer-by-layer deviation processing and directional compensation, can significantly improve the matching accuracy between the simulated spectrum and the target spectrum, while reducing the number of blind iterations, resulting in a more efficient and stable spectral control effect.

[0054] S105. Extract real-time switching rules from the final control parameter set.

[0055] Obtain the final set of control parameters. Using a pre-configured extraction template, extract real-time switching rules containing conditional statements from the final set of control parameters. For each extracted real-time switching rule, parse its judgment expression to determine the state transition condition and corresponding threshold basis for each rule. If the current business condition for a real-time switching rule is met, activate the execution entity of that rule. Based on the execution entity of the activated rule, obtain the corresponding control target object. Determine the specific execution parameters for this switch by matching the control target object with the current system state. After obtaining the specific execution parameters for this switch, immediately send the execution parameters to the system to complete the real-time switch.

[0056] For example, in business scenarios involving the processing and real-time switching of control parameters, a detailed analysis can be conducted on the entire process from parameter separation to execution, focusing on the acquisition and application of the final control parameter set. The topic of separating real-time switching rules using pre-configured extraction templates can be understood as the process of extracting specific conditional logic from a complex set of parameter data. Suppose that in an industrial control system, the final control parameter set contains control data for multiple indicators such as temperature and pressure. The extraction template will filter out temperature-related switching rules based on preset logical conditions, such as switching to cooling mode when the temperature exceeds 50 degrees Celsius. This separation method ensures the specificity of the rules and avoids interference from irrelevant data.

[0057] For example, the implementation of parsing the conditional expressions of real-time switching rules can be illustrated through a specific scenario. In a device operation monitoring scenario, suppose the conditional expression of a rule is "When the running time exceeds 2 hours and the load rate is higher than 80%, switch to energy-saving mode." The parsing process will clearly define the two threshold criteria of duration and load rate and compare them with the data at the current business moment. This parsing method can accurately locate the triggering conditions for state transition, providing a reliable basis for subsequent rule activation.

[0058] For example, in the process of activating rules and identifying the target object for regulation, an application on an automated production line can be envisioned. Suppose the current business condition meets the rule condition of "equipment needs to reduce speed after 3 hours of continuous operation," the system will activate the execution subject of this rule, thereby locking the target object as a high-speed motor. By matching the target object with the system state, the speed reduction is determined to be 70% of the original speed, thus forming specific execution parameters. This method ensures the targeted and operable nature of the switching action. The topic of sending execution parameters to the system to complete real-time switching can be illustrated with an example of a temperature control system. Suppose the switching parameter is "adjust the cooling fan speed to 800 revolutions per minute," the system will immediately send this parameter to the fan control module to achieve rapid temperature adjustment. This real-time nature ensures the stability of system operation while effectively avoiding potential risks caused by delays.

[0059] In one possible implementation, the specific process of matching switching rules with system status can be further refined. For example, in an energy management scenario, the system status shows that the current power load is at its peak. The matching switching rules require the shutdown of some non-essential equipment, with the execution parameters specifically specifying the shutdown of power supply to a particular group of lighting devices. This matching mechanism can dynamically adapt to business needs and improve resource utilization efficiency. Through these multifaceted examples, it can be seen that the entire process design, from parameter separation to rule activation and parameter distribution, can achieve precise control and rapid response in industrial control or equipment management. This implementation not only improves the system's automation level but also effectively reduces operational risks and ensures business continuity at critical moments.

[0060] S106. Determine whether the change in light quality triggers the real-time switching rule.

[0061] Step 1: Continuously collect light quality data from the environment using light quality monitoring equipment, recording changes in light quality in real time to obtain dynamic information on these changes. Step 2: Based on the dynamic information of light quality changes, compare it with a pre-established rule condition library to determine whether the light quality change reaches a preset threshold in the rule conditions. Step 3: If the light quality change reaches the preset threshold, trigger the change detection module to obtain detailed data on the change amplitude and frequency, determining whether a switching process is needed. Step 4: Using the change amplitude and frequency data, combined with the real-time judgment mechanism, analyze the rationality of the switching timing to obtain a preliminary conclusion on the switching trigger. Step 5: Based on the preliminary conclusion on the switching trigger, call the rule matching module to perform a deep comparison between the specific situation of the light quality change and the real-time switching rules to determine whether the conditions are met. Step 6: If the conditions of the real-time switching rules are met, generate a switching command, transmit it to the execution unit through the system interface, complete the switching operation, and determine the final execution status. Step 7: Based on the execution status of the switching operation, record the light quality change and switching trigger results, update the rule condition library, and provide a basis for subsequent monitoring frequency adjustments.

[0062] For example, in the business field of light quality monitoring, the implementation of the process for collecting and switching ambient light quality data can be analyzed and illustrated in detail from multiple perspectives to ensure that the logic is rigorous and has practical application value.

[0063] For example, continuous acquisition of light quality data can be achieved by deploying highly sensitive light quality sensors to record changes in light intensity and spectral distribution in the environment in real time. Suppose that in a greenhouse environment, the sensor records light quality data every 5 minutes, detecting a drop in light intensity from 1000 lux to 500 lux. This dynamic information provides a basis for subsequent judgment. This acquisition method can promptly capture light quality fluctuations, providing real-time data for system decision-making.

[0064] For example, in the comparison phase of the rule condition library, the preset threshold might be set as a change in light intensity exceeding 30% or a spectral shift exceeding a specific range. If a 40% decrease in light intensity is detected, the system will immediately compare this data with the threshold in the condition library to determine whether further processing is needed. This comparison mechanism ensures that only significant changes trigger subsequent processes, avoiding invalid operations.

[0065] For example, regarding the triggering of the change detection module, assuming the light intensity decreases three times per hour and the magnitude consistently exceeds 30%, the system will extract this detailed data and analyze whether a switching process needs to be initiated. This refined analysis helps to accurately identify the patterns of environmental changes and improve the rationality of decision-making.

[0066] For example, in the rationality analysis of switchover timing, the system may combine the frequency and magnitude of changes to determine whether the current time is the optimal time to switchover. If the frequency of changes is too high, it may mean that the environment is unstable, and the system will postpone the switchover to avoid wasting resources due to frequent operations. This mechanism ensures that the timing of switchover operations is more scientific.

[0067] For example, in the depth comparison of the rule matching module, if the change in light quality meets a certain rule, such as the light intensity being below 600 lux for more than 10 minutes, the system will confirm that the switching condition has been met. This depth comparison can filter out random fluctuations and ensure the necessity of switching.

[0068] For example, when generating a switching command and transmitting it to the execution unit, the system may generate a command to adjust the light source intensity, which is then sent to the lighting equipment inside the greenhouse via an interface, completing the switch from low-light mode to high-light mode. This operation ensures rapid adaptation to ambient light quality.

[0069] For example, when recording switching results and updating the rule condition library, the system stores the execution data for the current light intensity change from 500 lux to 800 lux, and adjusts the monitoring interval according to the frequency of change, such as shortening it from 5 minutes to 3 minutes. This update mechanism provides a more accurate reference for subsequent monitoring and optimizes the system's response efficiency. Through the above examples and analyses from multiple perspectives, it can be seen that the design of each link is closely centered on the core requirements of light quality monitoring and switching, progressing step by step to ensure the integrity and practicality of the process, while also providing strong support for environmental adaptability.

[0070] S107. If the change in light quality triggers the real-time switching rule, then obtain the dynamic control instruction.

[0071] Light quality change information is collected by sensor data, and the change data is preliminarily processed to obtain a judgment result of the light quality change. If the judgment result of the light quality change meets the preset switching rules, the real-time switching process is triggered, and the execution conditions for real-time switching are determined. Based on the execution conditions of real-time switching and combined with environmental adaptation data, the corresponding change response strategy is obtained, and the appropriate response mode is determined. If the change response strategy is consistent with the preset rule judgment, a dynamic control scheme is generated, and the content of the control command is determined. Through the command content of the dynamic control scheme and combined with the command generation mechanism, the final control command data is obtained, and the command is formatted. Based on the formatted control command data, it is transmitted to the target device, completing the execution process of real-time switching and dynamic control.

[0072] By collecting environmental spectral distribution and color temperature data in real time through sensors, a continuous light quality change curve is formed.

[0073] In one possible implementation, as natural light gradually transitions from warm white light in the morning to cool white light at noon, the system records the process of the color temperature rising from 2800K to 5500K and calculates the rate of change per unit time.

[0074] Specifically, if the light quality change assessment shows that the color temperature rises by more than 1800K within 30 minutes, and the proportion of blue light increases by more than 25%, then the triggering conditions are initially met. At this point, the system enters the real-time switching process.

[0075] It should be noted that the switch does not happen immediately, but rather is further verified based on the current indoor activity status and usage scenario.

[0076] For example, in an office setting, if the system detects that most people are engaged in detailed reading or computer work, it will prioritize accessing environmental adaptation data, specifically historical preference records showing that the user group has the highest acceptance of color temperatures between 4000K and 4500K. In this case, the change response strategy tends to gradually adjust to neutral white light rather than directly switching to daylight color temperature.

[0077] Preferably, a dynamic control scheme is generated when the judgment result is consistent with the preset rules.

[0078] In one embodiment, the scheme includes a three-stage color temperature transition path: first, the color temperature is linearly adjusted from the current color temperature to 4200K within 5 minutes, then stabilized for 10 minutes, and finally slowly approached 4600K based on real-time feedback. This step-by-step adjustment can effectively avoid discomfort caused by sudden visual changes.

[0079] In one possible implementation, the final control command explicitly specifies the grouped brightness ratio, target color temperature value, and transition time parameters for each luminaire, and encapsulates these parameters in a standard protocol format before sending them to the lighting controller. Upon receiving the command, the controller immediately executes the corresponding dimming action and simultaneously provides feedback on the execution status.

[0080] For example, if a sudden change in light quality is large, such as a sudden drop in color temperature of 1200K caused by cloud cover outdoors, the system will still prioritize determining whether the continuous monitoring threshold has been exceeded. Only if three consecutive samples exceed the limit will an adjustment command be generated, thereby avoiding frequent switching caused by brief interference.

[0081] Understandably, by combining the above-mentioned graded judgment with gradual control, we can ensure a high degree of responsiveness to changes in the light environment and natural light, while significantly reducing visual fatigue and discomfort among personnel, and improving comfort and work efficiency in long-term work scenarios.

[0082] S108. Drive the multilayer photonic crystal structure to perform the parallel spectroscopy task through the dynamic control command.

[0083] Based on the business content and extracted relevant attributes, the following business solution is generated, focusing on the goal of using dynamic control commands to drive a multilayer photonic crystal structure to complete parallel spectral tasks. The technical process is designed by combining attributes such as dynamic control, command-driven, multilayer structure, photonic crystal, parallel task, spectral analysis, control mechanism, task execution, structural design, photonic properties, command parsing, and task allocation. The following steps are logically linked, with the output of each step serving as the input for the next: The received dynamic control commands are decomposed and processed through a pre-established command parsing module to obtain specific control parameters and task objectives, determining the execution requirements for the multilayer photonic crystal structure. Based on the obtained control parameters, the interlayer configuration scheme of the multilayer photonic crystal structure is matched using a structural design database to obtain a structural adjustment scheme suitable for the parallel task. The structural adjustment scheme is transformed into specific driving signals through the control mechanism module, and the photonic properties of the multilayer structure are adjusted in real time to obtain the adjusted photonic crystal response state. If the adjusted response state meets the preset spectral analysis conditions, the parallel task is decomposed into multiple sub-task units through the task allocation module, determining the photonic crystal region corresponding to each unit. For each decomposed sub-task unit, an instruction-driven module sends execution signals to the corresponding region to acquire the real-time data stream of each region during the spectral analysis process. The data integration module processes the real-time data stream in parallel and, combined with the spectral analysis algorithm, obtains the final spectral task execution result. Based on the final execution result, corresponding task completion feedback information is generated and transmitted to the dynamic control system through the feedback channel, completing the entire parallel spectral task process.

[0084] For example, in applications where dynamic control commands drive multilayer photonic crystal structures to complete parallel spectral tasks, the implementation of the command parsing module can be understood from a theoretical perspective as the process of breaking down complex commands into executable units. Assuming the received command contains multiple control objectives, such as adjusting the photon response in a specific wavelength band and allocating task priorities, the parsing module will decompose it into control parameters within a wavelength range of 400-700 nm and a task execution order. This decomposition ensures that subsequent modules can accurately identify task requirements.

[0085] For example, when matching interlayer configuration schemes to a structural design database, a multilayer structure design corresponding to the control parameters can be found through a preset database. Assuming that the parameters require high transmittance in the 550nm band, the database will select crystal structure schemes with a layer thickness ratio of 1:2, and optimize the interlayer refractive index distribution in conjunction with the requirements of parallel tasks to ensure that the photonic properties meet the requirements of multi-task parallel processing.

[0086] For example, in the implementation of the control mechanism module, the structural adjustment scheme is transformed into electrical or thermal signals driving changes between crystal layers. Assuming the adjustment target is to change the photonic bandgap of a certain layer, the driving signal is applied to the corresponding region in the form of a 5V voltage, and the response status is monitored in real time to ensure the bandgap shifts to the target range. This real-time adjustment improves the accuracy of spectral analysis. For the task allocation module, assuming the parallel task involves three spectral analysis targets, the module will split the task into three sub-units, corresponding to three regions of the crystal structure, with each region responsible for data acquisition in one band. This allocation method ensures the high efficiency of task execution. In the implementation of the instruction-driven module, after sending an execution signal to the corresponding region, assuming a region is responsible for 600nm band analysis, the signal will trigger the photonic response in that region, acquiring the reflected light intensity data stream in real time. This precise driving ensures the reliability of data acquisition. For the data integration module, parallel processing algorithms can be used to integrate the data streams from each region. Assuming that the three regions acquire light intensity data in different bands, the integrated data forms a complete spectral distribution map, which is then combined with analysis algorithms to extract characteristic peaks. This integration method significantly improves the comprehensiveness of the task results. Finally, in generating the task completion feedback information, assuming the spectral analysis results show that the characteristic bands meet expectations, the feedback information will include the task completion rate and key parameters, and will be transmitted to the control system through the feedback channel for subsequent optimization. This feedback mechanism helps the system continuously improve its control strategy.

[0087] S109. Obtain the light quality feedback data after executing the parallel spectral task.

[0088] The completion identifier of the parallel spectral task is read from the execution end through a preset interface. Based on the completion identifier, all optical quality feedback data is retrieved from the corresponding buffer. The retrieved optical quality feedback data is then processed by multi-channel synchronous splitting to obtain the original spectral sequence of each channel. Fourier transform is used to extract frequency domain features from the original spectral sequences of each channel to obtain the spectral distribution result. Predefined key band intensity values ​​are separated from the spectral distribution result to form the current optical quality feature vector. The current optical quality feature vector is compared element-wise with the reference optical quality template associated with the task sequence. If the deviation of all key band intensity values ​​is less than a preset threshold, the optical quality feedback is deemed qualified; otherwise, it is deemed unqualified. Based on the judgment result, an optical quality feedback record containing the task number, channel number, and qualified status is generated.

[0089] By reading the completion flag of the parallel spectral task in real time from the execution end through a preset interface, it is possible to accurately determine whether the current task has entered the feedback acquisition stage.

[0090] In one embodiment, when the completion flag changes from 0 to 1, the subsequent data reading process is immediately triggered, avoiding the delay caused by polling and waiting.

[0091] In one possible implementation, all optical quality feedback data for this task is fully extracted from a dedicated buffer based on the completion identifier. This data typically includes raw optical intensity sequences acquired simultaneously across multiple channels. For example, for a four-channel system, four sets of time-series optical intensity records, each with 2048 points, can be obtained simultaneously, with each set corresponding to the response output of a different photonic crystal region. For the acquired optical quality feedback data, a multi-channel synchronous splitting processing technique is used to separate the composite data stream according to channel number, obtaining independent raw spectral sequences for each channel. For example, the channel 1 sequence reflects the response of the short-wavelength dominant region, while the channel 3 sequence mainly carries information from the mid-to-long-wavelength regions. All sequences are of consistent length and time-aligned, ensuring a unified basis for subsequent analysis.

[0092] Preferably, Fourier transform is used to extract frequency domain features from the original spectral sequences of each channel, thereby obtaining clear spectral distribution results.

[0093] Specifically, after transformation, a significant peak intensity of 0.87 was observed in Channel 1 near 450 nm, while Channel 2 reached an intensity of 0.92 at 620 nm. These peak positions and amplitudes directly reflect the current bandgap and transmission characteristics of the multilayer photonic crystal. Predefined key wavelength band intensity values ​​were separated from the spectral distribution results. For example, the 430–470 nm, 510–550 nm, and 600–640 nm bands were selected as the core monitoring targets, and their peak intensities were extracted to form the current optical quality feature vector. This vector has nine dimensions, with each set of three values ​​corresponding to the key performance characteristics of one channel.

[0094] Specifically, the current light quality feature vector is compared element by element with the baseline light quality template pre-stored in the task sequence.

[0095] In one embodiment, if the reference template requires an intensity of 0.85±0.04 at 430 to 470 nanometers, and the actual measured intensity is 0.87, the deviation is only 0.02, which is within the acceptable range; if the measured intensity at 620 nanometers in a certain channel is only 0.71, the deviation exceeds 0.20, which exceeds the preset threshold of 0.05.

[0096] It should be noted that when the intensity deviation of all nine key bands is less than the preset threshold of 0.05, it is determined that the current light quality feedback is qualified; otherwise, it is determined as unqualified. The qualified determination can directly support the confirmation of the task closed-loop, while the unqualified result triggers an abnormal alarm and a backtracking adjustment of the structural parameters. A light quality feedback record containing the task number, channel number, and qualified status is generated according to the determination result. For example, the record format can be "Task T20251224003, Channel 2, Unqualified, Deviation Exceeded Band 620nm", which is convenient for subsequent traceability and optimization of the regulation strategy. Through the above process, not only the precise quantitative evaluation of the light quality feedback is achieved, but also reliable data support is provided for the closed-loop iteration of the dynamic regulation instruction, effectively improving the stability and repeatability of the multi-layer photonic crystal parallel spectroscopy task.

[0097] The above content is only an example and illustration of the structure of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the structure of the invention or exceed the scope defined by this claim book, they shall fall within the protection scope of the present invention.

Claims

1. A dynamic control system and method for greenhouse light quality based on photonic crystal chips, characterized in that, The method includes: Obtain a parallel spectral task sequence for the integrated layout; Simulate the spectral output process of a multilayer photonic crystal structure according to the timing control protocol; If the deviation between the simulated spectral output and the target spectral distribution data exceeds a preset threshold, the band modulation intensity of the independent subtask is adjusted. Determine whether the adjusted deviation meets the requirements to obtain the final set of control parameters; Extract real-time switching rules from the final set of control parameters; Determine whether the change in light quality triggers the real-time switching rule; If the change in light quality triggers the real-time switching rule, then a dynamic control command is obtained; The multilayer photonic crystal structure is driven to perform the parallel spectral task by the dynamic control commands; Obtain the optical quality feedback data after executing the parallel spectral task.

2. The greenhouse light quality dynamic control system and method based on photonic crystal chips according to claim 1, characterized in that, The step of obtaining the parallel spectral task sequence for the integrated layout includes: Based on the business content and the extracted relevant attributes, the following business solution is generated, which revolves around the generation of integrated layout and parallel spectral task sequence, and forms a tight logical chain between the steps: For integrated layout, the initial layout scheme is obtained from the pre-established layout template library; By initially screening the layout templates, a layout structure that meets the requirements of parallel spectral processing was determined. Based on the selected layout structure, obtain the corresponding set of spectral analysis tasks; The data acquisition module is used to classify and organize the task set, resulting in classified task groups; For the categorized task groups, resource scheduling is performed for each group through a parallel processing mechanism; If the workload of a certain group exceeds a preset threshold, the tasks are split into sub-task units. Based on the decomposed task units, a preliminary task sequence is generated; A sequence generation tool is used to sort the task units by time to determine the execution order of the sorted tasks. Based on the sorted task execution order, the dependency relationships between tasks are analyzed through the information integration module; If dependency conflicts are found between tasks, the execution order is adjusted to obtain an optimized task sequence; Based on the optimized task sequence and in conjunction with the layout optimization strategy, the integrated layout is dynamically adjusted. The layout adjustment tool is used to balance resource allocation and determine the final layout configuration; For the final layout configuration, generate a sequence of parallel spectral tasks; The task allocation module binds the sequence with the layout configuration to obtain an executable task sequence scheme.

3. The greenhouse light quality dynamic control system and method based on photonic crystal chips according to claim 1, characterized in that, The process of simulating the spectral output of a multilayer photonic crystal structure according to the timing control protocol includes: Initial incident light information is obtained by pre-establishing a wavelength sequence; The transmission matrix method is used to calculate the transmittance and reflectance data of the first layer material parameters. The first layer transmittance and reflectance data are used as input, and the second layer transmittance and reflectance data are calculated by combining the second layer material parameters. Based on the current cumulative transmittance data and the current cumulative reflectance data, the material parameters of the next layer are input sequentially under time control to obtain the transmittance data and reflectance data of the next layer. If the current layer sequence has not reached the last layer, the cumulative transmittance and cumulative reflectance data will continue to be updated using the transfer matrix method. When the layer sequence reaches the last layer, the transmission spectrum curve of the complete multilayer photonic crystal is determined by the final cumulative transmittance data. Based on the transmission spectrum curve and phase matching results of the complete multilayer photonic crystal, the final spectral simulation results are output.

4. The greenhouse light quality dynamic control system and method based on a photonic crystal chip according to claim 1, characterized in that, If the deviation between the simulated spectral output and the target spectral distribution data exceeds a preset threshold, the band modulation intensity of the independent subtask is adjusted, including: Obtain the point-by-point difference sequence between the current simulated spectral output and the target spectral distribution data in each band; Calculate the overall deviation based on the point-by-point difference sequence and compare it with a preset threshold; If the overall deviation exceeds the preset threshold, then extract the positive deviation band set and the negative deviation band set from the point-by-point difference sequence. For a set of positive deviation bands, determine the current band modulation intensity of the corresponding independent sub-task; For the set of negative bias bands, determine the current band modulation intensity of the corresponding independent sub-task; Based on the deviation direction between the positive deviation band set and the negative deviation band set, the band modulation intensity of the corresponding independent sub-task is either increased or decreased. The k-nearest neighbor regression model is used to obtain the band modulation intensity values ​​after adjustment for each independent subtask based on the records of historical modulation intensity and corresponding deviation.

5. The greenhouse light quality dynamic control system and method based on a photonic crystal chip according to claim 1, characterized in that, The determination of whether the adjusted deviation meets the requirements yields the final set of control parameters, including: The process involves acquiring raw deviation data, filtering it within a range using a preset threshold to obtain a deviation sequence that meets the initial conditions, calculating missing values ​​at intermediate positions using linear interpolation, resulting in a continuous deviation sequence, removing outliers based on the difference between adjacent points, and obtaining a smoothed deviation sequence. This smoothed deviation sequence is then scaled proportionally to obtain the adjusted deviation value. If the adjusted deviation value falls within a pre-defined target range, the current control parameter set is directly output; otherwise, reverse compensation calculation is performed based on the deviation direction to obtain a corrected control parameter set. A control command sequence is then generated based on the final control parameter set.

6. The greenhouse light quality dynamic control system and method based on photonic crystal chips according to claim 1, characterized in that, The step of extracting real-time switching rules from the final control parameter set includes: Obtain the final set of control parameters; By using a pre-configured extraction template, real-time switching rules containing conditional statements are extracted from the final set of control parameters; For the real-time switching rules obtained from the separation, their judgment expressions are analyzed one by one to determine the state transition conditions and corresponding threshold basis for each rule; If the current business condition meets the state transition condition of a real-time switching rule, then the execution entity of that rule is activated; Based on the executing entity of the activated rule, obtain the corresponding control target object; By adjusting the matching relationship between the target object and the current system state, the specific execution parameters for this switch can be determined; Once the specific execution parameters for this switch are obtained, the system is immediately notified to send these parameters to complete the real-time switch.

7. The greenhouse light quality dynamic control system and method based on photonic crystal chips according to claim 1, characterized in that, Determining whether the change in light quality triggers the real-time switching rule includes: Step 1: Continuously collect light quality data in the environment using light quality monitoring equipment, record changes in light quality in real time, and obtain dynamic information on changes in light quality; Step 2: Based on the dynamic information of light quality changes, compare the results using a pre-established rule condition library to determine whether the light quality changes have reached the preset threshold in the rule conditions. Step 3: If the change in light quality reaches the preset threshold, the change detection module is triggered to obtain detailed data on the change amplitude and frequency to determine whether it is necessary to enter the switching process. Step 4: By analyzing the data on the magnitude and frequency of change, combined with the real-time judgment mechanism, analyze the rationality of the switching timing and obtain a preliminary conclusion on the switching trigger. Step 5: Based on the preliminary conclusions of the switching trigger, call the rule matching module to perform a deep comparison between the specific details of the light quality change and the real-time switching rules to determine whether the conditions are met. Step Six: If the conditions of the real-time switching rules are met, a switching instruction is generated, transmitted to the execution unit through the system interface, the switching operation is completed, and the final execution status is determined; Step 7: Based on the execution status of the switching operation, record the light quality change and the result data of the switching trigger, update the rule condition library, and provide a basis for subsequent monitoring frequency adjustment.

8. The greenhouse light quality dynamic control system and method based on photonic crystal chips according to claim 1, characterized in that, If the change in light quality triggers the real-time switching rule, a dynamic control instruction is obtained, including: Information on changes in light quality is collected by sensor data, and the changes are preliminarily processed to obtain a determination result of the changes in light quality. If the determination result of the change in light quality meets the preset switching rules, the real-time switching process is triggered to determine the execution conditions for real-time switching. Based on the execution conditions of real-time switching and combined with environmental adaptation data, obtain the corresponding change response strategy and determine the appropriate response mode. If the change response strategy is consistent with the preset rule judgment, a dynamic control plan is generated to determine the content of the control instruction; By combining the instructions of the dynamic control scheme with the instruction generation mechanism, the final control instruction data is obtained, and the instruction formatting process is completed. Based on the formatted control command data, it is transmitted to the target device to complete the execution process of real-time switching and dynamic control.

9. A greenhouse light quality dynamic control system and method based on a photonic crystal chip according to claim 1, characterized in that, The process of driving the multilayer photonic crystal structure to perform the parallel spectroscopy task through the dynamic control command includes: Based on the business content and the extracted relevant attributes, the following business solution is generated, which focuses on the goal of using dynamic control instructions to drive a multilayer photonic crystal structure to complete parallel spectral tasks. The technical process is designed by combining attributes such as dynamic control, instruction drive, multilayer structure, photonic crystal, parallel task, spectral analysis, control mechanism, task execution, structural design, photonic properties, instruction parsing and task allocation. The following steps form a tight logical connection, and the output of each step serves as the input of the next step: through the pre-established instruction parsing module, the received dynamic control instructions are decomposed and processed to obtain specific control parameters and task objectives, and to determine the execution requirements of the instructions for the multilayer photonic crystal structure. Based on the obtained control parameters, the interlayer configuration scheme of the multilayer photonic crystal structure is matched with the structural design database to obtain a structural adjustment scheme suitable for parallel tasks; The structural adjustment scheme is transformed into a specific driving signal through the control mechanism module, and the photonic properties of the multilayer structure are adjusted in real time to obtain the adjusted photonic crystal response state. If the adjusted response state meets the preset spectral analysis conditions, the parallel task is decomposed into multiple sub-task units through the task allocation module, and the photonic crystal region corresponding to each unit is determined. For the decomposed sub-task units, an instruction-driven module is used to send execution signals to the corresponding regions to obtain the real-time data stream of each region during the spectral analysis process; The real-time data stream is processed in parallel by the data integration module, and combined with the spectral analysis algorithm, the final spectral task execution result is obtained. Based on the final execution result, corresponding task completion feedback information is generated and transmitted to the dynamic control system through the feedback channel to complete the entire parallel spectroscopy task process.

10. A greenhouse light quality dynamic control system and method based on a photonic crystal chip according to claim 1, characterized in that, The acquisition of optical quality feedback data after executing the parallel spectral task includes: The completion flag of the parallel spectral task is read from the execution end through a preset interface; Retrieve all optical quality feedback data from the corresponding cache area based on the completion indicator; The acquired optical quality feedback data is simultaneously split into multiple channels to obtain the original spectral sequences of each channel. The frequency domain features of the original spectral sequences of each channel were extracted using Fourier transform to obtain the spectral distribution results; The intensity values ​​of predefined key bands are separated from the spectral distribution results to form the current optical quality feature vector; The current optical quality feature vector is compared element by element with the baseline optical quality template associated with the task sequence; If the deviation of the intensity values ​​of all key bands is less than the preset threshold, the optical quality feedback is deemed qualified; otherwise, it is deemed unqualified. Based on the judgment result, an optical quality feedback record containing the task number, channel number, and qualified status is generated.

Images