Plant factory light intensity integrated control system

Abstract

The application discloses a plant factory light intensity integrated control system, which comprises an environment sensing module, a central control unit and a partition light adjusting driving module. The core of the application is that a Bayesian optimization module and a digital PID controller are integrated in the central control unit. The Bayesian optimization module takes PID parameters as optimization variables, and a comprehensive performance index as a target function, and automatically optimizes through a Bayesian optimization algorithm. The target function comprises a function for improving response speed and accuracy, a function for suppressing light overshoot, and a function for protecting equipment and reducing energy consumption. The application realizes adaptive setting of PID parameters, solves the problem of insufficient control precision of a traditional method, and improves the control quality of a plant factory light environment.

Description

Technical Field

[0001] This application relates to the field of agricultural environmental control technology, and in particular to an integrated control system for light intensity in plant factories. Background Technology

[0002] Plant factories are highly efficient agricultural systems that achieve continuous year-round crop production by precisely controlling indoor environmental conditions, without the influence of outdoor environments. Among these, light is a key environmental factor for the normal operation of a plant factory.

[0003] Currently, the use of PID controllers for light control in plant factories has the following drawbacks: tuning is time-consuming and labor-intensive, and it is difficult to find globally optimal parameters; when crop varieties are changed, growth stages progress, or the environment changes dynamically, fixed PID parameters cannot always maintain optimal control performance, resulting in insufficient control precision and poor adaptability. To address these issues, we propose an integrated light intensity control system for plant factories that can automatically, efficiently, and accurately tune PID controller parameters, thereby achieving higher control precision, better stability, and stronger adaptability. Summary of the Invention

[0004] The main purpose of this application is to provide an integrated control system for light intensity in plant factories, which aims to solve the technical problems of insufficient control precision, poor adaptability, cumbersome tuning process, and potential impact on crop growth in existing light control systems for plant factories.

[0005] This application provides an integrated control system for light intensity in a plant factory based on a Bayesian-optimized PID parameter self-tuning algorithm, employing the following technology: An integrated control system for light intensity in a plant factory, comprising an environmental sensing module, a central control unit, and a zoned dimming drive module, characterized in that: The environmental sensing module independently deploys sensors in each zone to directly measure the actual light intensity received by the plants and send the data to the central control unit. The central control unit has pre-stored target light intensity settings corresponding to different plant varieties and growth stages. It receives plant light intensity measurements from the environmental sensing module and compares the measurements reported by the environmental sensing module with the target light intensity settings using a built-in digital PID controller and Bayesian optimization module to calculate control commands. Send instructions to the local dimming driver module; The local dimming drive module receives and decodes instructions from the central control unit. The dimming device is driven to change according to the instructions.

[0006] Optionally, the central control unit is equipped with a digital PID controller and a Bayesian optimization module: the Bayesian optimization module uses the parameter combination of the digital PID controller. To optimize variables, a preset comprehensive performance index was used. Given the objective function, the Bayesian optimization algorithm is used to automatically find a function that satisfies the objective function. Optimal PID parameter combination The optimized parameter set is configured to the digital PID controller; the built-in digital PID controller compares the measured values ​​reported by the environmental sensing module with the target setpoint to calculate the control command. It sends commands to the zone dimming driver module.

[0007] Optionally, the central control unit has a built-in digital PID controller whose control algorithm is an incremental PID algorithm, and outputs the control command. .

[0008] Optionally, the objective function for: ,in, To assess the parameters of the closed-loop control system A comprehensive and quantitative evaluation of dynamic response quality and steady-state performance; The integral of time multiplied by the absolute error is used to measure the system's response speed and steady-state accuracy. , This is due to the error in light intensity. This is the overshoot, used to suppress light intensity overshoot. This represents the maximum measured light intensity during the system response process. It is the system target illumination intensity value; This is a penalty term for drastic changes in control output, used to protect the actuator and save energy. ; Optional They are respectively , , The weighting coefficients are determined; an initial set of weights is set, and Bayesian optimization is run to obtain a set of benchmark PID parameters and corresponding control performance. The response curve of the benchmark control is observed, analyzed, and decisions are made. The corresponding weights are adjusted, and the entire Bayesian optimization process is rerun using the new weights to find new PID parameters. The new control response curve is observed to verify whether the requirements are met and whether the unadjusted weights are within an acceptable range, until a combination of PID parameters that meets all requirements is found. weighting coefficient combination .

[0009] Optionally, the Bayesian optimization module receives the objective function with weights specified. In the predefined parameter space Internally, a set of initial parameter configurations is selected through a random sampling strategy. For each set of parameters The objective function value is obtained by running it on the control system and evaluating its performance. This constitutes the initial dataset. Based on this dataset, a Gaussian process is constructed as a surrogate model to predict... ; Selected parameters A digital PID controller is configured to run the control system in a closed-loop environment for a predetermined period of time, during which the light intensity is collected, and the comprehensive performance index is calculated based on this data. Numerical values, pairing new data Add to dataset To expand the observation sample; After the iteration process terminates, from the final dataset Selecting from the options makes the objective function The combination of parameters that achieves the minimum value As an optimization result, it is locked into the final parameters of the PID controller.

[0010] Optionally, the digital PID controller is based on Formulas, real-time calculation of control commands ; For proportional calculations, ; in For error, ; For integration operations, ; in From the first time to the... Sum of all errors, For the first Error The sampling period; For differentiation operations, ; in for Error in time; Calculations were performed using parameter combinations Digital PID controller Control commands Send to the corresponding partition dimming driver module.

[0011] Optionally, the Bayesian optimization module is configured to automatically trigger the optimization process when the system starts up or the growth phase switches.

[0012] Optionally, the system is configured with an independent digital PID controller and Bayesian optimization process for each planting zone.

[0013] Optionally, the zone dimming drive module includes a dimming device, which is a plurality of LED light groups, each independent LED light group corresponding to a planting zone.

[0014] This application proposes an integrated control system for light intensity in plant factories. This system uses an environmental sensing module to acquire environmental information about the plant factory, a central control unit to compare this information with plant growth information and issue commands, and a zoned dimming drive module to connect the environmental sensing module and the zoned LED supplementary lighting module. This enables precise conversion from digital signals to power control. The zoned LED supplementary lighting module adjusts the plant environment according to the commands, solving the problems of insufficient control precision and intelligence in traditional methods. It achieves precise light control while simultaneously providing comprehensive benefits of energy saving and extended lifespan. Attached Figure Description

[0015] Figure 1 A schematic diagram of a method for an integrated control system for light intensity in a plant factory, provided in an embodiment of this application; Figure 2 for Figure 1 A schematic diagram of the working process of the zoned dimming drive module in the integrated control system for light intensity in a plant factory. The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] The main solution in this application's embodiments is: An integrated control system for light intensity in a plant factory includes an environmental sensing module, a central control unit, and a zoned light-adjusting drive module. The central control unit is equipped with a digital PID controller and a Bayesian optimization module. The Bayesian optimization module uses a combination of parameters from the digital PID controller. To optimize variables, a preset comprehensive performance index was used. Given the objective function, the Bayesian optimization algorithm is used to automatically find a function that satisfies the objective function. Optimal PID parameter combination The Bayesian optimization module is connected to the digital PID controller and is used to configure the optimal parameter set obtained from the optimization to the digital PID controller.

[0017] The Bayesian optimization module optimizes the multi-objective synthesis function. Able to take into account both , and This results in a control effect that far surpasses traditional methods in terms of overall performance; the system can bring the illumination to the set value faster without overshoot, while the LED driver operates more smoothly, effectively extending the equipment life and reducing energy consumption.

[0018] The environmental sensing module treats each independent planting bed, planting rack, or single-layer vertical cultivation rack as a partition, and deploys sensors independently in each partition to ensure precise and targeted control; the light sensors deployed in the plant canopy directly measure the actual light intensity received by the plants and send the data to the central control unit.

[0019] The central control unit receives raw data from the environmental sensing module and transmits the processed data to the digital PID controller. PID control comprises three parts: proportional, integral, and derivative. Proportional control provides a fast response to errors; integral control eliminates steady-state errors; and derivative control improves the system's dynamic performance and stability. When the PID controller is running on the system, the system calculates the objective function at that moment. When the system starts up for the first time and the growth phase progresses, the Bayesian optimization process is triggered to find the objective function. Minimum parameter The light intensity is configured to be stable and quickly stabilized at the target value by a digital PID controller for closed-loop control.

[0020] The local dimming drive module receives instructions from the central control unit. The dimming device is driven to change according to instructions; the dimming device includes multiple LED light groups, each independent LED light group corresponding to a planting area, and the LED light groups change according to instructions. When the drive signal is high, current forms a path and flows through the LED group, making it light up; when the drive signal is low, the current path is cut off and the LED group is turned off; through zoned deployment, light energy is accurately delivered to the required area.

[0021] Currently, when plant factories use PID controllers, the PID parameters are fixed values, which makes it difficult to cope with the complex nonlinear and time-varying environment of plant factories. Some self-tuning methods may cause the system to go out of control during the tuning process, which may interfere with plant growth. When using other optimization algorithms (such as genetic algorithms and particle swarm optimization algorithms) to tune PID, a large number of fitness functions need to be evaluated, the convergence speed is slow, and it may be sensitive to the parameters of the algorithm itself.

[0022] This application provides a solution that enables the integrated light intensity control system for plant factories to achieve better dynamic performance and steady-state accuracy than traditional tuning methods; it possesses strong adaptability; it extends equipment life and reduces energy consumption; and it enables plant factories to provide a highly controllable, stable, and optimal light environment for crop growth.

[0023] Reference Figure 1 This application provides an embodiment of an integrated control system method for light intensity in a plant factory. The plant factory is divided into three independent planting zones: A, B, and C. Zone A is for the seedling stage, zone B is for the growth stage, and zone C is for the harvesting stage. Taking zone B as an example, the control process from system startup to stable operation is described in detail: After the system starts, the user selects the crop and growth stage via the touchscreen display. In this embodiment, area B is selected for description. The crop planted in area B is lettuce, and the growth stage is the growing season. The sensor above the plant canopy in area B in the environmental sensing module collects the light intensity of the plant canopy at a sampling period of 80ms, which is 280. The current signal output by the sensor is sent to the signal conditioning circuit. The signal conditioning circuit includes an instrumentation amplifier and a filter. The current signal is amplified by the AD620 instrumentation amplifier, and then filtered by a second-order active low-pass filter composed of TL084 operational amplifiers to remove high-frequency and power frequency noise. The signal conditioning circuit converts the original signal into a standard analog voltage signal and effectively filters out 50Hz power frequency interference and other high-frequency noise from the environment. The processed analog voltage signal is sent to the ADC pin of the microcontroller corresponding to the zone for analog-to-digital conversion. The converted light intensity data is then uploaded to the central control unit via an RS-485 communication bus.

[0024] The central control unit includes a microcontroller, a digital PID controller, a Bayesian optimization module, and a memory. The central control unit loads a preset crop light intensity database from the memory: the target light intensity for area A is 150. The target illumination intensity in area B is 300. The target illumination intensity in area C is 250. The central control unit retrieves the corresponding target light intensity setpoint from the database, receives plant light measurement values ​​from the environmental sensing module, compares the received real-time light measurement values ​​for each zone with the stored target setpoint, and calculates the error. .

[0025] Error in this embodiment .

[0026] Automatic Bayesian optimization is triggered during system startup: The central control unit uses an STM32F407ZGT6 microcontroller (MCU), which integrates a power management unit (PMU) and a system initialization controller. The Bayesian optimization module is integrated into the MCU's floating-point unit (FPU) and interacts with the initialization controller via a GPIO interface. After the system is connected to a 220V AC power supply, it is converted to 5V DC by a switching power supply. The MCU powers on and executes a reset procedure. After the reset is complete, the MCU starts the initialization sequence. When all initialization steps are completed, the initialization controller outputs a high-level trigger signal (lasting 200ms, amplitude 3.3V, meeting TTL level standards) to the Bayesian optimization module via the GPIO pin (PA1), triggering the Bayesian optimization process.

[0027] During the rapid growth phase of lettuce, rapid biomass accumulation is required, necessitating a high light response speed. Simultaneously, excessive light exposure must be avoided to prevent leaf burn, while ensuring the dimming stability of the LED lighting system. The core performance requirements are: Steady-state accuracy: Steady-state error 150 ,correspond 150 ; Overshoot suppression: Overshoot amount 8%; Controlling output stability: Controlling output penalty term 120.

[0028] In this example, the Bayesian weight optimization process is as follows: Considering the core requirement of "response speed priority" during the rapid growth phase, the initial weights are set with an emphasis on enhancing steady-state accuracy and response speed, while reserving basic weights for overshoot suppression and equipment protection. The initial weights are set to... , Higher weight than other weights, priority protection ITAE index; To meet the basic requirements for overshoot suppression; It also takes into account the protection of LED light groups and avoids excessive and frequent dimming.

[0029] Based on the light regulation requirements during the rapid growth period, the parameter space is set as follows: , , ; In the predefined parameter space Within, five sets of initial parameter configurations are selected through a random sampling strategy. , , , , .

[0030] Each set of parameters was configured to the digital PID controller in sequence, and the system was run in a closed loop for 30 minutes. The environmental sensing module collected the measured light intensity values ​​at an 80ms sampling period. The data is transmitted to the central control unit via an RS-485 bus. The central control unit calculates the performance indicators corresponding to each set of parameters. ; ; ( For the first (Control commands for the next sampling) Constructing the initial dataset A proxy model is constructed based on Gaussian processes, and the expected improvement criterion (EI) is used to iteratively select the next set of parameters to be evaluated. Until three consecutive iterations Value change 0.01, iteration terminates. EI strikes a balance between "exploring the unknown parameter region" and "utilizing the known optimal region" by quantizing parameters. Based on the expected performance improvement, select the next set of parameters with the greatest optimization potential to be evaluated, thereby efficiently converging to the global optimum.

[0031] First-round optimal PID parameters (baseline parameters): The corresponding control performance indicators are: ; ; ; Objective function value: .

[0032] After obtaining the baseline PID parameters and corresponding control performance, analysis and decision-making are performed: Overshoot This exceeds the crop's tolerance range and poses a risk of leaf burn; and The indicators have met the requirements, and There is redundancy; In the initial weights The value is too low, causing the Bayesian optimization process to overemphasize This sacrifices overshoot suppression performance; If it is too high, it can be reduced to release weight resources and increase the weight of overshoot. corresponding Once the indicators are met, improvements can be made to verify the effectiveness. Will The weight was reduced from 0.6 to 0.4 to retain the minimum weight required to meet steady-state accuracy requirements. The priority of overshoot suppression was increased from 0.2 to 0.3. The weighting is increased from 0.2 to 0.3, and the new weight combination is: .

[0033] The central control unit will combine the new weights. Write to the target function configuration register and update the target function to... .

[0034] Reset the dataset cache of the Bayesian optimization module to avoid interference from historical data, and randomly sample 5 sets of initial PID parameters again within the same parameter space: , , , , Keep the optimization iteration rules and termination conditions unchanged.

[0035] Second round of optimal PID parameters: The corresponding control performance indicators are: ; ; ; Objective function value: , The optimization effect is significant.

[0036] Based on the results of the second round of optimization, the final weight combination is determined. This combination satisfies the core performance indicators while achieving a balance between response speed, overshoot suppression, and device protection. To ensure the stability of the weight combination, a third round of Bayesian optimization (based on the final weights) is initiated, repeating the optimization process to verify the consistency and reliability of the optimal parameters. Third round of optimal PID parameters: The corresponding control performance indicators are: , , All indicators met the requirements, and the results differed from those of the second round. 5%, verifying the stability and reliability of the weight combination.

[0037] All core performance metrics ( 、 、 All requirements are met; the metrics corresponding to the unadjusted weights are within acceptable ranges; the weight combinations in the two consecutive rounds of optimization have not changed, and the performance metrics do not fluctuate. The convergence state was reached at 5%. Therefore, the final weight combination that satisfies all requirements is determined as follows: .

[0038] In this embodiment, the execution flow of the Bayesian optimization module is as follows: In this embodiment, the target light intensity The weight combination is The PID parameter space is defined as follows: The complete execution process of the Bayesian optimization module: The Bayesian optimization module reads the weight coefficients and objective function formula, automatically parses the formula and loads it into the floating-point unit (FPU) to complete the initialization of the computational logic; it reads the preset parameter range, verifies the rationality of the parameter space (minimum value < maximum value, step size adapts to accuracy requirements), and locks the parameter space after confirming that there are no errors. .

[0039] In the predefined parameter space Within, five sets of initial parameter configurations are selected through a random sampling strategy. , , , , Each set of parameters is configured to the digital PID controller in sequence, and the following process is executed for each set of parameters: Closed-loop operation: The PID controller operates based on current parameters. The system outputs dimming commands to control the LED light group to adjust the light intensity, and runs continuously for 30 minutes in a closed-loop environment. Data Acquisition: The environmental sensing module collects measured values ​​of light intensity at an 80ms sampling period. The data is transmitted to the central control unit via the RS-485 bus and stored in the memory buffer. Performance index calculation: Calculation ; ; ( For the first (Control commands for the next sampling) Constructing the initial dataset The details are as follows: A proxy model is constructed based on Gaussian processes, and the expected improvement criterion (EI) is used to iteratively select the next set of parameters to be evaluated. The criterion formula is ,in The minimum objective function value for the current dataset. Based on model predictions and EI value calculations, the selected... .

[0040] Will Write the parameter register of the PID controller, and collect the illumination data after running in closed loop for 30 minutes. Computational performance metrics: , , 105.2, objective function value ;Pair the new data Add to dataset The first iteration of expansion has been completed.

[0041] The iteration terminates when the change in the objective function value is ≤0.01 for three consecutive iterations. In this example, the termination condition is met on the 7th iteration, as detailed below: After the iteration terminates, the module traverses the final dataset. (Including 5 initial sets and 7 iterations, for a total of 12 sets of data), select the data that make the objective function... Minimal parameter combination: The corresponding objective function value Performance indicators , ... .

[0042] Will The final parameters of the PID controller are locked and sent to the digital PID controller of the planting area in Zone B via the RS-485 bus until the next optimization trigger.

[0043] Digital PID controllers use The digital PID controller uses internal parameters to perform closed-loop control of the system. Formulas, real-time calculation of control commands : This is a proportional operation, which generates a control action that is proportional to the current error magnitude. .

[0044] The integral operation generates a control action that is proportional to the integral of the error over time, used to eliminate the system's steady-state error. .

[0045] Differential operations generate a control effect proportional to the rate of change of the error, which can predict the future trend of the error and apply a reverse suppression effect, thereby reducing overshoot, suppressing oscillations, and improving the stability of the system. , .

[0046] Calculations were performed using parameter combinations Digital PID controller Control commands Send to the corresponding partition dimming driver module.

[0047] Above the cultivation racks in Zone B, a set of LED lights is installed. These lights are all connected to the output of the "Zone Dimming Driver Module B", forming the "LED Light Group B". Similarly, Zones A and C also have their own dedicated "LED Light Group A" and "LED Light Group C", which are completely independent in electrical connection and are driven by "Driver Module A" and "Driver Module C" respectively.

[0048] The zone dimming drive module receives the target control command from the central control unit and sends it to "zone dimming drive module B". The slave MCU generates a PWM signal with a duty cycle of 55.9% (initial duty cycle 50%). After optocoupler isolation, the PWM signal is sent to the MOSFET driver for current and voltage amplification. The amplified drive signal controls the gate of the power MOSFET, causing it to turn on and off at high speed, thereby driving LED lamp group B in the dimming device and increasing the light intensity to 300. The environmental sensing module detects changes in light intensity, generates a new PV value, and initiates a new adjustment cycle.

[0049] The central control unit has a built-in real-time clock (RTC) module and a growth cycle configuration register. During system initialization, after the user selects a crop variety (lettuce) via the touchscreen, the system automatically retrieves the preset cycles for each growth stage of the crop from the built-in database and writes the cycle parameters into the configuration register. When planting begins, the user clicks "Start Growth Cycle," and the RTC module starts timing and records the growth start timestamp. Every 10 seconds, the RTC module reads the current time, calculates the cumulative growth time with the start timestamp, and compares it with the stage cycle in the configuration register. When the cumulative time reaches the seedling stage cycle (7 days), the RTC module outputs a high-level signal (lasting 200ms) through the GPIO pin (PB0), and simultaneously updates the growth stage code from "seedling stage" to "growing stage," writing it into the stage status register. The Bayesian optimization module listens for changes in the PB0 pin level through the interrupt controller. Upon detecting a high level, it triggers an interrupt request, confirms the growth stage switch, and initiates the optimization process.

[0050] By implementing the above process, an integrated control system for light intensity in plant factories is developed. Through an environmental sensing module, a central control unit, and a zoned dimming drive module, the system achieves centralized control of light intensity in plant factories, solving the problem of insufficient control precision in traditional methods and improving the control quality and intelligence level of the light environment in plant factories.

[0051] In the description of this application, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0052] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0053] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0054] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0055] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0056] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

[0057] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

Claims

1. An integrated control system for light intensity in a plant factory, comprising an environmental sensing module, a central control unit, and a zoned dimming drive module, characterized in that: The environmental sensing module independently deploys sensors in each zone to directly measure the actual light intensity received by the plants and send the data to the central control unit. The central control unit has pre-stored target light intensity settings corresponding to different plant varieties and growth stages. It receives plant light intensity measurements from the environmental sensing module and compares the measurements reported by the environmental sensing module with the target light intensity settings using a built-in digital PID controller and Bayesian optimization module to calculate control commands. Send instructions to the local dimming driver module; The local dimming drive module receives instructions from the central control unit. It drives the dimming device to change according to instructions.

2. The system according to claim 1, characterized in that, The central control unit is equipped with a digital PID controller and a Bayesian optimization module: the Bayesian optimization module uses the parameter combination of the digital PID controller. To optimize variables, a preset comprehensive performance index was used. Given the objective function, the Bayesian optimization algorithm is used to automatically find a function that satisfies the objective function. Optimal PID parameter combination The optimized parameter set is configured to the digital PID controller; the built-in digital PID controller compares the measured values ​​reported by the environmental sensing module with the target setpoint to calculate the control command. It sends commands to the zone dimming driver module.

3. The integrated control system for light intensity in a plant factory according to claim 1, characterized in that, The central control unit has a built-in digital PID controller, which uses an incremental PID algorithm to output the control commands. .

4. The system according to claim 2, characterized in that, The objective function for: ,in, To assess the parameters of the closed-loop control system A comprehensive and quantitative evaluation of dynamic response quality and steady-state performance; The integral of time multiplied by the absolute error is used to measure the system's response speed and steady-state accuracy. , This is due to the error in light intensity. This is the overshoot, used to suppress light intensity overshoot. This represents the maximum measured light intensity during the system response process. It is the system target illumination intensity value; This is a penalty term for drastic changes in control output, used to protect the actuator and save energy. .

5. The system according to claim 4, characterized in that, They are respectively , , Weighting coefficients; Set an initial set of weights, run Bayesian optimization to obtain a set of baseline PID parameters and corresponding control performance, perform analysis and decision-making, adjust the corresponding weights, use the new weights, and rerun the entire Bayesian optimization process to find new PID parameters, verify whether they meet the requirements, and whether the unadjusted weights are within an acceptable range, until a combination of PID parameters that meets all requirements is found. weighting coefficient combination .

6. The system according to claim 1, characterized in that, The Bayesian optimization module receives the objective function with weights specified. In the predefined parameter space Internally, a set of initial parameter configurations is selected through a random sampling strategy. For each set of parameters The objective function value is obtained by running it on the control system and evaluating its performance. This constitutes the initial dataset. Based on this dataset, a Gaussian process is constructed as a surrogate model to predict... ; Selected parameters A digital PID controller is configured to run the control system in a closed-loop environment for a predetermined period of time, during which the light intensity is collected, and the comprehensive performance index is calculated based on this data. Numerical values, pairing new data Add to dataset To expand the observation sample; After the iteration process terminates, from the final dataset Selecting from the options makes the objective function The combination of parameters that achieves the minimum value As an optimization result, it is locked into the final parameters of the PID controller.

7. The system according to claim 1, characterized in that, Digital PID controller based on Formulas, real-time calculation of control commands ; For proportional calculations, ; in For error, ; For integration operations, ; in From the 1st to the 1st Sum of all errors, For the first Error The sampling period; For differentiation operations, ; in for Error in time; Calculations were performed using parameter combinations Digital PID controller Control commands Send to the corresponding partition dimming driver module.

8. The system according to claim 1, characterized in that, The Bayesian optimization module is configured to automatically trigger the optimization process when the system starts up or the growth phase switches.

9. The system according to claim 1, characterized in that, The system is configured with an independent digital PID controller and Bayesian optimization process for each planting zone.

10. The system according to claim 1, characterized in that, The zone dimming drive module includes a dimming device, which consists of multiple LED light groups, each of which corresponds to a planting zone.

Images