Intelligent agricultural monitoring system and method based on cloud platform

Abstract

The invention relates to the technical field of intelligent agriculture and plant factory environment regulation and control, in particular to an intelligent agriculture monitoring system and method based on a cloud platform, and the method comprises the steps: collecting multi-dimensional state data and hidden environment input parameters of a crop group; calculating to obtain a corrected pulse synchronization rate through a pulse synchronization rate correction model; calculating a group suffocation risk index through a group suffocation risk index calculation model; determining a risk level index; the method comprises the following steps: identifying a non-linear damping coefficient, and carrying out standardization processing on the non-linear damping coefficient and a preset critical damping coefficient value to obtain a system toughness factor; generating a control intensity correction factor based on the risk level index and the system toughness factor; according to the risk grade index and the control intensity correction factor, executing a grading avoidance strategy, and outputting a regulation and control signal of the environment regulation and control system; according to the invention, accurate intervention is realized, production safety and resource efficiency are effectively balanced, and insufficiency or overreaction of a traditional control mode is avoided.

Description

Technical Field

[0001] This invention relates to the field of smart agriculture and plant factory environmental control technology, specifically to a cloud-based smart agriculture monitoring system and method. Background Technology

[0002] In the environmental monitoring of fully artificial light plant factories, existing technologies mainly rely on monitoring and threshold alarms for macroscopic environmental parameters such as temperature and humidity. This approach ignores the implicit environmental harmonics generated by equipment operation, such as the pulse width modulation spectrum of the growth light source and the infrasound spectrum of the ventilation system. These harmonics may resonate with the physiological rhythms of crops, inducing microenvironmental boundary layer locking and ultimately leading to population suffocation disaster. In addition, traditional methods lack the assessment of the system's own dynamic recovery ability, i.e., system resilience, making it difficult to provide effective early warning and intervention before disaster occurs. Therefore, how to integrate implicit environmental stress with system resilience assessment to achieve accurate prediction and proactive avoidance of population suffocation risk is a technical problem that urgently needs to be solved. Summary of the Invention

[0003] The purpose of this invention is to provide a cloud-based smart agriculture monitoring system and method. By integrating latent environmental harmonic analysis and system resilience assessment, it aims to solve the technical problem of existing technologies that rely solely on macroscopic environmental parameters and cannot accurately predict and intervene in the risk of population suffocation caused by micro-environment boundary layer locking. This allows for accurate prediction and proactive avoidance of this risk. Specifically, the technical solution of this invention is as follows: A cloud-based smart agriculture monitoring method includes: Collect multi-dimensional state data and implicit environmental input parameters of crop populations; Based on the implicit environmental input parameters, the corrected pulse synchronization rate is calculated using the pulse synchronization rate correction model. By combining the corrected pulse synchronization rate with multi-dimensional state data, the group asphyxiation risk index is calculated using the group asphyxiation risk index calculation model. Risk level indicators are determined based on the mass asphyxiation risk index and preset coordination limits; By applying standardized small perturbations to the environmental system and monitoring the response curves of the system parameters, the nonlinear damping coefficient is identified, and then standardized with the preset critical damping coefficient value to obtain the system toughness factor. Based on risk level indicators and system resilience factors, a control strength correction factor is generated. Based on the risk level indicators and control intensity correction factors, a graded avoidance strategy is implemented, and control signals of the environmental control system are output.

[0004] Optional, multi-dimensional state data, including: Microscopic pulse synchronization rate obtained using a blade porosity sensor array; Mesoscopic canopy turbulent kinetic energy spectrum entropy calculated using high-frequency three-dimensional ultrasonic anemometer array data; Macroscopic nonlinear damping coefficients obtained through system identification techniques; The Shannon entropy, a chemical fingerprint of root exudates quantified in hydroponic nutrient solution, was analyzed using liquid chromatography-mass spectrometry.

[0005] Optional, implicit environment input parameters include: Pulse width modulation spectrum of the growth light source; The infrasound spectrum of the ventilation system.

[0006] Optional, risk level indicators may be determined, including: The mass asphyxiation risk index was normalized to obtain a risk level index. Based on the numerical range of the risk level indicators, the system status is assessed as safe, Level 1 warning, and Level 2 warning.

[0007] Optional, tiered avoidance strategies include: If the system is in a safe state, the current environmental control strategy will be maintained. If the system status reaches Level 1 warning, then the synchronization suppression mode is activated; If the system status is a Level 2 warning, the system will initiate a mode disruption and system reset procedure.

[0008] Optional synchronization suppression modes include: The dimming frequency of the growth light source is actively deviated, and the amount of active deviation is dynamically adjusted by a control intensity correction factor. Random disturbances are introduced into the control signal of the ventilation system, and the intensity of the random disturbances is dynamically adjusted by the control intensity correction factor.

[0009] Optional, mode destruction and system reset procedures include: Drive the ventilation system to execute a turbulence pulse program; After the turbulent pulse program ends, it recovers from the preset safe initial point to the target operating point along a preset nonlinear path.

[0010] A cloud-based smart agriculture monitoring system includes: The data acquisition module is used to collect multi-dimensional status data and implicit environmental input parameters of the crop population; The risk assessment module is used to calculate the mass asphyxiation risk index based on multi-dimensional state data and implicit environmental input parameters, and to determine the risk level indicators and the corresponding system state. The system resilience monitoring module is used to determine the system resilience factor by applying disturbances to the environmental system and monitoring the response; The closed-loop avoidance control module is used to generate control strength correction factors based on system state, risk level indicators and system resilience factors, execute hierarchical avoidance strategies, and output control signals of the environmental control system.

[0011] Optional risk assessment module, including: The stress analysis unit is used to calculate the corrected pulse synchronization rate based on the implicit environment input parameters and through the pulse synchronization rate correction model. The risk calculation unit is used to combine the corrected pulse synchronization rate with multi-dimensional state data to calculate the mass asphyxiation risk index through the mass asphyxiation risk index calculation model. The risk level determination unit is used to determine the risk level indicators and assess the system status based on the mass asphyxiation risk index and preset coordination limits.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention reveals and quantifies the latent environmental harmonic stress generated by equipment such as growth light sources and ventilation systems. By constructing a model to analyze its resonance effect with crop physiological rhythms, it improves the depth and accuracy of risk identification and overcomes the shortcomings of existing technologies that rely solely on macroscopic environmental parameters and cannot provide early warning of such latent risks. 2. This invention constructs a more comprehensive and three-dimensional system state profile by integrating multi-dimensional data from microscopic physiology, mesoscopic physics, macroscopic systems, and biochemical levels, calculates a comprehensive mass asphyxiation risk index, and profoundly reveals the intrinsic mechanism of the system's evolution from a steady state to a critical state. Compared with the threshold alarm method of traditional technology, the assessment results are more accurate and reliable. 3. This invention innovatively introduces system resilience assessment. By actively applying small disturbances and monitoring the response, it quantifies the dynamic ability of crops, i.e. environmental systems, to resist external shocks and recover on their own. This elevates the monitoring strategy from static risk assessment to dynamic system health management, enabling forward-looking prediction of system stability. 4. This invention constructs a hierarchical closed-loop avoidance control strategy based on risk level and system resilience, realizing the transformation from passive response to active avoidance; this strategy can dynamically adjust the intensity of control measures according to the urgency of the risk and the system's tolerance, achieving precise intervention, effectively balancing production safety and resource efficiency, and avoiding the shortcomings or overreaction of traditional control methods. Attached Figure Description

[0013] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a structural diagram of the system of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0015] Example 1: Please see Figure 1 A cloud-based smart agriculture monitoring method includes: Collect multi-dimensional state data and implicit environmental input parameters of crop populations; Based on the implicit environmental input parameters, the corrected pulse synchronization rate is calculated using the pulse synchronization rate correction model. By combining the corrected pulse synchronization rate with multi-dimensional state data, the group asphyxiation risk index is calculated using the group asphyxiation risk index calculation model. Risk level indicators are determined based on the mass asphyxiation risk index and preset coordination limits; By applying standardized small perturbations to the environmental system and monitoring the response curves of the system parameters, the nonlinear damping coefficient is identified, and then standardized with the preset critical damping coefficient value to obtain the system toughness factor. Based on risk level indicators and system resilience factors, a control strength correction factor is generated. Based on the risk level indicators and control intensity correction factors, a graded avoidance strategy is implemented, and control signals of the environmental control system are output.

[0016] This embodiment provides a cloud-based smart agriculture monitoring method. The method constructs a complete technical link from multi-dimensional data acquisition, risk modeling, resilience assessment to closed-loop avoidance, aiming to accurately warn and proactively avoid the possible mass suffocation disaster of crop groups in fully artificial light plant factories. This phenomenon is defined as a sharp decline in physiological exchange function caused by the locking of the microenvironment boundary layer. Collecting multi-dimensional state data and implicit environmental input parameters of crop populations is the data foundation for model-driven and risk assessment. In this embodiment, data collection is achieved through a multimodal sensor network deployed in the plant factory. Based on implicit environmental input parameters, the corrected pulse synchronization rate is calculated using a pulse synchronization rate correction model. This step aims to quantify the resonant driving effect of specific frequency harmonics generated by lighting and ventilation equipment on the physiological rhythms of crop populations. The implicit environmental input parameters refer to the pulse width modulation (PWM) spectrum of the growth light source and the infrasound spectrum of the ventilation system. Their function is to reveal potential environmental stress sources, which are obtained through real-time spectral analysis of equipment operating signals. To achieve this step, this embodiment constructs a pulse synchronization rate correction model, the purpose of which is to explicitly model the impact of the aforementioned implicit stresses on key physiological indicators. This model is specifically expressed by the following formula: ; in, The corrected pulse synchronization rate is a key physiological indicator that integrates the effects of latent environmental stress. It is dimensionless, calculated by this formula, and serves as the core input for the subsequent calculation of the group asphyxiation risk index. Baseline synchronization rate refers to the baseline synchronization level of stomatal conductance pulses in crop leaves measured under ideal conditions without significant light or acoustic harmonic interference. It is dimensionless and is obtained through pre-calibration in a control environment. Normalized amplitude of light source refers to the peak amplitude of a specific harmonic frequency band, such as 3-5kHz, in the PWM spectrum of the growth light source that has a resonance effect on crop physiology. It is dimensionless after normalization and is obtained through real-time spectrum analysis. Normalized amplitude of ventilation system refers to the peak amplitude of a specific resonant frequency band, such as 10-15Hz, in the infrasound spectrum generated by the ventilation system. It is dimensionless after normalization and is obtained through real-time spectrum analysis. and The coupling response coefficient characterizes the sensitivity of a specific crop variety to light and acoustic harmonic excitation; it is dimensionless. Its calibration process involves applying different frequencies to the crop while keeping other environmental parameters constant. The optical or acoustic excitation is used to measure and record a set of signals consisting of the excitation frequency and its corresponding steady-state pulse synchronization rate. The calibration dataset is constructed, and its values ​​are determined by fitting the dataset using the nonlinear least squares method. This model enables the dynamic correlation between equipment operating parameters and core physiological indicators, thereby quantifying the sources of system risk at a deeper level and providing more accurate and physically meaningful input for subsequent risk calculations. By combining the corrected pulse synchronization rate with multi-dimensional state data, the mass asphyxiation risk index is calculated using a mass asphyxiation risk index calculation model. The core of this step lies in integrating multi-scale information to construct a comprehensive indicator that fully reflects the system state and predicts disaster risks. Therefore, this embodiment constructs a mass asphyxiation risk index calculation model, using the corrected pulse synchronization rate... It is the main driving force of disasters, exhibiting a nonlinear acceleration effect; while other dimensions of state data reflect the system's buffering capacity, playing a role in suppressing risks. The model is specifically expressed by the following formula: ; in, The population asphyxiation risk index is a multi-factor fusion index for assessing the risk of sudden changes in system state. It is dimensionless and serves as the direct basis for risk assessment and graded control in this technical solution. The corrected pulse synchronization rate, whose physical meaning and source have been defined in the previous steps, is calculated by the pulse synchronization rate correction model; Synchronization rate critical threshold refers to the experimentally determined threshold at which a system undergoes a sudden phase transition. The critical point, dimensionless, is determined by conducting progressively increasing stress-induced experiments on specific crops. The canopy turbulent kinetic energy spectrum entropy reflects the efficiency and uniformity of gas exchange within the canopy. It is dimensionless and is derived from calculations based on data from a high-frequency three-dimensional ultrasonic anemometer array. Spectral entropy: The maximum health baseline value, indicating the crop under optimal growth conditions. The reference value is dimensionless and is obtained through historical data statistics or control experiments. It is typically... No more than If it appears In such a case, it might mean that the airflow within the canopy is too turbulent, which is also not conducive to efficient gas exchange. In this case, the model... A negative value indicates an increased risk. Shannon entropy, a chemical fingerprint of root exudates, reflects the health status and physiological diversity of the root system. It is dimensionless and is obtained by analyzing hydroponic nutrient solution using liquid chromatography-mass spectrometry. Shannon entropy reference baseline value, referring to the crop's health status. The reference value is dimensionless and is obtained through historical data statistics or control experiments; : Weight coefficients, which correspond to the contributions of the risk-driving term, turbulence buffer term, and root health term, respectively. They are dimensionless and are determined by optimizing the historical dataset containing mass asphyxiation event samples through multiple regression analysis. : Nonlinear amplification factor, used to characterize The accelerated amplification effect of risk after exceeding the critical value is usually greater than 1, dimensionless, and calibrated according to the sensitivity of different crop varieties to stress. This formula constructs a multi-factor coupling model that includes nonlinear amplification, positive suppression, and reverse buffering, which can more profoundly reveal the intrinsic mechanism of the system's evolution from a steady state to a critical state; Based on the mass asphyxiation risk index and preset harmonized limits, risk level indicators are determined; this step aims to address the continuously changing risk index. Transformed into discrete, decision-making-friendly risk levels; coordination limits, denoted as... This refers to the critical threshold at which an irreversible boundary layer-locked phase transition occurs in a system, determined based on extensive experimental data. Value, risk level indicator The calculation method is to Normalization is performed: ; This risk level indicator Transform the original risk index into an intuitive decision-making basis with 1 as the critical point; By applying a standardized small disturbance to the environmental system and monitoring the response curves of the system parameters, the nonlinear damping coefficient is identified and standardized with a preset critical damping coefficient value to obtain the system resilience factor. This step aims to assess the system's ability to resist disturbances and self-recover. In practice, the control system executes a standardized small disturbance, such as a brief ventilation pulse. By monitoring the response curves of parameters such as canopy humidity from deviation to recovery to steady state, and based on a second-order system response model, the nonlinear damping coefficient of the system is identified. For ease of application, this embodiment further defines the system resilience factor. The calculation formula is as follows: ; in, The system resilience factor is a standardized index that measures the dynamic recovery capability of a system. The boundary treatment of this formula ensures that its value range is between 0 and 1. The closer it is to 1, the better the resilience. It is dimensionless and serves as one of the core inputs for modifying control strength. The nonlinear damping coefficient at the current moment is obtained through perturbation response experiments. Critical damping coefficient value refers to the damping value when the system is completely unresponsive to external disturbances after entering a fully locked state. It is dimensionless and is calibrated through extreme condition tests. Introducing system resilience factor This allows the monitoring method of the present invention to go beyond static risk assessment and enter the level of dynamic health management; Based on risk level indicators and system resilience factors, a control strength correction factor is generated; this step aims to achieve intelligent and dynamic avoidance control; this embodiment designs a control strength correction factor. It integrates two dimensions: the current risk level and the system resilience. Its calculation formula is as follows: ; in, : Control strength correction factor, a dimensionless modulator used to dynamically modulate the amplitude and strength of various control measures in subsequent evasion strategies; Risk level indicators, whose definitions and sources are as described above; System vulnerability indicators The smaller the value, the larger this value, indicating that the system is more fragile; and : Weighting coefficient, used to adjust the proportion of risk level and system resilience in the final decision, dimensionless, and its source is optimized based on expert experience or reinforcement learning algorithm. The correction factor This ensures that the intensity of control measures is precisely matched with the actual urgency of the risks and the system's capacity to withstand them; Based on risk level indicators and control intensity correction factors, a graded avoidance strategy is implemented, and control signals from the environmental control system are output. This step is the final execution end of the entire technical closed loop, taking corresponding, adjustable control actions based on the assessment results. The method described in this embodiment, through the construction of a complete technical system from implicit environmental parameter mining and multi-scale data fusion risk modeling, to the generation of dynamic control factors combined with system resilience assessment, and finally to graded closed-loop avoidance, can achieve early, accurate warning and proactive intervention for the coordinated asphyxiation disaster of crop groups in plant factories. Compared with the existing technology's management method that only uses threshold alarms based on macroscopic environmental parameters, this invention reveals and quantifies the implicit stress caused by equipment harmonics and integrates system resilience assessment, enabling risk management to shift from reactive response to predictive, adaptive closed-loop avoidance, thereby significantly improving the production stability, resource utilization efficiency, and crop yield of fully artificial light plant factories.

[0017] Example 2: Multi-dimensional state data, including: Microscopic pulse synchronization rate obtained using a blade porosity sensor array; Mesoscopic canopy turbulent kinetic energy spectrum entropy calculated using high-frequency three-dimensional ultrasonic anemometer array data; Macroscopic nonlinear damping coefficients obtained through system identification techniques; The Shannon entropy, a chemical fingerprint of root exudates quantified in hydroponic nutrient solution, was analyzed using liquid chromatography-mass spectrometry.

[0018] This embodiment further defines the method described in Embodiment 1. Specifically, the multi-dimensional state data includes the following four core parameters across scales: The micro-pulse synchronization rate, obtained using a leaf stomatal conductance sensor array, refers to the degree of coordination and consistency of the stomatal opening and closing rhythms of multiple leaves in a crop population. In this embodiment, multiple contact-type leaf stomatal conductance sensors are deployed within the canopy to collect and analyze the time series of their signals in real time, and the synchronization rate is calculated. The purpose of this parameter is to capture changes in population coordination at the most basic physiological unit level, which is essential for subsequent model correction. The direct source; Mesoscopic canopy turbulent kinetic energy spectrum entropy calculated using high-frequency three-dimensional ultrasonic anemometer array data: Mesoscopic canopy turbulent kinetic energy spectrum entropy, i.e., in the aforementioned formula , is a physical quantity that describes the degree of chaos and disorder in the airflow field inside the canopy; in this embodiment, high-resolution three-dimensional wind speed data is collected by deploying a high-frequency three-dimensional ultrasonic anemometer array above and inside the canopy, and the spectrum analysis is performed to calculate the distribution entropy of turbulent kinetic energy at different frequency scales; the purpose of this parameter is to quantify the gas exchange efficiency between the canopy and the environment. The macroscopic nonlinear damping coefficient obtained through system identification technology: the macroscopic nonlinear damping coefficient, which is the aforementioned coefficient used to calculate the toughness factor. It is a macroscopic dynamic parameter that characterizes the ability of the entire crop-environment system as a whole to recover to a steady state after being subjected to external disturbances; the purpose of this parameter is to evaluate the stability and robustness of the system from a macroscopic perspective of system control theory. The chemical fingerprint and Shannon entropy of root exudates in hydroponic nutrient solution were analyzed using liquid chromatography-mass spectrometry: the chemical fingerprint and Shannon entropy of root exudates are expressed as shown in the aforementioned formula. The chemical fingerprint of metabolites is an indicator that measures the diversity of metabolites secreted by crop roots into the hydroponic nutrient solution. In this embodiment, nutrient solution samples are periodically extracted and analyzed using liquid chromatography-mass spectrometry to obtain chemical fingerprints of metabolites, and Shannon entropy is calculated accordingly. The purpose of this parameter is to monitor the health status of crop roots from a biochemical perspective. By specifically defining the above four multi-dimensional state data spanning microphysiology, mesophysics, macrosystems, and biochemistry, this invention can construct a more comprehensive and three-dimensional system state profile than traditional methods, greatly improving the accuracy and reliability of the risk assessment model.

[0019] Example 3: Implicit environment input parameters include: Pulse width modulation spectrum of the growth light source; The infrasound spectrum of the ventilation system.

[0020] This embodiment further defines the method described in Embodiment 1. Specifically, the implicit environment input parameters include: Pulse Width Modulation (PWM) Spectrum of the Growth Light Source: The PWM spectrum of the growth light source refers to the frequency distribution of the PWM signal used to adjust the brightness of the LED growth light source. This invention discovers that a PWM signal of a specific frequency or its harmonics may resonate with the natural opening and closing rhythm of stomata in crop leaves, thereby exacerbating the synchronicity of the population. In this embodiment, the PWM signal is directly monitored by a sensor or read from the light source control system, and a fast Fourier transform is performed on it to obtain the real-time spectrum, which serves as the basis for the aforementioned model. The source; The infrasound spectrum of a ventilation system refers to the frequency distribution of sound waves with frequencies below 20Hz generated by the operation of ventilation system components. This low-frequency vibration can also drive the physiological rhythms of crops. In this embodiment, by deploying infrasound sensors in the environment, collecting environmental acoustic signals and performing spectrum analysis, the energy amplitude of specific frequency bands is extracted as the basis for the aforementioned model. The source of these environmental stresses is revealed by innovatively defining these two usually overlooked parameters as implicit environmental input parameters. By incorporating these parameters into the risk assessment model, the early warning capability of the model is greatly enhanced, providing a direct and physically meaningful basis for risk tracing and subsequent proactive avoidance and control.

[0021] Example 4: Risk level indicators include: The mass asphyxiation risk index was normalized to obtain a risk level index. Based on the numerical range of the risk level indicators, the system status is assessed as safe, Level 1 warning, and Level 2 warning.

[0022] This embodiment further defines the method described in Embodiment 1. Specifically, the steps for determining the risk level indicators include: The mass asphyxiation risk index is normalized to obtain a risk level index; this step has been described in Example 1 using the formula. A description was provided, the purpose of which is to describe a primitive index with complex physical units and a wide range of numerical values. This is transformed into a standardized, dimensionless index with 1 as the critical threshold. ; Based on the numerical range of the risk level indicators, the system status is assessed as safe, Level 1 warning, and Level 2 warning. In this embodiment, the assessment logic is specifically set as follows: Safe state: when At that time, the system is assessed as being in this state; this threshold It is determined based on historical data statistical analysis, and aims to ensure that an early warning is initiated when the system shows a statistically significant risk trend, thereby achieving a balance between sensitivity and avoiding false alarms; Level 1 warning: When When the system is assessed as being in this state, it indicates that the system has entered the critical warning zone and there is a significant risk of sliding towards the boundary layer lock-in state. Level 2 warning: When When the system is assessed as being in this state, it indicates that the risk index has exceeded the critical threshold, and the system is about to or has already experienced boundary layer locking. This implementation method establishes a clear, explicit, and operable risk level classification standard, transforming complex model outputs into simple decision inputs, thus forming a key transformation link from modeling and evaluation to control.

[0023] Example 5: Tiered avoidance strategies include: If the system is in a safe state, the current environmental control strategy will be maintained. If the system status reaches Level 1 warning, then the synchronization suppression mode is activated; If the system status is a Level 2 warning, the system will initiate a mode disruption and system reset procedure.

[0024] This embodiment further defines the method described in Embodiment 4. Its core lies in defining a graded avoidance strategy corresponding to the aforementioned risk levels, including: If the system is in a safe state, the current environmental control strategy will be maintained. In response to a Level 1 warning system status, a synchronization suppression mode is activated. The purpose of this mode is to actively disrupt the external environmental rhythms that may lead to enhanced resonance by introducing non-periodic, mild disturbances, thereby pushing the system back from the critical zone to the safe zone without significantly affecting the normal growth of crops. In response to a level 2 warning in the system status, the system reset procedure is initiated. This is a pre-set, powerful nonlinear control sequence designed to physically break the already formed boundary layer lock-in state and guide the system to recover from an absolutely safe point to a normal operating point via a safe path. The hierarchical avoidance strategy established in this embodiment achieves precise matching between control behavior and risk level, avoiding the drawbacks of traditional control strategies that either fail to act or overreact, and achieving a balance between system safety and production efficiency.

[0025] Example 6: Synchronization suppression modes include: The dimming frequency of the growth light source is actively deviated, and the amount of active deviation is dynamically adjusted by a control intensity correction factor. Random disturbances are introduced into the control signal of the ventilation system, and the intensity of the random disturbances is dynamically adjusted by the control intensity correction factor.

[0026] This embodiment is a further refinement of the method described in Embodiment 5, specifically describing the implementation of the synchronization suppression mode, including: The dimming frequency of the growth light source is actively deviated, with the deviation amount dynamically adjusted by a control intensity correction factor; this operation is called light source avoidance. Specifically, the system will adjust the PWM dimming frequency of the growth light source... Actively deviating from the identified resonant frequency range, offset amount The magnitude of the control intensity correction factor calculated above is related to the magnitude of the control intensity correction factor. For positive correlation, the following adjustment logic is adopted: ; in, The current PWM dimming frequency of the growth light source is the reference frequency before the deviation operation was performed; It is a preset, dimensionless light source frequency adjustment coefficient; this means that when the risk level... Higher or system resilience The worse it is, The larger the value, the greater the frequency deviation, thus achieving a stronger desynchronization effect; A random disturbance is introduced into the control signal of the ventilation system, the intensity of which is dynamically adjusted by a control intensity correction factor; this operation is called ventilation avoidance. Specifically, the system transforms the control signal of the ventilation system from a steady-state output into a dynamic signal containing the base wind speed and a random disturbance term, with the instantaneous wind speed at the vent being... It can be determined by the following formula: ; in, It is the base target wind speed. It is a white noise signal with a standard normal distribution, and the disturbance intensity is... Control intensity correction factor Dynamically determined: ; in It is a preset, dimensionless ventilation disturbance intensity coefficient; The larger the value, the more severe the introduced random fluctuations in wind speed, thus more effectively disrupting the synchronicity of the canopy microclimate. Through the specific means of avoiding light sources and ventilation mentioned above, the synchronization suppression mode can precisely perturb potential resonance sources in an intelligent, dynamic, and collaborative manner. Its core advantage lies in the fact that the strength of all avoidance actions is adjusted by a factor that integrates risk and resilience. Real-time dynamic adjustment enables refined management of the Level 1 warning status.

[0027] Example 7: Mode destruction and system reset procedures include: Drive the ventilation system to execute a turbulence pulse program; After the turbulent pulse program ends, it recovers from the preset safe initial point to the target operating point along a preset nonlinear path.

[0028] This embodiment is a further refinement of the method described in Embodiment 5, specifically describing the implementation of the mode violation and system reset procedure, including: The ventilation system is driven to execute a turbulence pulse program; this operation is called mode disruption. When the system enters a level-two warning state, it immediately reduces the light intensity to a safe level that is harmless to crops and drives the ventilation system to execute a preset, time-limited... For example, a 30-60 second turbulence pulse program; in this program, the ventilation system outputs maximum power and introduces a preset phase difference between different fans to generate strong, chaotic, non-laminar winds; its physical purpose is to forcibly disperse the stagnant air boundary layer that has been locked around the crop canopy with powerful external energy. After the turbulence pulse procedure ends, the system recovers from a preset safe initial point along a preset nonlinear path to the target operating point; this operation is called a system reset. If the mode is violated, the system does not directly recover to the previous target operating point; instead, it first sets all environmental parameters to a preset, absolutely safe initial point, and then follows a pre-planned nonlinear path that avoids known critical regions over a relatively long period of time. For example, within 5-10 minutes, the environmental parameters are slowly restored to the normal target working point. The design of this procedure greatly improves the robustness and fault recovery capability of the entire smart agriculture monitoring system, and is a key technical link to ensure that plant factories are not subject to major losses in extreme situations.

[0029] Example 8: Please see Figure 2 A cloud-based smart agriculture monitoring system includes: The data acquisition module is used to collect multi-dimensional status data and implicit environmental input parameters of the crop population; The risk assessment module is used to calculate the mass asphyxiation risk index based on multi-dimensional state data and implicit environmental input parameters, and to determine the risk level indicators and the corresponding system state. The system resilience monitoring module is used to determine the system resilience factor by applying disturbances to the environmental system and monitoring the response; The closed-loop avoidance control module is used to generate control strength correction factors based on system state, risk level indicators and system resilience factors, execute hierarchical avoidance strategies, and output control signals of the environmental control system.

[0030] This embodiment provides a cloud-based smart agriculture monitoring system. This system is a combination of hardware and software designed to implement the methods described in any of the foregoing embodiments. The system is architectured on a cloud platform, utilizing the powerful computing and storage capabilities of the cloud for complex model calculations and big data analysis. It includes the following core modules: The data acquisition module aims to provide real-time, multi-dimensional data input for the entire system. This module consists of various sensors, analytical instruments, and interfaces with environmental controllers deployed within the plant factory. It is responsible for collecting multi-dimensional status data and implicit environmental input parameters of the crop population and transmitting this raw data to the cloud platform through an IoT gateway. The risk assessment module aims to perform in-depth processing and analysis of the collected data to assess the real-time risk status of the system. This module is an algorithm processing unit deployed on the cloud platform, responsible for calculating the mass asphyxiation risk index based on multi-dimensional state data and implicit environmental input parameters, and determining the risk level indicators and the corresponding system status. The system resilience monitoring module aims to assess the health status and dynamic recovery capability of the system itself. This module is a software subsystem that works in conjunction with the environmental control system and data acquisition module. It is responsible for determining the system resilience factor by applying disturbances to the environmental system and monitoring the response. The closed-loop avoidance control module aims to intelligently decide and execute corresponding control strategies based on evaluation results. This module is a decision control unit connecting analysis and execution. It is responsible for generating control strength correction factors based on system status, risk level indicators, and system resilience factors, executing graded avoidance strategies, and outputting control signals from the environmental regulation system. This system places complex data analysis and decision-making logic in the cloud, while using field sensors and actuators as terminals, achieving efficient resource utilization and strong scalability. It provides a system-level solution that can ensure the efficient, stable, and safe operation of modern plant factories.

[0031] Example 9: The risk assessment module includes: The stress analysis unit is used to calculate the corrected pulse synchronization rate based on the implicit environment input parameters and through the pulse synchronization rate correction model. The risk calculation unit is used to combine the corrected pulse synchronization rate with multi-dimensional state data to calculate the mass asphyxiation risk index through the mass asphyxiation risk index calculation model. The risk level determination unit is used to determine the risk level indicators and assess the system status based on the mass asphyxiation risk index and preset coordination limits.

[0032] This embodiment further defines the system described in Embodiment 8. Specifically, the internal structure of the risk assessment module is divided in detail, including: The stress analysis unit is designed to process raw latent environmental input parameters and quantify their impact on core physiological indicators. This unit is a dedicated algorithm subroutine that receives PWM spectrum data from the growth light source and infrasound spectrum data from the ventilation system from the data acquisition module. Based on the latent environmental input parameters, it calculates the corrected pulse synchronization rate using a pulse synchronization rate correction model. ; The risk calculation unit is designed to integrate multi-source information and calculate a comprehensive risk index; this unit receives the corrected pulse synchronization rate calculated by the stress analysis unit. In addition to other multi-dimensional state data provided by the data acquisition module, it is used to combine the corrected pulse synchronization rate with the multi-dimensional state data to calculate the group asphyxiation risk index through the group asphyxiation risk index calculation model. ; The risk level determination unit is designed to transform the raw risk index into a risk level that can be used for decision-making; this unit receives the herd asphyxiation risk index output by the risk calculation unit. And used for group asphyxiation risk index and preset coordination limits. Determine risk level indicators The system status is assessed, and the output of this unit is directly sent to the closed-loop avoidance control module as the basis for triggering the corresponding control strategy. By further dividing the risk assessment module into functionally independent units, this embodiment makes the internal logic flow of risk assessment clearer and more modular, improving the interpretability and reliability of the system.

[0033] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart agriculture monitoring method based on a cloud platform, characterized in that, include: Collect multi-dimensional state data and implicit environmental input parameters of crop populations; Based on the implicit environmental input parameters, the corrected pulse synchronization rate is calculated using the pulse synchronization rate correction model. By combining the corrected pulse synchronization rate with multi-dimensional state data, the group asphyxiation risk index is calculated using the group asphyxiation risk index calculation model. Risk level indicators are determined based on the mass asphyxiation risk index and preset coordination limits; By applying standardized small perturbations to the environmental system and monitoring the response curves of the system parameters, the nonlinear damping coefficient is identified, and then standardized with the preset critical damping coefficient value to obtain the system toughness factor. Based on risk level indicators and system resilience factors, a control strength correction factor is generated. Based on the risk level indicators and control intensity correction factors, a graded avoidance strategy is implemented, and control signals of the environmental control system are output.

2. The smart agriculture monitoring method based on a cloud platform according to claim 1, characterized in that, Multi-dimensional state data, including: Microscopic pulse synchronization rate obtained using a blade porosity sensor array; Mesoscopic canopy turbulent kinetic energy spectrum entropy calculated using high-frequency three-dimensional ultrasonic anemometer array data; Macroscopic nonlinear damping coefficients obtained through system identification techniques; The Shannon entropy, a chemical fingerprint of root exudates quantified in hydroponic nutrient solution, was analyzed using liquid chromatography-mass spectrometry.

3. The smart agriculture monitoring method based on a cloud platform according to claim 1, characterized in that, Implicit environment input parameters include: Pulse width modulation spectrum of the growth light source; The infrasound spectrum of the ventilation system.

4. The smart agriculture monitoring method based on a cloud platform according to claim 1, characterized in that, Risk level indicators include: The mass asphyxiation risk index was normalized to obtain a risk level index. Based on the numerical range of the risk level indicators, the system status is assessed as safe, Level 1 warning, and Level 2 warning.

5. The smart agriculture monitoring method based on a cloud platform according to claim 4, characterized in that, Tiered avoidance strategies include: If the system is in a safe state, the current environmental control strategy will be maintained. If the system status reaches Level 1 warning, then the synchronization suppression mode is activated; If the system status is a Level 2 warning, the system will initiate a mode disruption and system reset procedure.

6. The smart agriculture monitoring method based on a cloud platform according to claim 5, characterized in that, Synchronization suppression modes include: The dimming frequency of the growth light source is actively deviated, and the amount of active deviation is dynamically adjusted by a control intensity correction factor. Random disturbances are introduced into the control signal of the ventilation system, and the intensity of the random disturbances is dynamically adjusted by the control intensity correction factor.

7. The smart agriculture monitoring method based on a cloud platform according to claim 5, characterized in that, Mode destruction and system reset procedures include: Drive the ventilation system to execute a turbulence pulse program; After the turbulent pulse program ends, it recovers from the preset safe initial point to the target operating point along a preset nonlinear path.

8. A cloud-based smart agriculture monitoring system, based on the cloud-based smart agriculture monitoring method according to any one of claims 1-7, characterized in that, include: The data acquisition module is used to collect multi-dimensional status data and implicit environmental input parameters of the crop population; The risk assessment module is used to calculate the mass asphyxiation risk index based on multi-dimensional state data and implicit environmental input parameters, and to determine the risk level indicators and the corresponding system state. The system resilience monitoring module is used to determine the system resilience factor by applying disturbances to the environmental system and monitoring the response; The closed-loop avoidance control module is used to generate control strength correction factors based on system state, risk level indicators and system resilience factors, execute hierarchical avoidance strategies, and output control signals of the environmental control system.

9. A smart agriculture monitoring system based on a cloud platform according to claim 8, characterized in that, The risk assessment module includes: The stress analysis unit is used to calculate the corrected pulse synchronization rate based on the implicit environment input parameters and through the pulse synchronization rate correction model. The risk calculation unit is used to combine the corrected pulse synchronization rate with multi-dimensional state data to calculate the mass asphyxiation risk index through the mass asphyxiation risk index calculation model. The risk level determination unit is used to determine the risk level indicators and assess the system status based on the mass asphyxiation risk index and preset coordination limits.

Images