An internet of things driven tobacco growth environment multi-parameter intelligent regulation and control method

Abstract

The application discloses a kind of tobacco growth environment multi-parameter intelligent regulation and control methods driven by Internet of Things, it is related to agricultural Internet of Things and environmental intelligent regulation and control technical field, the method includes: based on the environment parameters of tobacco planting area and crop sap flow data collected in real time, in combination with the current growth stage of tobacco and quality target, the target concentration change trajectory of the key metabolite representing the internal physiological activity of tobacco is dynamically generated;With target concentration change trajectory as optimization goal, crop physiological mechanism model describing the relationship between environment and physiological process is used to solve reversely, and the target trajectory of the environment parameter setting value required to realize optimization goal is calculated with time change;According to target trajectory, in combination with pre-established equipment disturbance propagation model, the coordinated action sequence of environmental regulation equipment is generated by solving a field control inversion problem;The coordinated action sequence of equipment is executed to control corresponding environmental regulation equipment to work.

Description

Technical Field

[0001] This application relates to the field of agricultural Internet of Things and intelligent environmental control technology, and more specifically to the field of precise control of the growth environment in tobacco agriculture. In particular, it relates to an Internet of Things-driven method for intelligent control of multiple parameters of the tobacco growth environment. Background Technology

[0002] Tobacco, as an important economic crop in my country, directly impacts the economic benefits and market competitiveness of the tobacco industry through its quality and yield. Growing environmental conditions are key factors influencing tobacco growth, development, and the formation of its intrinsic quality. Tobacco exhibits significantly different requirements for environmental parameters such as temperature, humidity, light, CO2 concentration, and soil moisture at different growth stages. If environmental conditions deviate from suitable ranges, it will not only slow down tobacco growth and increase the risk of pests and diseases, but may also lead to an imbalance in the internal chemical composition of tobacco leaves, reducing raw material utilization. With the advancement of modern agricultural intelligence, utilizing Internet of Things (IoT) technology to achieve precise control of the tobacco growing environment has become an important way to overcome the limitations of traditional manual management, ensure the stability of tobacco quality, and improve production efficiency. This has significant practical implications for promoting the development of tobacco agriculture towards high efficiency, high quality, and sustainability.

[0003] While existing tobacco growth environment control technologies have gradually incorporated intelligent monitoring methods, significant shortcomings remain. Most control schemes rely on preset environmental parameter thresholds or simple empirical models, failing to fully consider the dynamic changes in the tobacco's own growth state and thus making accurate control difficult.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides an Internet of Things-driven intelligent multi-parameter control method for tobacco growth environment to solve the above-mentioned technical problems.

[0006] This application provides an IoT-driven method for intelligent multi-parameter control of tobacco growth environment, including:

[0007] Based on real-time collected environmental parameters and crop stem flow data of the tobacco planting area, combined with the current growth stage and quality targets of tobacco, a target concentration change trajectory of key metabolites characterizing the internal physiological activities of tobacco is dynamically generated. Using this target concentration change trajectory as the optimization objective, a crop physiological mechanism model describing the relationship between the environment and physiological processes is used for inverse solving to calculate the target trajectory of environmental parameter setpoints changing over time to achieve the optimization objective. Based on this target trajectory, and combined with a pre-established equipment disturbance propagation model, a coordinated action sequence of environmental control equipment is generated by solving a field control inversion problem. The equipment disturbance propagation model defines the spatiotemporal disturbance characteristics of each equipment action on the environmental state at different locations within the planting space. The coordinated action sequence of the equipment is executed to control the corresponding environmental control equipment. The control effect is evaluated based on the temporal changes in both environmental parameters and crop stem flow data, and the key physiological parameters in the crop physiological mechanism model and the spatial transmission characteristics in the equipment disturbance propagation model are dynamically calibrated based on the evaluation results.

[0008] Based on the embodiments provided in this application, by dynamically generating target concentration change trajectories of key metabolites based on real-time environmental parameters and crop stem flow data, combined with the current growth stage and quality goals of tobacco, the regulatory targets can be directly anchored to the core needs of internal physiological activities in tobacco. This avoids the disconnect between the regulatory direction and the intrinsic growth mechanism of tobacco, making environmental regulation more aligned with the physiological essence of tobacco forming target quality at different growth stages. By using the target concentration change trajectory as the optimization objective and inversely solving the target trajectory using a crop physiological mechanism model, the regulatory targets of environmental parameters can be adaptively adjusted according to the dynamic changes in tobacco metabolic activities. This overcomes the limitations of relying on fixed thresholds or empirical models, ensuring that environmental parameters are always precisely matched with the real-time physiological needs of tobacco.

[0009] By combining a device disturbance propagation model and generating a sequence of coordinated actions of the devices through solving the field control inversion problem, the spatiotemporal disturbance differences of each device's actions on the environmental state at different locations in the planting space are fully considered. This effectively reduces mutual interference when multiple devices are working, improves the spatial uniformity of the environmental state in the planting area, and avoids the impact of local environmental fluctuations on the uniformity of tobacco growth. By evaluating the regulation effect based on the temporal changes of environmental parameters and stem flow data, and dynamically calibrating the key parameters of the crop physiological mechanism model and the spatial transfer characteristics of the device disturbance propagation model, a closed-loop optimization mechanism is formed. This mechanism enables the regulation model to continuously adapt to changes in tobacco growth status and device operating characteristics, ensuring the stability and accuracy of long-term regulation effects and contributing to the reliable achievement of tobacco quality goals. Attached Figure Description

[0010] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0011] Figure 1 This is a flowchart of an optional IoT-driven intelligent multi-parameter control method for tobacco growth environment according to an embodiment of this application;

[0012] Figure 2 This is a flowchart of another optional IoT-driven intelligent multi-parameter control method for tobacco growth environment according to an embodiment of this application;

[0013] Figure 3 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application.

[0014] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0016] According to one aspect of the embodiments of this application, such as Figure 1 As shown, this application provides an IoT-driven intelligent multi-parameter control method for tobacco growth environment, including:

[0017] S101, based on real-time collected environmental parameters and crop stem flow data of tobacco planting areas, combined with the current growth stage and quality targets of tobacco, dynamically generates the target concentration change trajectory of key metabolites that characterize the internal physiological activities of tobacco.

[0018] In S101, the regulatory target shifts from external environmental parameters to the crop's own physiological activities. By integrating real-time environmental data with stem flow—a key physiological signal—and combining growth stage and quality targets, the trajectory of target concentration changes, representing internal physiological activities, is dynamically deduced. This is equivalent to establishing a quantified, dynamically changing physiological target centered on the crop's internal state for the entire regulatory system, rather than a fixed environmental set point. Its significance lies in transforming regulatory behavior from "maintaining the environment" to "achieving the ideal physiological state of the crop."

[0019] It needs to be explained that the "dynamic" nature of step S101 is reflected in two aspects. First, it is data-driven and triggered in real time. The generation process is driven by real-time collected environmental parameters and crop stem flow data. Data is acquired every 5-10 minutes (one monitoring cycle). When the data indicates a change in the crop's physiological state or a trend change in the external environment, the recalculation and update of the target concentration change trajectory is triggered. Second, it involves online refreshing of model parameters. The key parameters of the core model used to generate the trajectory are not fixed but are dynamically called and fine-tuned from a preset parameter library based on the crop's current growth stage and historical environmental data, so that the generated trajectory can adapt to the dynamic physiological needs of the crop.

[0020] The core principle is to take quality targets as the ultimate guide and real-time data as the calibration basis. First, a basic trajectory is generated based on photosynthetic models and metabolic allocation strategies. Then, a stress detection and response mechanism is used to correct this basic trajectory in real time, ultimately outputting a target concentration change trajectory that closely matches the actual needs of the crop.

[0021] S102, taking the target concentration change trajectory as the optimization objective, the crop physiological mechanism model describing the relationship between the environment and physiological processes is used for inverse solution to calculate the target trajectory of the environmental parameter set value required to achieve the optimization objective as a function of time.

[0022] Using real-time collected environmental parameters and stem flow data as model input and feedback, the model is solved in reverse using a crop physiological mechanism model, and the target trajectory changes continuously over time.

[0023] In this invention, the crop physiological mechanism model is an integrated concept. It does not refer to a single, fixed set of mathematical equations, but rather to a computational framework for describing key physiological processes in tobacco (such as photosynthesis, material transport, and water transport) and their interactions with environmental factors. The specific implementation of this model can be manifested in the collaborative work of components such as a dynamic physiological causal graph, a simplified metabolic flow model for rapid calculation, and a stem flow prediction model.

[0024] In S102, a reverse thinking and mechanism model-driven solution method was employed. Using the generated internal physiological target (target concentration change trajectory) as the endpoint, a crop physiological mechanism model was used to reverse-engineer the required external environment (target trajectory) to drive the crop to achieve this internal target. This is equivalent to deducing "what the environment should provide" from "what the crop wants" in a virtual space. The effect is that the generated target trajectory has a solid physiological basis, ensuring that any environmental intervention directly serves a specific physiological purpose, greatly improving the targetedness and scientific rigor of regulation.

[0025] S103, based on the target trajectory and combined with the pre-established equipment disturbance propagation model, a coordinated action sequence of environmental control equipment is generated by solving a field control inversion problem; wherein, the equipment disturbance propagation model defines the spatiotemporal disturbance characteristics of each equipment action on the environmental state at different locations within the planting space;

[0026] The field control inversion problem of this invention is mathematically formulated as a constrained spatiotemporal optimization problem. Its objective function is to minimize the root mean square error between the actual environmental field and the desired environmental field in the main region of the crop canopy. The constraints include physical constraints of the equipment itself, energy consumption economic constraints, and logical mutual exclusion constraints between the actions of different equipment.

[0027] The impact of equipment actions on the planting space is not uniform but exhibits spatiotemporal perturbation characteristics. In S103, a "spatial expectation field increment" is formed by comparing the target trajectory with the current state. Using this increment as the objective, the optimal cooperative action sequence is solved from a predefined equipment action primitive library. This achieves a mapping from a one-dimensional environmental target to three-dimensional spatial control, and avoids mutual interference between equipment through the cooperative action sequence, ensuring that the environment within the entire space can accurately and efficiently approximate the target trajectory.

[0028] S104, Execute the coordinated action sequence of the equipment to control the corresponding environmental control equipment to work;

[0029] S105, evaluate the regulatory effect based on the temporal changes of environmental parameters and crop stem flow data, and dynamically calibrate the key physiological parameters in the crop physiological mechanism model and the spatial transmission characteristics in the equipment disturbance propagation model based on the evaluation results.

[0030] In S105, a self-learning, adaptive feedback optimization mechanism was constructed. It does not execute all at once, but continuously monitors the temporal changes in environmental parameters and crop stem flow, evaluating the regulatory effect from two dimensions: "environmental tracking accuracy" and "physiological response effect." Based on the evaluation results, key parameters in the crop physiological mechanism model and the equipment disturbance propagation model are dynamically calibrated. Its core effect is to enable the system to overcome uncertainties caused by initial model errors, changes in equipment performance, and crop growth dynamics, allowing the model to increasingly conform to actual conditions over time, thereby ensuring the stability and reliability of long-term regulatory effects.

[0031] Furthermore, based on real-time collected environmental parameters and crop stem flow data from tobacco-growing areas, combined with the current growth stage and quality targets of tobacco, the target concentration change trajectory of key metabolites characterizing the internal physiological activities of tobacco is dynamically generated, including:

[0032] The light intensity and temperature data from the real-time collected environmental parameters are input into the preset primary productivity model of photosynthesis to output the basic synthesis rate curve of carbohydrates over time.

[0033] As a specific example, this invention can use a non-rectangular hyperbolic light response model as the core of the primary productivity model of photosynthesis. This model can fit the light saturation phenomenon well, and its mathematical expression includes parameters such as photosynthetically active radiation flux density, maximum total photosynthetic rate, apparent quantum efficiency, convexity factor, and dark respiration rate, all of which are well-known technologies and will not be elaborated upon in this embodiment.

[0034] By inputting real-time collected light intensity and temperature data into the model, the basal carbohydrate synthesis rate curve can be output. The model parameter values ​​are determined based on publicly available physiological research literature on different tobacco varieties.

[0035] The current growth stage of tobacco is used as an index to query the preset metabolic allocation strategy database to obtain allocation parameters; the basic synthesis rate curve is decomposed into a first synthesis rate sub-trajectory for growth and a second synthesis rate sub-trajectory for accumulation and transport using the allocation parameters.

[0036] The metabolic allocation strategy database is implemented in the form of a two-dimensional lookup table, with row indexes representing reproductive stages and column indexes representing allocation targets. The data in the table is derived from a combination of long-term field trials and literature data.

[0037] To clearly define the allocation process, this invention introduces an allocation coefficient function with reproductive stage as the variable. Using this allocation parameter, the basic synthesis rate curve was plotted. The specific formula for decomposition is as follows:

[0038]

[0039]

[0040] in, It is the first synthetic rate sub-trajectory used for growth. It is the second synthetic rate sub-trajectory used for accumulation and transport. It is the basic synthesis rate curve that varies with time. It is a growth allocation coefficient function related to the reproductive stage. Based on this formula, the abstract "metabolic allocation strategy" is transformed into a precise mathematical model. The core calculation step to achieve "on-demand regulation" is to guide the flow of photosynthetic products through dynamically changing coefficients.

[0041] By combining the nicotine content set in the quality target with the number of days from the current date to the planned harvest date, the average daily demand rate of nicotine precursor substances is calculated to generate a translocation rate curve with a baseline level.

[0042] The purpose of this step is to break down the ultimate quality target (nicotine content) into the crop's daily physiological tasks. The process is as follows:

[0043] Input: The nicotine content per tobacco plant or unit area at the planned harvest, as set in the quality objectives (e.g., 2.5% of dry weight), and the total number of days from the current date to the planned harvest date (e.g., 60 days).

[0044] Calculation: First, based on crop variety and historical data, estimate the overall conversion efficiency (e.g., 40%) from nicotine precursors (such as nicotinic acid) to nicotine synthesis and eventual accumulation. Next, calculate the total amount of nicotine precursors required to synthesize over the remaining growing season. Conceptually, the formula is: Total precursor requirement = Target nicotine content / Overall conversion efficiency. Finally, distribute the total requirement evenly across each day to obtain the daily demand rate. That is: Daily demand rate = Total precursor requirement / Remaining days.

[0045] Output: Based on the average daily demand rate, the system generates a flat or slightly fluctuating baseline translocation rate curve. This curve indicates that, ideally, crops need to stably synthesize and translocate a certain amount of precursor substances each day to achieve the target quality at harvest.

[0046] Based on the basic synthesis rate curve, the first synthesis rate sub-trajectory, the second synthesis rate sub-trajectory and the transport rate curve, the initial set of metabolic flux trajectories is determined.

[0047] The real-time crop stem flow data is compared with the preset normal range; when the data is abnormal, the attenuation coefficient is obtained according to the preset stress response coefficient table to globally adjust all curves in the initial metabolic flow trajectory set and output the target concentration change trajectory.

[0048] The stress response coefficient table is a crucial empirical database. For example, when the system detects that the stem flow rate is consistently 10%-20% below the normal range, it is classified as mild water stress, with a corresponding attenuation coefficient of 0.9. The physiological basis for this is that partial stomatal closure restricts the supply of photosynthetic substrates, necessitating a moderate downsizing of the synthesis target. When the stem flow rate is consistently more than 20% below the normal range, it is classified as severe water stress, with an attenuation coefficient of 0.7. This is based on the fact that significant stomatal closure severely inhibits photosynthesis. When the system determines that it has entered a certain stress state, it reads the corresponding attenuation coefficient from this table and globally adjusts all curves in the initial metabolic flow trajectory set.

[0049] It should be noted that, taking the stem flow rate of mature tobacco as an example, its preset normal range can be set to 800 to 1500 grams per hour. This range is a statistical range derived from years of planting practice by statistically analyzing the stem flow data of healthy tobacco plants with sufficient water supply on typical sunny days (for example, taking the interval from the 5th to the 95th percentile).

[0050] The process of determining the initial set of metabolic flow trajectories is not to generate a single trajectory, but rather to construct an initial set of schemes that includes multiple objectives.

[0051] Basic trajectory: The basic synthesis rate curve is directly calculated from the photosynthesis model and allocation strategy.

[0052] Growth-preferred trajectory: On top of the base trajectory, the amplitude of the "first synthesis rate sub-trajectory" used for growth is enhanced to form a scheme that is biased towards promoting plant growth.

[0053] Quality-first trajectory: On the base trajectory, the amplitude of the "second synthesis rate sub-trajectory" and the "transport rate curve" used for accumulation and transport is enhanced to form a scheme biased towards the accumulation of secondary metabolites.

[0054] This set provides a variety of feasible initial options for subsequent reverse engineering.

[0055] Based on the embodiments provided in this application, external environmental monitoring (such as light and temperature) is coupled with internal physiological activities (such as carbohydrate synthesis and nicotine precursor transport) at multiple levels. A preset model and database are used to achieve refined decomposition and adjustment of metabolic flow. First, a basic synthesis rate is output through a photosynthesis model. Then, metabolic pathways are dynamically allocated according to the growth stage. Finally, a stress response coefficient is introduced through stem flow anomaly detection for global calibration. This achieves simulation of internal metabolic activities in tobacco, enabling the regulatory system to make decisions from a physiological level rather than just an environmental level, thus improving the accuracy of regulation. By comparing real-time stem flow data with normal ranges, stress states can be quickly identified and metabolic trajectories can be automatically adjusted, enhancing adaptability in variable environments. Converting nicotine content targets into daily demand rates ensures the continuity and traceability of quality control, avoiding the regulatory lag caused by ambiguous targets in traditional methods.

[0056] Furthermore, the method also includes:

[0057] When the light intensity data is below the light compensation point threshold and the stem flow rate data shows a downward trend for three consecutive monitoring cycles, it is determined that the state of weak light stress has been entered.

[0058] Under low light stress, the peak value of the second synthesis rate sub-trajectory is reduced by a first preset ratio, while the duration of its synthesis rate being greater than zero is extended by a fixed period of time.

[0059] When the canopy temperature data exceeds the high temperature threshold and the fluctuation range of the stem flow rate data exceeds the normal fluctuation range in two consecutive monitoring cycles, it is determined that the state of high temperature stress has been entered.

[0060] Specifically, for the flue-cured tobacco variety K326 used in this scheme, the light compensation point threshold was determined to be 35 μmol·m at a canopy temperature of 25℃. -2 ·s -1 This value was obtained by consulting the plant physiology manual for this variety, indicating that under this light intensity, the amount of carbohydrates produced by photosynthesis in the leaves is balanced with the amount consumed by respiration.

[0061] In this embodiment, the high-temperature threshold is set at 32°C. This threshold is determined based on the consensus in tobacco cultivation: when the canopy temperature consistently exceeds this value, the efficiency of crop photosystem II decreases significantly, respiration consumption intensifies, and growth is inhibited. For mature tobacco plants, under conditions of sufficient water supply and sunny weather, the normal fluctuation range of stem flow rate is typically 800 to 1500 grams per hour. This range was obtained by statistically analyzing a large amount of stem flow data under historical normal growth conditions (e.g., using the 5th to 95th percentile).

[0062] It should be explained that three consecutive monitoring periods are chosen to determine low light stress because short-term light intensity fluctuations (such as a cloud passing by) do not immediately cause irreversible changes in the crop's physiological state. Three consecutive periods (approximately 15-30 minutes) of sustained low light intensity accompanied by a decrease in stem flow more reliably indicate that the crop has entered a sustained stress physiological response mode, rather than a momentary fluctuation.

[0063] Two consecutive monitoring cycles were chosen to determine high-temperature stress because the damage caused by high temperatures to crops is rapid and cumulative. When the canopy temperature exceeds the high-temperature threshold for two cycles (approximately 10-20 minutes) and stem flow exhibits abnormally high fluctuations (indicating that the transpiration cooling system is under stress), it indicates that the risk of heat stress is very urgent and the system needs to respond immediately to avoid irreversible damage to leaf tissues.

[0064] Under high-temperature stress, the peak occurrence period of the transport rate curve is adjusted from daytime to the lowest temperature period in the early morning of the next day, and its peak amplitude is adjusted according to the second preset ratio.

[0065] The first preset ratio, for example, 20%. Under low light stress, the system reduces the peak value of the "second synthesis rate sub-trajectory" used for accumulation and translocation by 20% to reflect the reality of reduced photosynthetic output. A fixed time period, for example, 2 hours. This extends the duration of the synthesis rate being greater than zero in this sub-trajectory by 2 hours, simulating the physiological compensation mechanism of crops in stress by extending effective metabolic time to compensate for insufficient efficiency. The second preset ratio under high temperature stress, for example, an increase of 15%. Under high temperature stress, the peak amplitude of the "translocation rate curve" is increased by 15%, and its occurrence time is adjusted to the cool early morning. This simulates the physiological adaptive behavior of crops to avoid high temperature stress periods and accelerate translocation tasks at suitable temperatures.

[0066] Based on the embodiments provided in this application, stress state determination is combined with metabolic allocation logic to optimize resource allocation by modifying the peak value and time period of the synthesis rate sub-trajectory. State determination is based on environmental parameters and stem flow trends within a continuous monitoring period, and the peak value and duration of the second synthesis rate sub-trajectory (used for accumulation and translocation) are adjusted under stress, or the time period of the translocation rate curve is replanned under high temperature. Under low light stress, by reducing the peak value and extending the duration, the energy-saving mechanism of plants under low light conditions is simulated, reducing ineffective metabolic consumption while maintaining basic growth requirements. Under high temperature stress, the translocation peak value is adjusted to a lower temperature period, utilizing the stability of nighttime physiological activity and avoiding the inhibition of nicotine precursor accumulation by high temperature, thereby improving the reliability of quality regulation.

[0067] In some embodiments, the following correspondence can be established: A dynamic physiological causal graph is the qualitative or semi-quantitative topological structure of a crop physiological mechanism model. It describes the causal relationship network between various physiological and environmental variables and forms the core skeleton of the model. A simplified metabolic flow model is the core of metabolic process calculations used for rapid validation in a crop physiological mechanism model. It is an engineering simplification of a complete, complex mechanism model, focusing on the prediction of metabolic flows. A stem flow prediction model is a submodule in a crop physiological mechanism model specifically used to simulate water transport and determine physiological stability.

[0068] Furthermore, such as Figure 2 As shown, taking the target concentration change trajectory as the optimization objective, a crop physiological mechanism model describing the relationship between the environment and physiological processes is used for inverse solution to calculate the target trajectory of the environmental parameter setpoints required to achieve the optimization objective over time, including:

[0069] S201, construct a dynamic physiological causal graph, whose nodes include environmental parameters and metabolic flow rates;

[0070] In the dynamic physiological causal graph, nodes include key environmental parameter nodes (light intensity, carbon dioxide concentration, canopy temperature, air humidity) and metabolite nodes (instantaneous carbohydrate synthesis rate, nicotine precursor transport rate). Edges represent causal relationships between nodes, pointing from cause to effect. For example, there is an edge from "light intensity" to "instantaneous carbohydrate synthesis rate," and an edge from "canopy temperature" to "nicotine precursor transport rate." Dynamic causal weight allocation is implemented, taking into account both the reproductive stage and stem flow stability. First, a basic weight matrix Wbase(Stage) is pre-stored within the system, where each element... This represents the base weight of the causal influence of node i on node j at a specific growth stage (Stage). Then, a stem flow stability adjustment factor γ(ApEn) is introduced, which is a function of the real-time stem flow approximate entropy ApEn. When ApEn is low (poor stability), the value of γ increases, thereby amplifying the weights of edges related to water and energy transport. The final dynamic weight Wdynamic = Wbase(Stage) multiplied by γ(ApEn).

[0071] S202, based on the current reproductive stage and real-time stalk flow stability, dynamically allocates the causal weights of the edges in the physiological causal graph; starting from the target node of the target concentration change trajectory, it performs an inverse graph search to identify the key environmental parameters and their sensitive periods that have the highest impact on achieving the target, thereby focusing the full parameter optimization problem on the key subset;

[0072] In some embodiments, the system stores a basic weight table that defines the baseline influence weights of various environmental parameters on metabolic flow at different growth stages (e.g., the highest weight for "light → carbohydrate synthesis" during the vigorous growth stage). During operation, the system incorporates real-time stem flow stability as a regulating factor. When stem flow stability is high, the system trusts the basic weights; when stability decreases, the system automatically increases the causal weights related to environmental parameters that directly affect water transpiration and transport, such as "canopy temperature" and "air humidity." This means that when the crop is "unwell," the control system prioritizes its core physiological stability.

[0073] Starting with the target node of the desired concentration change trajectory (e.g., the peak rate of carbon synthesis at 12 noon), the system traces backward in the dynamic causal graph. Along the causal edges, the system searches for environmental parameter nodes that point to the target node and currently have the highest dynamic weight (e.g., light intensity at 11 am and CO2 concentration at 10 am), and marks these parameters and their corresponding time periods (i.e., "sensitive periods"). This process focuses the optimization of thousands of variables onto a few key environmental parameters and critical time periods, significantly improving the efficiency of subsequent solutions.

[0074] S203 determines the physiological state of the crop based on real-time collected environmental parameters and crop stem flow data; based on different states, it calls the corresponding optimization algorithm from the strategy library, which includes global exploration strategy, conservative repair strategy and economic energy-saving strategy, and generates multiple different candidate environmental parameter trajectories in parallel.

[0075] The physiological conditions include: "Stable growth condition": canopy temperature is 22-28℃, stem flow approximate entropy is greater than 0.6, and light is sufficient. "Water stress condition": stem flow rate is consistently below 15% of the lower limit of the normal range, and stem flow approximate entropy is below 0.5.

[0076] The strategy library includes: "Global Exploration Strategy," which corresponds to a genetic algorithm and is suitable for "stable growth conditions," aiming to broadly search for potentially better new environmental solutions. "Conservative Repair Strategy," corresponding to a strongly constrained sequential quadratic programming method, is suitable for stress conditions and aims to quickly and smoothly restore crops to a safe state during physiologically sensitive periods. "Economic and Energy-Saving Strategy," corresponding to a particle swarm optimization algorithm with energy consumption as a penalty, is suitable when environmental goals are easily achieved, prioritizing the reduction of operating costs.

[0077] S204 uses a simplified metabolic flow model to verify and screen candidate environmental parameter trajectories, and outputs an initial solution set. For each trajectory in the initial solution set, an adjoint method channel is activated for adjustment. The adjoint method channel uses an attention mechanism to prioritize gradient calculation and optimization during a defined sensitive period. At the same time, an imitation learning channel is activated to perform pattern matching and fragment fusion between the trajectory and historical successful trajectories.

[0078] The Simplified Metabolic Flux Model is a real-time computational version of the crop physiological mechanism model. It linearizes complex differential equations around the operating point; focuses on core metabolic fluxes such as carbohydrate synthesis and transport, ignoring secondary pathways; and uses lookup tables for key physiological parameters instead of real-time calculation. It is a set of algebraic equations that take environmental parameters as input and predict metabolic flux values ​​as output, used to quickly verify the physiological plausibility of candidate environmental trajectories.

[0079] The adjoint method is a technique for efficiently calculating the gradient of the objective function by constructing an "adjoint system" corresponding to the original differential equation system. In this scenario, it is used to quickly determine how to adjust the environmental trajectory to minimize the error between the predicted metabolic flux and the target value.

[0080] The attention mechanism achieves focus on sensitive periods through a time-weighted function. Therefore, this invention constructs a loss function L that is solved inversely.

[0081]

[0082] in, This represents the loss function, used to quantify the overall bias. This represents the total number of time points within the optimization period. Indicates at time The attention weight is set to a higher value (e.g., 5.0) during sensitive periods and 1.0 during non-sensitive periods. Indicates at time Predicted metabolic flow rate (unit: μmol·m) -2 ·s - ¹). Indicates at time Target concentration change trajectory Target value (unit: μmol·m -2 ·s - ¹). Reference values ​​for metabolic flux (unit: μmol·m) -2 ·s - ¹), for example, determined based on the maximum or average metabolic flow rate in historical data, for normalization. Indicates at time Environmental parameter values ​​(e.g., temperature in °C, light intensity in μmol·m⁻²) -2 ·s - ¹). Reference values ​​representing changes in environmental parameters (units relative to environmental parameters) Consistency), for example, is determined based on the maximum or typical variation of environmental parameters allowed in historical data, and is used for normalization. This represents the weighting coefficient, used to balance the importance of metabolic flux tracking error and the penalty term for changes in environmental parameters.

[0083] The selection criteria for historical successful trajectories are: environmental regulation records that ultimately resulted in a metabolic flux achievement rate higher than 0.9 and low energy consumption under similar past reproductive stages and external weather types. Pattern matching employs a dynamic time warping algorithm to align candidate trajectories with historical successful trajectories on the time axis. Fragment fusion involves replacing or weighting corresponding portions of candidate trajectories with fragments from historical successful trajectories during similar time periods in the matching process, in order to introduce proven effective regulation patterns.

[0084] S205, based on the real-time stem flow stability index, the solutions output by the adjoint method channel and the solutions output by the imitation learning channel are weighted and fused to generate a preliminary target trajectory; the preliminary target trajectory is input into the stem flow prediction model to predict the stem flow sequence generated after executing the trajectory;

[0085] The stem flow prediction model can be implemented using a mechanism-based empirical regression model or a data-driven machine learning model. Such model construction methods are mature technologies in this field.

[0086] Specifically, one approach is to establish empirical formulas or lookup tables based on crop physiological mechanisms and field experiments, using environmental factors (such as light, temperature, and humidity) and crop growth stages as input variables, to achieve rapid estimation of stem flow rate. Another approach is to employ lightweight machine learning models, such as Support Vector Regression (SVR) or computationally efficient Gated Recurrent Unit (GRU) networks, trained using historical stem flow data and environmental data to obtain predictive models.

[0087] S206, calculate the approximate entropy of the predicted stem flow sequence. If the approximate entropy is lower than the safe entropy threshold, it is determined that the trajectory has the risk of causing physiological instability. When the risk of instability is detected, adjust the causal weights of the nodes related to water transport stability in the dynamic causal graph and trigger a new round of reverse solution process.

[0088] S207, if the predicted approximate entropy is higher than or equal to the safety entropy threshold, then output the current preliminary trajectory as the target trajectory.

[0089] Analysis of stem flow data from a large number of healthy tobacco plants revealed that their approximate entropy typically falls within a stable range. Therefore, 0.5 can be set as a safe entropy threshold. When the approximate entropy of the predicted stem flow sequence falls below 0.5, it is considered to pose a high risk of instability.

[0090] When an instability risk is detected, the system does not directly output the environmental trajectory of the risky environment. Instead, it initiates a closed-loop correction process:

[0091] Adjusting the causal graph: Significantly increase the causal weights of nodes directly related to water transport stability in the dynamic causal graph, such as "air humidity" and "canopy temperature".

[0092] Triggering a re-solution: Using the adjusted (now with a greater focus on stability) causal graph, perform a reverse graph search and subsequent optimization solution.

[0093] Iteration: The newly generated environmental trajectory instinctively avoids parameter combinations that cause instability in water transport (such as excessively high temperature and excessively low humidity), thus generating a safer and more robust new trajectory at the source. This process can be repeated until the predicted stem flow stability meets the target.

[0094] Based on the embodiments provided in this application, dynamic causal graphs and weight allocation enable the system to focus on environmental parameters that have the greatest impact on metabolic flux, reducing computational complexity and improving solution efficiency. Parallel generation of candidate trajectories combined with multiple optimization channels (such as adjoint methods and imitation learning) enhances the diversity and feasibility of trajectories, avoiding local optima. Through stem flow prediction and approximate entropy checking, physiological instability risks can be identified in advance and the solution recalculated, ensuring the safety of the regulatory process. Overall, this method combines reverse reasoning with machine learning to achieve a high-precision mapping from physiological goals to environmental parameters.

[0095] Furthermore, in the process of generating candidate environmental parameter trajectories and preliminary target trajectories, photosynthetic rhythm illumination constraints and carbon balance temperature constraints are introduced; among them,

[0096] The photosynthetic rhythm light constraint conditions include: when generating the light intensity parameter trajectory, the constraint curve of its change needs to simulate the diurnal variation pattern of natural light intensity, that is, the light intensity rises non-linearly from zero or the base value to the daytime peak in the first preset period after sunrise, and falls non-linearly from the daytime peak to the nighttime base value in the second preset period before sunset.

[0097] The nonlinear rise / fall is simulated using an S-shaped growth curve (Logistic function). This is because the diurnal variation of natural light intensity is not nonlinear, but rather a process of accelerated growth after sunrise, slowing growth around noon, and accelerated decay before sunset. The S-shaped curve can well fit this "slow-fast-slow" natural rhythm.

[0098] The carbon balance temperature constraints include: when generating the nighttime temperature parameter trajectory, the value of the parameter must be calculated based on the virtual amount of cumulative photosynthetic products during the day, so that the predicted nighttime respiration consumption of crops at this temperature and the virtual amount of cumulative photosynthetic products during the day are maintained within a ratio range dynamically set according to the quality target.

[0099] The first period (dawn onset) refers to the time from sunrise to noon, such as from local sunrise time to 11:00. The second period (twilight decay) refers to the time from noon to sunset, such as from 15:00 to local sunset time.

[0100] Daytime peak intensity: refers to the highest light intensity that needs to be reached between the first and second time periods. For example, for a light-loving crop like tobacco, the midday peak intensity can be set at 1200 μmol·m⁻² under sunny weather conditions. -2 ·s -1 Nighttime baseline: refers to the minimum light level required at night to maintain basic crop metabolism or certain special processes (such as dark fixation), for example, 0 μmol·m⁻². -2 ·s -¹50 μmol·m⁻¹ under complete darkness or supplemental light -2 ·s -1 .

[0101] Dynamically set ratio range: This refers to the ratio of nighttime respiration consumption to daytime photosynthetic products. This range is dynamically set based on quality objectives. For example, when pursuing a high carbon-to-nitrogen ratio and thick leaves, a lower ratio range (e.g., 10%-15%) is set to minimize nighttime carbon consumption and promote nutrient accumulation. When pursuing a normal growth rate, a medium ratio range (e.g., 15%-25%) is set.

[0102] Based on the embodiments provided in this application, the photosynthetic rhythm constraint requires the light intensity curve to simulate the nonlinear changes of sunrise and sunset, while the carbon balance constraint dynamically sets the nighttime temperature based on the virtual amount of daytime photosynthetic products to maintain the ratio of respiration consumption to photosynthetic products. The photosynthetic rhythm constraint makes artificial lighting closer to the natural light environment, reducing the stress response of plants caused by sudden changes in light intensity and promoting stable photosynthesis. The carbon balance temperature constraint ensures the matching of nighttime respiration consumption with daytime energy accumulation through virtual quantity calculation, avoiding energy waste or shortage, thereby supporting the achievement of quality goals. These constraints embed plant physiological laws into parameter generation, enhancing the biological rationality of the environmental trajectory.

[0103] Furthermore, based on the target trajectory and combined with a pre-established equipment disturbance propagation model, a coordinated action sequence of environmental control equipment is generated by solving a field control inversion problem, including:

[0104] It should be noted that the pre-established equipment disturbance propagation model in this invention refers to a mathematical model system used to quantify the spatiotemporal impact of environmental control equipment actions on the physical fields (such as temperature, humidity, and flow fields) inside the greenhouse. The specific implementation of this model system in the controller is represented by a library of equipment cooperative action primitives and a spatial effect map corresponding to each primitive. The spatial effect map is established through a prior system identification method and can be calibrated online during system operation. The process of generating a cooperative action sequence is essentially based on this equipment disturbance propagation model (specifically represented by the primitive library and spatial effect map), using matching and optimization to calculate the optimal equipment drive commands required to achieve the desired environmental changes.

[0105] Specifically, the equipment cooperative action primitives are the "excitation sources" or "input units" of the equipment disturbance propagation model. Each primitive represents a specific, executable combination of equipment actions. The spatial effect map is the "response database" or "output prediction" of the equipment disturbance propagation model. Each map precisely describes the model's output when the corresponding primitive is used as input—that is, the changes in environmental parameters caused throughout the entire space. The correspondence between the primitive library and the map set together constitutes a complete, query-based "equipment disturbance propagation model." The controller does not need to directly solve complex physical partial differential equations, but instead predicts the effects of equipment actions and combines solutions by querying this "primitive-map" database.

[0106] In this invention, the core mathematical meaning of solving a field control inversion problem is: given a desired environmental field change (i.e., spatial desired field increment) defined in the spatial and time domains, find an optimal equipment control sequence from a complex solution space composed of equipment physical actions, such that the error between the synthesized environmental field and the desired field generated by the sequence is minimized.

[0107] Traditional inversion methods may directly solve the inverse problem of partial differential equation systems, which is computationally complex and difficult to guarantee real-time performance. This invention creatively transforms this problem into a two-stage optimization problem for efficient solution:

[0108] Discrete inversion in the primitive space (corresponding to the screening step): The infinite-dimensional solution space of the device control sequence is reduced to a subset of the solution space consisting of a finite number of functionally defined device cooperative action primitives. By matching the spatial effect map with the spatial expectation field increment, a high-quality and feasible range of initial solutions is quickly identified. This step is equivalent to solving the existence and general form of the inverse problem.

[0109] Continuous inversion in parameter space (corresponding to optimization and adjustment steps): Based on the initial solution of the selected high-quality primitives, fine-tuning is performed in its continuous parameter space (opening degree, power, duration) to achieve the transition from coarse matching to fine matching, and finally the optimal control command is accurately solved.

[0110] Therefore, the screening and optimization adjustment steps together constitute the unique and efficient solution path designed by this invention for solving the field control inversion problem.

[0111] A set of equipment collaborative action primitives are predefined. Each primitive represents a collaborative action mode of a set of environmental control equipment under a specific time sequence. Through system identification methods, a spatial effect map is established for each primitive. The spatial effect map is used to describe the distribution of changes in environmental parameters in the three-dimensional spatial grid of the tobacco planting area over a future period after the execution of this primitive.

[0112] Among them, the equipment coordinated action primitives are predefined combinations of equipment coordinated actions. For example, the top cooling primitive: the action sequence is "open the skylight to 45% opening, delay for 90 seconds, start the two top circulating fans to 70% power, and run continuously for 8 minutes". The horizontal uniformity primitive: the action sequence is "simultaneously start the horizontal circulating fans on the east and west sides, set the power of the east fan to 60%, and the power of the west fan to 40%, and run continuously for 10 minutes".

[0113] For system identification methods, subspace identification methods can be used to build the model. Experimental procedure: During the system initialization phase, each device's collaborative action primitive is executed automatically in sequence, while data from all environmental sensors on the greenhouse's three-dimensional grid points are collected at a frequency of once per second. Based on these input-output data, a multi-input multi-output state-space model is fitted to each primitive using a subspace identification algorithm (such as N4SID). This model is the mathematical core of the spatial effect spectrum of that primitive.

[0114] In each control cycle, the change in the target trajectory in a short time domain is compared with the environmental parameter values ​​of each spatial grid point predicted based on the current sensor data. The expected change in environmental parameters that needs to be achieved at the main spatial grid points of the crop canopy in the short time domain is calculated and denoted as the spatial expected field increment.

[0115] In each control cycle (e.g., every 5 minutes), the system looks ahead to a short time domain (e.g., the next 30 minutes). It calculates the change in the target trajectory during that time period and compares it with the environmental parameter values ​​predicted for each grid point based on the current conditions to determine how much environmental quantity (e.g., a 2°C increase in temperature) needs to be added or reduced in the crop canopy area in the future. The spatial distribution of this required change is the spatial expected field increment.

[0116] From the cooperative action primitives of various devices, at least one candidate cooperative action primitive whose spatial effect map and spatial expected field increment best match the spatial distribution pattern in terms of spatial distribution is selected.

[0117] The spatial distribution pattern of the expected field increment (e.g., a significant warming in the central canopy and a slight warming at the edges) is matched with the spatial effect map of each primitive in the primitive library. This is achieved by calculating the spatial distribution correlation coefficient, and finally selecting several primitives whose effect patterns are most similar to the expected change patterns as candidates.

[0118] The motion parameters of the selected candidate cooperative action primitives are optimized and adjusted. The motion parameters include equipment opening degree, power and duration. The goal of the adjustment is to minimize the overall error between the synthesized spatial effect and the expected spatial field increment after the primitive is adjusted, while the total energy consumption of the equipment and the frequency of action are used as optimization penalty terms.

[0119] For the selected candidate primitives, their action parameters are fine-tuned (e.g., adjusting the fan's "70% power" to "65% power," or "operation time 8 minutes" to "7.5 minutes"). The goal of these adjustments is to minimize the overall error between the combined effect of these primitives and the "expected spatial field increment." Simultaneously, the total power consumption of the equipment and the frequency of equipment start-ups and shutdowns within a short period are considered as penalty factors in the optimization objective to ensure that the solution is accurate while also being economical and friendly to equipment lifespan.

[0120] The optimal collaborative action primitives and their parameters obtained through optimization and adjustment are converted into control commands for environmental control equipment and sent to the corresponding equipment for execution.

[0121] The spatial effect map can be mathematically represented as a linear time-invariant system. For a given device cooperative action primitive, the change in environmental parameters ΔE(x,y,z,τ) caused by it at a specific spatial point (x,y,z) can be modeled as: ΔE(x,y,z,τ)=H(x,y,z,τ)*U(τ);

[0122] ΔE(x,y,z,τ) represents the change in environmental parameters (e.g., temperature change) at a spatial point (x,y,z) and a future time τ. H(x,y,z,τ) represents the spatial impulse response function of this primitive at point (x,y,z). It encapsulates the spatiotemporal characteristics of disturbance propagation. Its physical meaning is: the change in environmental parameters at a spatial point (x,y,z) and time τ caused by a unit pulse of the device input signal U(τ). Therefore, the dimension of H is the environmental parameter change dimension / time (e.g., for temperature control, its unit can be °C / s). The input signal U(τ) is the normalized device action intensity, ranging from 0 to 1. U(τ) represents the input signal of the device's cooperative action primitive (e.g., the change in power percentage over time). * indicates a convolution operation.

[0123] This model quantitatively describes the dynamic relationship between equipment actions and changes in the spatial environment. It forms the basis for solving field control inversion problems, enabling the system to predict the spatiotemporal consequences of actions.

[0124] Regarding the best match in spatial distribution patterns, the quantitative standard for matching is the calculation of spatial cross-correlation coefficients. The system calculates the Pearson correlation coefficient between the spatial effect map of each primitive and the spatial expected field increment in two dimensions: one is the spatial distribution pattern, and the other is the trend over time. The primitive with the highest comprehensive correlation coefficient is selected as a candidate.

[0125] Based on the embodiments provided in this application, a spatial effect map is established for each primitive to describe the disturbance characteristics of equipment actions on three-dimensional spatial environmental parameters. Then, the primitive parameters are screened and optimized by calculating the spatial expected field increment. The equipment cooperative action primitive abstracts complex equipment operations into standardized patterns, simplifies the logic of multi-equipment cooperative control, and improves the generation efficiency of action sequences. The spatial effect map enables the system to predict the spatiotemporal distribution of equipment actions within the planting area, achieving precise spatial control of environmental parameters and avoiding local over-adjustment or under-adjustment. The optimization process considers equipment energy consumption and action frequency, reducing operating costs. At the same time, the matching degree between the action sequence and the environmental target is ensured by solving the inversion problem.

[0126] Furthermore, the method also includes feedforward disturbance cancellation control based on weather forecasts, including:

[0127] It receives short-term weather forecast data and converts the predicted external environmental disturbances that will occur in the future into a predicted disturbance field with a specific incoming direction, spatial distribution pattern and intensity variation waveform;

[0128] The construction of the predicted disturbance field includes the process of converting short-term weather forecast data into a predicted disturbance field with spatiotemporal characteristics. This is a quantification process from point to surface, from scalar to field quantity. Specifically:

[0129] Direction of entry: The main direction of entry for the disturbance is determined based on the wind direction forecast. For example, if the forecast is a southeasterly wind, the disturbance field is set to propagate from the southeast side of the greenhouse to the northwest side.

[0130] Spatial distribution pattern: Based on the greenhouse structure and the type of disturbance, different spatial distribution patterns are preset. For example, for instantaneous strong light, it is modeled as a two-dimensional planar field that propagates uniformly downward from the top of the greenhouse; for gusts, it is modeled as a gradient field that penetrates inward from the side window on the windward side and decays with distance.

[0131] Intensity variation waveform: Based on the data of the predicted intensity changing over time, the "waveform" of the disturbance is generated. For example, if the forecast shows that a thick cloud will block the sun in 10 minutes and last for 20 minutes, the light disturbance is quantified as a trapezoidal wave in which the intensity linearly decreases from the current value to a low value after 10 minutes, remains there for 20 minutes, and then linearly recovers.

[0132] Through the above implementation method, an abstract weather forecast data is concretized into a physical field model with a clear propagation path, distribution range and intensity variation within the greenhouse physical space.

[0133] With the goal of offsetting the effect of the predicted disturbance field in the crop canopy area, a pre-action sequence is formed by backward searching one or more equipment cooperative action primitives from each equipment cooperative action primitive; the design goal of the pre-action sequence is to minimize the net environmental change after the regulation field generated by it is superimposed with the predicted disturbance field in the main spatial area of ​​the crop canopy.

[0134] It should be noted that the reverse search process is structured as an optimization problem with the goal of offsetting effects. The objective of this optimization problem is to find one or more primitives from the device cooperative action primitive library such that the superposition effect of the resulting control field and the predicted disturbance field in the crop canopy region is minimized. Since the primitive library is discrete, directly solving the combinatorial optimization problem would be computationally very expensive.

[0135] Therefore, this embodiment employs a two-stage greedy search algorithm. The first stage performs rapid filtering, calculating the maximum offsetting potential of each primitive acting alone, and selecting the top N primitives with the highest potential. The second stage performs combination evaluation, permuting and combining these N primitives, evaluating the joint offsetting effect of different combinations, and finally selecting the primitive combination with the optimal effect and the fewest total device actions to form the "pre-action sequence." This method significantly improves search efficiency while ensuring the offsetting effect.

[0136] The pre-action sequence and the coordinated action sequence are weighted and fused; the weight coefficient of the feedforward control is dynamically calculated based on the average deviation between the meteorological forecast data and the actual monitoring data in the past preset period. The smaller the deviation, the higher the weight.

[0137] Based on the embodiments provided in this application, meteorological forecast data is received, a predicted interference field is generated, and then the cooperative action primitive of the equipment is searched with the goal of counteracting the interference. The pre-action sequence and the feedback control sequence are weighted and fused. Feedforward control can proactively respond to impending changes in the external environment (such as temperature fluctuations or wind speed changes), reducing the lag of feedback control and improving the stability of environmental parameters. The weighted fusion mechanism dynamically adjusts the feedforward weights according to the forecast accuracy, avoiding control inaccuracies caused by forecast errors. Through the superposition and optimization of the interference field and the control field, more uniform environmental control is achieved in the crop canopy area, improving the overall control quality.

[0138] Furthermore, the feedforward disturbance cancellation control also includes the following safety strategies:

[0139] Set a safety boundary based on historical operating data for the single action range of each environmental control device to ensure that the action does not cause drastic fluctuations in environmental parameters;

[0140] The safety boundary for a single operation of the equipment is set based on a combination of two layers of information. The first layer is the physical limits of the equipment, derived from the equipment manual, such as the maximum power of the heater and the maximum opening of the skylight. The second layer is the physiological safety range of the crop, derived through analysis of historical operating data. This involves statistically analyzing the operating ranges that do not cause drastic fluctuations in crop stem flow rate or overshooting of environmental parameters, and using their upper and lower limits as soft constraints for the safety boundary. The final safety boundary is the more stringent value between the physical limits and the physiological safety range.

[0141] Establish a device action logic conflict detection mechanism. When the calculated action sequence includes mutually exclusive actions, automatically block or replace low-priority actions based on the priority adjusted by environmental parameters.

[0142] Among these, conflicts in equipment operation logic mainly include functional mutual exclusion and resource mutual exclusion. Functional mutual exclusion refers to actions with opposite physical effects, such as starting a heater and opening a wet curtain for cooling, or releasing carbon dioxide and opening a skylight for ventilation. Resource mutual exclusion refers to situations where, due to limitations in power capacity or water system, certain high-power or high-water-consumption equipment cannot operate simultaneously.

[0143] The priority rules for environmental parameter regulation follow these principles: first, ensure crop survival; second, promote crop growth; and finally, optimize energy consumption. Specifically, relieving stress (such as high or low temperatures) has the highest priority; next are parameters affecting core metabolic processes (such as photosynthesis), mainly light and carbon dioxide; then come parameters affecting microenvironment comfort, such as humidity and airflow. When conflicts occur, the system automatically shields or delays low-priority actions to ensure the achievement of high-priority goals.

[0144] Set confidence conditions for feedforward control to take effect. When the predicted interference intensity is lower than the sensor noise level or key meteorological data is missing, feedforward control will be automatically suspended, and the system will be maintained by feedback control alone.

[0145] Based on the embodiments provided in this application, a safety boundary limits the amplitude of a single device action, a conflict detection mechanism identifies and automatically handles mutually exclusive actions, and a confidence condition suspends feedforward when the forecast is unreliable. This design achieves the following technical effects: the safety boundary prevents equipment damage or sudden environmental changes that may be caused by excessive device actions; the conflict detection mechanism eliminates contradictions between device actions through priority logic, avoiding system oscillations caused by control command conflicts; and the confidence condition dynamically manages feedforward control based on the deviation between sensor data and forecasts, automatically downgrading to feedback control when the data is unreliable, thus improving reliability in uncertain environments.

[0146] Furthermore, the regulatory effect is evaluated based on the temporal changes in both environmental parameters and crop stem flow data. Based on the evaluation results, key physiological parameters in the crop physiological mechanism model and spatial transfer characteristics in the equipment disturbance propagation model are dynamically calibrated, including:

[0147] Within each evaluation period, three evaluation indicators are calculated: environmental parameter tracking error, metabolic flow realization degree, and stem flow stability maintenance degree.

[0148] Among them, the environmental parameter tracking error is obtained by calculating the root mean square error between the actual monitored environmental parameter values ​​and the corresponding target trajectory values ​​at each spatial location within the sliding time window; the metabolic flow realization degree is obtained by inputting the actual environmental parameter data into a simplified metabolic flow model for forward simulation to obtain the metabolic flow prediction value, and then calculating the correlation coefficient between the prediction value and the corresponding target concentration change trajectory value; the stem flow stability maintenance degree is obtained by calculating the approximate entropy of the actual collected stem flow rate data within the time window and comparing it with the preset expected stability range.

[0149] Among them, the environmental parameter tracking error is calculated as the root mean square value of the difference between the actual monitored environmental parameter values ​​(such as temperature and humidity) at each spatial location and the corresponding values ​​calculated from the target trajectory within a sliding time window (such as the past hour). It directly reflects the accuracy of environmental control.

[0150] Metabolic flux achievement: This indicator uses correlation coefficients instead of tracking error, based on the core idea of ​​this invention. Metabolic flux is the ultimate intrinsic goal pursued by this invention, while environmental parameters are the means. The correlation coefficient can measure the consistency between the trend of metabolic flux changes driven by the actual environment and the trajectory of target concentration changes. A high positive correlation coefficient means that even if the absolute value deviates due to uncontrollable factors, the system correctly guides the metabolic flux to be high when needed and low when needed, which is more in line with the essence of physiological regulation. The calculation method is to input actual environmental data into a simplified metabolic flux model to obtain a predicted metabolic flux sequence, and then calculate the Pearson correlation coefficient between this sequence and the target concentration change trajectory sequence.

[0151] Stem flow stability maintenance: This index is assessed by calculating the approximate entropy of actual stem flow rate data collected within the same time window and comparing it with a preset expected stability range (e.g., 0.5 to 1.0). Its purpose is to ensure that regulatory behavior maintains the stability of the crop water transport system and avoids physiological instability.

[0152] Based on the embodiments provided in this application, the root mean square error, correlation coefficient, and approximate entropy within the sliding time window are calculated, reflecting the accuracy of environmental control, the degree of achievement of physiological goals, and the stability of plant physiological state, respectively. Multi-index evaluation covers the entire chain of regulatory effects from environment to physiology, helping the system identify weak links; environmental parameter tracking error directly measures the accuracy of environmental control, providing a basis for equipment adjustment; metabolic flow achievement degree verifies the achievement of physiological goals through model simulation, ensuring the effectiveness of the regulatory strategy; stem flow stability maintenance degree quantifies plant water transport health using approximate entropy, preventing the risk of physiological stress.

[0153] Furthermore, the method also includes:

[0154] Based on the evaluation indicators, parameter sensitivity analysis was performed on the crop physiological mechanism model and the equipment disturbance propagation model. The top K physiological parameters that have the greatest impact on the realization of metabolic flux were identified as the set of physiological parameters to be calibrated, and the top M spatial transfer parameters that have the greatest impact on the tracking error of environmental parameters were identified as the set of equipment parameters to be calibrated.

[0155] In some embodiments, the Morris screening method is used for parameter sensitivity analysis. This method is a highly efficient global sensitivity analysis approach. By systematically perturbing multiple parameters within a reasonable space and observing changes in model outputs (such as metabolic flux realization and environmental parameter tracking errors), it quickly identifies the key parameters that have the greatest impact on the output. Compared to the more accurate but computationally expensive Sobol method, the Morris method is more suitable for the computational efficiency requirements of this online calibration scenario while ensuring the identification of the main sensitive parameters.

[0156] Using a recursive least squares method with a forgetting factor, with environmental parameter tracking error and metabolic flux realization as joint optimization objectives, the parameters in the set of physiological parameters to be calibrated and the set of equipment parameters to be calibrated are estimated and updated online. The updated parameter values ​​are then written into the crop physiological mechanism model and the equipment disturbance propagation model, respectively, to complete the model calibration.

[0157] Among them, recursive least squares with a forgetting factor was used for online estimation of key model parameters selected from the pool. The application process is as follows: First, the model to be calibrated is linearized around the operating point, transforming its parameter estimation problem into a linear regression problem. Then, the recursive least squares method iteratively updates the estimated values ​​of the model parameters using new monitoring data (such as environmental parameters and stem flow data). The "forgetting factor" assigns higher weight to new data, enabling the algorithm to track the dynamic characteristics of the system caused by slow changes in crop growth or the environment. The initial value of the forgetting factor is set to 0.99, and fine-tuned at different growth stages: during rapid crop changes, such as the transplanting and seedling establishment period, a smaller forgetting factor (e.g., 0.95) is used to quickly track changes; during stable stages, such as the maturity and harvest period, a larger forgetting factor (e.g., 0.995) is used to maintain the stability of parameter estimation and prevent excessive fluctuations.

[0158] The dynamic calibration process also includes the following adaptive mechanisms:

[0159] Set the start conditions for model calibration. When the moving average of the tracking error of environmental parameters exceeds the first threshold for three consecutive evaluation periods, or the metabolic flow realization is lower than the second threshold for two consecutive evaluation periods, or the stem flow stability maintenance exceeds the expected stability range for two consecutive evaluation periods, the model calibration process is triggered.

[0160] Different calibration intensity parameters are set according to different growth stages. During the transplanting and seedling establishment period, the forgetting factor of the recursive least squares method is increased to enable the model parameters to adapt to changes in crop status more quickly. During the maturity and harvest period, the forgetting factor is reduced to maintain the stability of the model parameters.

[0161] Set reasonable physical range constraints for each parameter to be calibrated to ensure that the calibrated parameter values ​​conform to the physiological characteristics of crops and the physical characteristics of equipment. When the calibration results exceed the reasonable range, use boundary values ​​to replace them and record the abnormal state.

[0162] Based on the embodiments provided in this application, a set of key parameters that have the greatest impact on evaluation indicators is identified, and then recursive estimation is performed with tracking error and achievement rate as the objectives. This design achieves the following technical benefits: dynamic calibration enables the model to adapt to physiological changes and environmental drift during tobacco growth, maintaining the model's predictive accuracy; parameter sensitivity analysis focuses on key parameters, reducing the computational burden of calibration and improving efficiency; the recursive least squares method with a forgetting factor ensures the real-time nature of parameter estimation and a reasonable balance with historical data, allowing the model to be continuously optimized to adapt to actual regulatory needs.

[0163] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described IoT-driven multi-parameter intelligent control method for tobacco growth environment is also provided. This electronic device may be... Figure 3The terminal device or server shown. This embodiment uses this electronic device as an example of a server. Figure 3 As shown, the electronic device includes a memory 402, a processor 404, and a transmission device 406. The memory 402 stores a computer program, and the processor 404 is configured to execute the steps in any of the above method embodiments through the computer program.

[0164] Optionally, in this embodiment, the aforementioned electronic device may be located in at least one of a plurality of network devices in a computer network.

[0165] Optionally, the transmission device 406 is used to receive or send data via a network. Specific examples of the network described above may include wired and wireless networks. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can be connected to other network devices and a router via a network cable to communicate with the Internet or a local area network. In another example, the transmission device 406 is a Radio Frequency (RF) module used to communicate with the Internet wirelessly. Furthermore, the electronic device also includes a display 408 and a connection bus 410, which connects the various module components within the electronic device.

[0166] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for intelligent multi-parameter control of tobacco growth environment driven by the Internet of Things, characterized in that, include: Based on real-time collected environmental parameters and crop stem flow data of tobacco planting areas, combined with the current growth stage and quality targets of tobacco, the target concentration change trajectory of key metabolites characterizing the internal physiological activities of tobacco is dynamically generated. Using the target concentration change trajectory as the optimization objective, the crop physiological mechanism model describing the relationship between the environment and physiological processes is used for inverse solution to calculate the target trajectory of the environmental parameter setpoints required to achieve the optimization objective over time. Based on the target trajectory and combined with a pre-established equipment disturbance propagation model, a coordinated action sequence of environmental control equipment is generated by solving a field control inversion problem; wherein, the equipment disturbance propagation model defines the spatiotemporal disturbance characteristics of each equipment action on the environmental state at different locations within the planting space; Execute the coordinated action sequence of the device to control the corresponding environmental control equipment to operate; The regulatory effect is evaluated based on the temporal changes of environmental parameters and crop stem flow data, and the key physiological parameters in the crop physiological mechanism model and the spatial transmission characteristics in the equipment disturbance propagation model are dynamically calibrated based on the evaluation results. The method of dynamically generating target concentration change trajectories of key metabolites characterizing internal physiological activities of tobacco, based on real-time collected environmental parameters and crop stem flow data of the tobacco planting area, combined with the current growth stage and quality targets of the tobacco, includes: The light intensity and temperature data from the real-time collected environmental parameters are input into the preset primary productivity model of photosynthesis to output the basic synthesis rate curve of carbohydrates over time. Using the current growth stage of tobacco as an index, a preset metabolic allocation strategy database is queried to obtain allocation parameters; the basic synthesis rate curve is decomposed into a first synthesis rate sub-trajectory for growth and a second synthesis rate sub-trajectory for accumulation and transport using the allocation parameters. The average daily demand rate of nicotine precursors is calculated by combining the nicotine content set in the quality target with the number of days from the current date to the planned harvest date, in order to generate a translocation rate curve. Based on the basic synthesis rate curve, the first synthesis rate sub-trajectory, the second synthesis rate sub-trajectory, and the transport rate curve, an initial set of metabolic flux trajectories is determined. The real-time collected crop stem flow data is compared with a preset normal range; when the data is abnormal, the attenuation coefficient is obtained according to the preset stress response coefficient table to globally adjust all curves in the initial metabolic flow trajectory set and output the target concentration change trajectory.

2. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 1, characterized in that, The method further includes: When the light intensity data is below the light compensation point threshold and the stem flow rate data shows a downward trend for three consecutive monitoring cycles, it is determined that the state of weak light stress has been entered. Under low light stress, the peak value of the second synthesis rate sub-trajectory is reduced by a first preset ratio, while the duration of its synthesis rate being greater than zero is extended by a fixed period of time. When the canopy temperature data exceeds the high temperature threshold and the fluctuation range of the stem flow rate data exceeds the normal fluctuation range in two consecutive monitoring cycles, it is determined that the state of high temperature stress has been entered. Under high-temperature stress, the peak occurrence period of the transport rate curve is adjusted from daytime to the lowest temperature period in the early morning of the next day, and its peak amplitude is adjusted according to a second preset ratio.

3. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 1, characterized in that, The optimization objective is to use the target concentration change trajectory as the optimization target, and to perform inverse solving using a crop physiological mechanism model describing the relationship between the environment and physiological processes, to calculate the target trajectory of the environmental parameter setpoints required to achieve the optimization objective over time, including: Construct a dynamic physiological cause-and-effect graph, whose nodes include environmental parameters and metabolic flow rates; Based on the current reproductive stage and real-time stalk flow stability, the causal weights of the edges in the physiological causal graph are dynamically allocated; a reverse graph search is performed starting from the target node of the target concentration change trajectory to identify the key environmental parameters and their sensitive periods that have the highest impact on achieving the target, thereby focusing the full parameter optimization problem on the key subset; The physiological state of the crop is determined based on real-time collected environmental parameters and crop stem flow data; based on different states, corresponding optimization algorithms are called from the strategy library, including global exploration strategy, conservative repair strategy and economic energy-saving strategy, to generate multiple different candidate environmental parameter trajectories in parallel. The candidate environmental parameter trajectories are validated and screened using a simplified metabolic flow model, and an initial solution set is output. For each trajectory in the initial solution set, an adjoint method channel is activated for adjustment. The adjoint method channel prioritizes gradient calculation and optimization during a defined sensitive period through an attention mechanism. At the same time, an imitation learning channel is activated to perform pattern matching and segment fusion between the trajectory and historical successful trajectories. Based on the real-time stem flow stability index, the solution output by the accompanying method channel and the solution output by the imitation learning channel are weighted and fused to generate a preliminary target trajectory. The preliminary target trajectory is input into the stem flow prediction model to predict the stem flow sequence generated after executing the trajectory; The approximate entropy of the predicted stem flow sequence is calculated. If the approximate entropy is lower than the safe entropy threshold, it is determined that the trajectory poses a risk of physiological instability. When the risk of instability is detected, the causal weights of the nodes related to water transport stability in the dynamic causal graph are adjusted, and a new round of reverse solution process is triggered. If the predicted approximate entropy is higher than or equal to the safety entropy threshold, then the current preliminary target trajectory is output as the target trajectory.

4. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 3, characterized in that, In the process of generating the candidate environmental parameter trajectories and the preliminary target trajectories, photosynthetic rhythm illumination constraints and carbon balance temperature constraints are introduced; wherein... The photosynthetic rhythm light constraint conditions include: when generating the light intensity parameter trajectory, the constraint curve of its change needs to simulate the diurnal variation pattern of natural light intensity, that is, the light intensity rises non-linearly from zero or the base value to the daytime peak value in a preset first period after sunrise, and falls non-linearly from the daytime peak value to the nighttime base value in a preset second period before sunset. The carbon balance temperature constraint conditions include: when generating the nighttime temperature parameter trajectory, constraining its value to be calculated based on the virtual amount of cumulative photosynthetic products during the day, so that the predicted nighttime respiration consumption of crops at this temperature and the virtual amount of cumulative photosynthetic products during the day are maintained within a ratio range dynamically set according to the quality target.

5. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 3, characterized in that, The step involves generating a coordinated action sequence of environmental control equipment by solving a field control inversion problem based on the target trajectory and a pre-established equipment disturbance propagation model, including: A set of equipment collaborative action primitives are predefined, each primitive representing a collaborative action mode of a set of environmental control equipment under a specific time sequence; a spatial effect map is established for each primitive through a system identification method, the spatial effect map being used to describe the distribution of changes in environmental parameters over a future period of time on a three-dimensional spatial grid of the tobacco planting area after the execution of this primitive; In each control cycle, the change of the target trajectory in a short time domain in the future is compared with the environmental parameter values ​​of each spatial grid point predicted based on the current sensor data. The expected change of environmental parameters that needs to be achieved at the main spatial grid points of the crop canopy in the short time domain is calculated and denoted as the spatial expected field increment. From the cooperative action primitives of various devices, at least one candidate cooperative action primitive whose spatial effect map best matches the spatial distribution pattern of the spatial expected field increment is selected. The motion parameters of the selected candidate cooperative action primitives are optimized and adjusted; wherein, the motion parameters include equipment opening degree, power and duration; the goal of the adjustment is to minimize the overall error between the synthesized spatial effect of the primitives after adjustment and the expected spatial field increment, while taking the total energy consumption of the equipment and the frequency of action as optimization penalty terms; The optimal collaborative action primitives and their parameters obtained through optimization and adjustment are converted into control commands for environmental control equipment and sent to the corresponding equipment for execution.

6. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 5, characterized in that, The method also includes feedforward disturbance cancellation control based on weather forecasts, including: It receives short-term weather forecast data and converts the predicted external environmental disturbances that will occur in the future into a predicted disturbance field with a specific incoming direction, spatial distribution pattern and intensity variation waveform. With the goal of counteracting the effect of the predicted interference field in the crop canopy area, a pre-action sequence is formed by backward searching one or more device cooperative action primitives from each device cooperative action primitive; the design goal of the pre-action sequence is to minimize the net environmental change after the regulation field it generates is superimposed on the predicted interference field in the main spatial area of ​​the crop canopy. The pre-action sequence and the coordinated action sequence are weighted and fused; wherein, the weight coefficient of the feedforward control is dynamically calculated based on the average deviation between the meteorological forecast data and the actual monitoring data in the past preset period. The smaller the deviation, the higher the weight.

7. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 6, characterized in that, The feedforward interference cancellation control also includes the following security strategies: Set a safety boundary based on historical operating data for the single action range of each environmental control device. Establish a device action logic conflict detection mechanism. When the calculated action sequence includes mutually exclusive actions, automatically block or replace low-priority actions based on the priority adjusted by environmental parameters. Set confidence conditions for feedforward control to take effect. When the predicted interference intensity is lower than the sensor noise level or key meteorological data is missing, feedforward control will be automatically suspended, and the system will be maintained by feedback control alone.

8. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 3, characterized in that, The evaluation of the regulatory effect based on the temporal changes of both environmental parameters and crop stem flow data, and the dynamic calibration of key physiological parameters in the crop physiological mechanism model and spatial transfer characteristics in the equipment disturbance propagation model based on the evaluation results, includes: Within each evaluation period, three evaluation indicators are calculated: environmental parameter tracking error, metabolic flow realization degree, and stem flow stability maintenance degree. The environmental parameter tracking error is obtained by calculating the root mean square error between the actual monitored environmental parameter values ​​at each spatial location within the sliding time window and the corresponding target trajectory values; the metabolic flow realization degree is obtained by inputting the actual environmental parameter data into the simplified metabolic flow model for forward simulation to obtain the metabolic flow prediction value, and then calculating the correlation coefficient between the prediction value and the corresponding target concentration change trajectory value; the stem flow stability maintenance degree is obtained by calculating the approximate entropy of the actual collected stem flow rate data within the time window and comparing it with the preset expected stability range.

9. The IoT-driven intelligent multi-parameter control method for tobacco growth environment according to claim 8, characterized in that, The method further includes: Based on the evaluation indicators, parameter sensitivity analysis was performed on the crop physiological mechanism model and the equipment disturbance propagation model to identify the top K physiological parameters that have the greatest impact on the realization of metabolic flux as a set of physiological parameters to be calibrated, and the top M spatial transfer parameters that have the greatest impact on the tracking error of environmental parameters as a set of equipment parameters to be calibrated. Using a recursive least squares method with a forgetting factor, and taking environmental parameter tracking error and metabolic flux realization as joint optimization objectives, the parameters in the set of physiological parameters to be calibrated and the set of equipment parameters to be calibrated are estimated and updated online. The updated parameter values ​​are then written into the crop physiological mechanism model and the equipment perturbation propagation model, respectively, to complete the model calibration.

Images