Garden vegetation growth state monitoring system based on nonlinear time sequence prediction

Abstract

The invention discloses a garden vegetation growth state monitoring system based on nonlinear time sequence prediction, and relates to the technical field of data analysis, a data acquisition part is used for acquiring growth characteristic parameters of each vegetation in each observation time window in a garden and acquiring environmental parameters of an environment where the vegetation is located in each observation time window; the vegetation photosynthetic efficiency evaluation part is used for evaluating the vegetation growth potential, the photosynthetic efficiency and the water utilization efficiency of the vegetation in each observation time window according to the growth characteristic parameters and the environmental parameters; and the vegetation growth state evaluation part is used for obtaining the vegetation growth amount of the current time window and predicting the vegetation biomass of the next observation time window according to the vegetation growth potential, the photosynthesis efficiency and the water utilization efficiency. According to the method, the future growth state can be accurately predicted based on the historical growth data, and potential risks such as water stress, high temperature stress or nutrient insufficiency can be identified in advance.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and more particularly to a monitoring system for the growth state of garden vegetation based on non - linear time series prediction. Background Art

[0002] In the process of fine management and ecological restoration of garden vegetation, the dynamic monitoring and prediction of the growth state of vegetation have always been the key points and difficulties in the development of technology. With the acceleration of the urbanization process, a large number of landscaping projects, wetland ecological restoration, and vegetation maintenance in urban parks require more accurate data support to ensure the good growth state of vegetation and the stable operation of the ecosystem. However, since the growth of vegetation is jointly affected by multiple environmental factors, including light, water, temperature, soil nutrients, pH, and salt stress, etc., these factors interact with each other and present complex non - linear relationships. Therefore, the existing methods for monitoring the growth state of vegetation often cannot accurately reflect the dynamic changes in vegetation growth under different environmental conditions, resulting in a large deviation in the assessment of the health state of vegetation.

[0003] In terms of monitoring the dynamic changes in biomass, the Logistic growth model and the Gompertz growth model, as common vegetation growth prediction models, usually use fixed growth rate parameters to perform static simulation on biomass. These models describe the growth trend of vegetation by defining a fixed maximum biomass level and growth rate, but they cannot fully reflect the dynamic effects of multiple factors such as photosynthesis efficiency, water use efficiency, and respiratory metabolism consumption on biomass changes. In practical applications, due to the coupling effect of complex factors such as photosynthesis rate, water supply, temperature deviation, and respiratory metabolism on the growth state of vegetation, simply relying on growth models with fixed parameters cannot accurately reflect the biomass accumulation law of vegetation under different environmental conditions. Especially under high - temperature, drought, or salt - stress conditions, the basal respiratory metabolism rate will increase significantly, significantly inhibiting the biomass accumulation rate, and the existing growth models fail to introduce a temperature - sensitive respiratory metabolism correction mechanism, resulting in a large error in the biomass prediction results. Summary of the Invention

[0004] To solve the above - mentioned technical problems, a monitoring system for the growth state of garden vegetation based on non - linear time series prediction is provided, which can accurately predict the future growth state based on historical growth data and identify potential risks such as water stress, high - temperature stress, or nutrient deficiency in advance. Compared with traditional biomass growth models, the present invention significantly improves the accuracy of growth state assessment.

[0005] To achieve the above - mentioned purpose, the technical solution adopted by the present invention is as follows:

[0006] A monitoring system for the growth state of garden vegetation based on non - linear time series prediction, comprising:

[0007] A data acquisition part, which is used to acquire the growth characteristic parameters of each type of vegetation in the garden in each observation time window, and acquire the environmental parameters of the environment where the vegetation is located in each observation time window;

[0008] A vegetation photosynthetic efficiency evaluation part, which is used to evaluate the vegetation growth potential, photosynthesis efficiency and water use efficiency of the vegetation in each observation time window according to the growth characteristic parameters and environmental parameters;

[0009] A vegetation growth status evaluation part, which is used to evaluate the vegetation biomass increment of the current observation time window compared with the previous observation time window according to the vegetation growth potential, photosynthesis efficiency and water use efficiency, and accumulate the vegetation biomass of the previous observation time window to obtain the vegetation growth amount of the current time window and predict the vegetation biomass of the next observation time window.

[0010] Further, in the first observation time window, the biomass of each type of vegetation is set to an initial value corresponding to that type of vegetation; the vegetation growth amount of the second observation time window is obtained by accumulating the vegetation growth amount of the first observation time window plus the vegetation biomass increment between the first observation time window and the second observation time window, and so on, to obtain the vegetation growth amount of each observation time window and predict the vegetation growth amount of the next observation time window of the current observation time window.

[0011] Further, the growth characteristic parameters include: leaf transpiration rate Tr(t), with the unit of mmolH2O / m 2 ·s, vegetation light saturation constant K P , with the unit of μmol / m 2 ·s; vegetation maximum photosynthetic rate P max , with the unit of μmolCO2 / m 2 ·s; vegetation growth minimum temperature T min , with the unit of °C; vegetation growth optimum temperature T opt , with the unit of °C; vegetation growth maximum temperature T max , with the unit of °C; vegetation growth minimum pH value pH min ; vegetation growth optimum pH value pH opt ; among them, the leaf transpiration rate Tr(t) is measured by a porometer; the vegetation light saturation constant K P is determined by a light response curve, and the value range is from 200 to 500; the vegetation maximum photosynthetic rate P max is determined by a photosynthesis meter; t represents the t-th observation time window; the length of each observation time window is equal.

[0012] Further, the environmental parameters include: photosynthetically active radiation PAR(t), with the unit of μmol / m 2·s, obtained by measuring with a quantum sensor; soil water content SM(t), obtained by measuring with a soil moisture sensor; soil electrical conductivity EC(t), unit: dS / m, obtained by measuring with a conductivity sensor; soil nitrogen content N(t), unit: mg / kg, obtained by measuring with a soil nutrient sensor; photosynthetic photon flux density PPFD(t), unit: μmol / m 2 ·s, obtained by measuring with a quantum sensor; soil pH value pH(t), obtained by measuring with a pH sensor; vapor pressure deficit VPD(t), unit: kPa, obtained by calculating after measuring with a temperature and humidity sensor; relative humidity RH(t), obtained by measuring with a humidity sensor; current temperature T(t), unit: °C, obtained by measuring with a temperature sensor.

[0013] Furthermore, the vegetation photosynthetic efficiency evaluation part evaluates the vegetation growth potential of the vegetation in each observation time window as:

[0014]

[0015] Among them, G(t) represents the vegetation growth potential in the t-th observation time window, and its value is a real number greater than 1 and less than 0; FC is the field capacity, and its value range is different according to the different soil types where the vegetation is located; α is the salt stress coefficient, unit: m / dS, and its value range is from 0.2 to 0.4.

[0016] Furthermore, if the soil type where the vegetation is located is sandy soil, the value range of FC is from 0.05 to 0.12; if the soil type where the vegetation is located is loam soil, the value range of FC is from 0.20 to 0.35; if the soil type where the vegetation is located is clay loam soil, the value range of FC is from 0.30 to 0.45; if the soil type where the vegetation is located is clay soil, the value range of FC is from 0.35 to 0.50.

[0017] Furthermore, the vegetation photosynthetic efficiency evaluation part evaluates the photosynthesis efficiency of the vegetation in each observation time window as:

[0018]

[0019] Among them, P(t) represents the photosynthesis rate in the t-th observation time window, unit: μmolCO2 / m 2 ·s; K N represents the nitrogen half-saturation constant, unit: mg / kg, and its value range is from 25 to 40; β represents the pH deviation sensitivity coefficient, and its value range is from 0.3 to 0.6.

[0020] Furthermore, the vegetation photosynthetic efficiency evaluation part evaluates the water use efficiency of the vegetation in each observation time window as:

[0021]

[0022] Among them, W(t) represents the water use efficiency of the t-th observation time window, with the unit of mmolCO2 / molH2O; γ represents the water stress response coefficient, and its value range is from 2.5 to 4.0; δ represents the temperature deviation sensitivity coefficient, with the unit of 1 / °C, and its value range is from 0.08 to 0.12.

[0023] Furthermore, the vegetation growth state evaluation part evaluates the vegetation biomass of the t-th observation time window as:

[0024]

[0025] Among them, B(t) is the vegetation biomass of the t-th observation time window; B(t - 1) is the vegetation biomass of the (t - 1)-th observation time window; s is the time integration variable; R0 represents the basal respiration rate of the vegetation, and its value range is from 0.01 to 0.02; Q 10 is the respiration temperature sensitivity coefficient, and its value is 2.0.

[0026] Compared with the prior art, the beneficial effects of the present invention are as follows: Compared with the traditional biomass evaluation model, the present invention can not only dynamically model the vegetation growth potential, photosynthesis efficiency and water use efficiency within different observation time windows, but also accurately depict the non-linear amplification effect between the basal respiration metabolic rate and temperature changes, ensuring that the biomass update result is closer to the actual growth state. At the same time, by introducing a sliding time window adaptive adjustment mechanism, the present invention can dynamically adjust the length of the observation window according to the changes in the vegetation growth stage, thereby increasing the data acquisition frequency during the rapid growth period and extending the time window during the stable growth period to reduce the calculation burden, achieving a dynamic balance between data accuracy and calculation efficiency. In addition, through the non-linear time series modeling and prediction algorithm, the present invention can accurately predict the future growth state based on historical growth data, identify risks such as possible water stress, high temperature stress or nutrient deficiency in advance, and help managers take intervention measures such as irrigation, fertilization or cooling in a timely manner, improving the stability and stress resistance of vegetation growth. Compared with traditional static growth models with fixed growth rates such as the Logistic model and the Gompertz model, the present invention adopts an integral form of biomass dynamic update equation, combines the dynamic balance of the photosynthesis accumulation term and the temperature-sensitive respiration metabolism consumption term, realizes the dynamic tracking and multi-dimensional correction of the biomass growth process, and thus greatly improves the accuracy and reliability of the growth state evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the system structure of the garden vegetation growth state monitoring system based on non-linear time series prediction proposed by the present invention. Detailed implementation mode

[0028] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0029] Referring to Figure 1 As shown, the garden vegetation growth status monitoring system based on non-linear time series prediction in the embodiment of the present invention includes:

[0030] A data acquisition part, which is used to acquire the growth characteristic parameters of each type of vegetation in the garden at each observation time window, and acquire the environmental parameters of the environment where the vegetation is located at each observation time window;

[0031] The core technology of the data acquisition part lies in the dynamic acquisition and preprocessing of vegetation growth characteristic parameters and environmental parameters. Vegetation growth characteristic parameters include leaf transpiration rate, light saturation constant, maximum photosynthetic rate, growth temperature threshold, and adaptive pH range, etc. These parameters directly reflect the photosynthesis potential, biomass accumulation rate, and environmental adaptability of vegetation. In the actual implementation process, parameters are measured by high-precision devices such as a porometer, a photosynthesis meter, and a pH sensor, and the original data is uploaded to the data storage unit through a wireless data transmission module to form a time series database. And environmental parameters include photosynthetically active radiation, soil water content, conductivity, photon flux density, temperature, humidity, vapor pressure deficit, soil nitrogen content, and pH value, etc. These data are acquired through a multi-modal sensor array. These sensor devices are precisely calibrated and calibrated to ensure the measurement accuracy of the data and avoid system deviations caused by environmental changes. At the same time, in order to adapt to the growth characteristics of vegetation in different growth stages, the present invention also designs a dynamic time window adaptive adjustment mechanism, which dynamically optimizes the length of the data acquisition time window according to the growth cycle of the vegetation, the environmental change frequency, and the microclimate characteristics. This mechanism can effectively avoid the problems of insufficient sampling accuracy or data redundancy caused by the traditional fixed time window method, thereby improving the sensitivity and reliability of data acquisition.

[0032] In the specific implementation process, the present invention adopts a multi-layer data fusion architecture. The underlying data is collected in real time through a wireless sensor network (WSN), and each sensing node is responsible for collecting environmental parameter or vegetation characteristic parameter collection tasks in a specific area. The intermediate layer data is preliminarily preprocessed through edge computing devices, including operations such as data denoising, outlier removal, and linear interpolation filling. In order to cope with the complex non-linear relationship between vegetation growth parameters and environmental factors, the present invention introduces a non-linear dimensionality reduction algorithm in data preprocessing, projects high-dimensional multi-parameter data onto a low-dimensional feature space, so as to reduce the computational complexity and improve the training efficiency of the model. The top layer data is stored and managed through a cloud computing platform, providing full-scale data support for subsequent photosynthesis efficiency modeling and growth state prediction.

[0033] The data acquisition part of the present invention adopts a multi-parameter collaborative modeling mechanism. By synchronously aligning vegetation characteristic data and environmental parameters in different time windows, a multi-dimensional time series matrix is formed. This matrix not only includes the growth characteristic change information in the time dimension, but also integrates the environmental change trend and growth state response characteristics, thus providing a rich data basis for the construction of a non-linear time series model. Compared with the traditional monitoring system that only relies on single-time-point data for static analysis, this system realizes continuous collection and analysis of multi-window data through a dynamic time window sliding mechanism, thus capturing the dynamic characteristics of non-linear changes in the vegetation growth process. Specifically, through multi-dimensional spatio-temporal coupling modeling, the present invention dynamically maps the correlation between vegetation characteristic data and environmental factors into a time series state vector and updates its state within each time window. Combining key environmental parameters such as photosynthetically active radiation, soil moisture, and temperature, the present invention realizes environmental-vegetation dynamic interaction modeling, which can accurately reflect the non-linear coupling relationship between vegetation growth potential and environmental changes.

[0034] The vegetation photosynthetic efficiency evaluation part is used to evaluate the vegetation growth potential, photosynthesis efficiency, and water use efficiency of vegetation in each observation time window according to growth characteristic parameters and environmental parameters;

[0035] The core of the vegetation photosynthetic efficiency evaluation part lies in the dynamic response modeling of photosynthesis. Its modeling mechanism is based on a non-linear temperature-photosynthesis coupling model, which fully simulates the complex relationship between photosynthesis and temperature changes. The system first collects photosynthetically active radiation (PAR) through a quantum sensor and calculates the photosynthesis temperature response function of vegetation in the current time window in combination with the environmental temperature data collected by the temperature and humidity sensor. This function simulates the temperature gradually changing from the lowest growth temperature (T min ) to the optimum temperature (T opt)The promotion effect of photosynthesis during the change process, and the photosynthetic attenuation trend after the temperature exceeds the optimal temperature is described by a Gaussian decay function. This non-linear temperature - photosynthesis modeling method enables the system to accurately evaluate the photosynthesis state of vegetation under different environmental temperatures and dynamically adjust the growth potential evaluation results.

[0036] In addition to temperature factors, the vegetation photosynthetic efficiency evaluation part also fully considers the effects of soil moisture and salt stress. The system collects the soil water content through a soil moisture sensor, compares it with the field capacity, and calculates the water use efficiency. When the soil water content is close to the field capacity, the photosynthesis of the vegetation reaches the best state, while when the soil moisture is in a shortage or surplus state, the photosynthesis efficiency is inhibited. At the same time, the system also obtains the conductivity data through a soil conductivity sensor, judges the soil salt level, and dynamically adjusts the photosynthesis evaluation results through the salt stress inhibition term. This water - salt interaction inhibition modeling mechanism enables the system to precisely adjust the photosynthetic potential of vegetation under different soil conditions and incorporates the non-linear effects of soil stress into the overall evaluation model.

[0037] The dynamic modeling of photosynthesis efficiency also introduces a non-linear feedback mechanism of photon utilization efficiency and water use efficiency. The system measures the leaf transpiration rate through a porometer and calculates the water use efficiency (WUE) in combination with the vapor pressure deficit (VPD), so as to dynamically optimize the transpiration - photosynthesis balance state during the photosynthesis process. This part of the evaluation process not only considers the balance relationship between water transpiration and carbon dioxide fixation during photosynthesis, but also combines the feedback regulation of transpiration by environmental relative humidity changes to achieve the dynamic non-linear optimization of photosynthesis efficiency. This multi-factor dynamic coupling model can accurately reflect the dynamic changes of photosynthesis of vegetation under different environmental conditions, thus providing more reliable data support for the subsequent growth state evaluation.

[0038] In the actual implementation process, the present invention adopts a multi-layer data fusion and non-linear regression modeling technology. The original data collected by multi-modal sensors is preprocessed, and the photosynthesis response model is trained through a Bayesian non-linear fitting algorithm. The photosynthesis-related parameters obtained by the system in the data acquisition part, including multi-dimensional data such as photosynthetically active radiation, soil moisture, temperature, conductivity, etc., are preprocessed through a time series interpolation and outlier rejection mechanism to ensure the reliability of data input. After data cleaning and dimensionality reduction, the system calculates the dynamic response values of photosynthesis efficiency, growth potential and water use efficiency through a non-linear regression model, and inputs them as a time series state vector into the vegetation growth state evaluation part for further analysis.

[0039] The vegetation growth status assessment part is used to evaluate the vegetation biomass increment in the current observation time window compared with the previous observation time window according to the vegetation growth potential, photosynthesis efficiency, and water use efficiency, accumulate the vegetation biomass in the previous observation time window, obtain the vegetation growth amount in the current time window, and predict the vegetation biomass in the next observation time window.

[0040] In the garden vegetation growth status monitoring system based on non-linear time series prediction of the present invention, the vegetation growth status assessment part is a key module for realizing the dynamic prediction of vegetation biomass and the analysis of growth trends. This part performs non-linear modeling on the vegetation growth potential, photosynthesis efficiency, and water use efficiency, dynamically accumulates the data of different observation time windows, and dynamically updates the vegetation biomass based on the non-linear biomass growth model in integral form. The core of vegetation growth status assessment lies in integrating the coupling effect of photosynthesis and water transpiration, the temperature sensitivity of respiratory metabolism, and the dynamic feedback of environmental stress factors into a time-varying integral equation, enabling the system to accurately predict the biomass change of vegetation at different growth stages. This model not only considers the direct contribution of vegetation photosynthesis to biomass growth but also combines the inhibitory effect of the basal respiration consumption term on the growth rate, thus realizing the multi-dimensional dynamic assessment of vegetation growth status.

[0041] The growth characteristic parameters and environmental parameters obtained by the system in each observation time window are used as model inputs. These parameters include the photosynthesis rate, transpiration rate, photosynthetically active radiation, soil moisture, conductivity, temperature, humidity, etc. of the vegetation. Through non-linear time series modeling, the system maps these multi-dimensional parameters into the state space of vegetation growth potential, photosynthesis efficiency, and water use efficiency, providing data support for subsequent growth status assessment. The key to growth status assessment lies in dynamically calculating the growth increment of vegetation through the photosynthesis-water use synergistic growth mechanism and correcting the growth increment through the respiratory metabolism dynamic feedback mechanism. In this process, the system uses the integral form of the biomass dynamic update equation to accumulate the vegetation growth amount with the data of the previous observation time window, thus forming a time series model that can dynamically track the biomass change trend.

[0042] The key technology of vegetation growth status assessment lies in the non-linear metabolism correction model based on temperature sensitivity and respiratory inhibition mechanism. The system collects environmental temperature data through temperature and humidity sensors and combines the respiratory temperature sensitivity coefficient Q 10 to dynamically correct the basal respiration rate. When the temperature rises, the basal respiration rate increases with Q 10The coefficient shows exponential growth, leading to an increase in biomass consumption. When the temperature approaches the optimal growth temperature, the inhibitory effect of the respiratory metabolism term on biomass growth weakens. The system introduces a basal respiration inhibition term into the vegetation biomass growth model, integrates it with the photosynthesis and water use terms, and thus dynamically corrects the dynamic changes of vegetation biomass. This mechanism ensures that under different temperature conditions, the dynamic change trend of the vegetation growth state can be accurately captured, thereby improving the accuracy of growth state prediction.

[0043] In the actual implementation process, the vegetation growth state assessment part performs multi-dimensional spatio-temporal coupling on the data obtained by different sensors through a multi-layer data fusion mechanism, and adopts an adaptive time window adjustment algorithm to dynamically optimize the data acquisition frequency. When the vegetation is in the rapid growth period, the system automatically shortens the length of the observation time window and increases the data acquisition frequency to capture the growth characteristic changes in the rapid growth stage; when the vegetation is in the stable growth period or the growth stagnation period, the system automatically increases the length of the observation time window and reduces the data acquisition frequency, thereby optimizing the energy consumption and storage resource allocation of the system. This time window adaptive mechanism enables the system to dynamically adjust the monitoring strategy according to the changes in the vegetation growth stage, thereby improving the operation efficiency of the system while ensuring the data quality.

[0044] To further improve the accuracy of growth state assessment, the present invention introduces a photosynthesis-water use efficiency synergistic growth mechanism into the biomass growth integration model, calculates the water use efficiency through the leaf transpiration rate and the vapor pressure deficit (VPD), and synergistically optimizes it with the photosynthesis efficiency. When the water use efficiency is high, the system automatically increases the contribution of photosynthesis to biomass growth, and when the water use efficiency decreases, the system inhibits the biomass accumulation rate by reducing the weight of the photosynthesis term. This mechanism effectively simulates the dynamic balance relationship between water and photosynthesis, and can accurately evaluate the vegetation growth state under extreme conditions such as drought and over-wetness.

[0045] Furthermore, in the first observation time window, the biomass of each type of vegetation is set to an initial value corresponding to that type of vegetation; the vegetation growth amount in the second observation time window is obtained by adding the vegetation growth amount in the first observation time window to the vegetation biomass growth amount between the first and the second observation time windows, and so on, to obtain the vegetation growth amount in each observation time window and predict the vegetation growth amount in the next observation time window of the current observation time window.

[0046] In this process, the system adopts a time - cumulative integration mechanism, modeling the dynamic changes in the vegetation growth state as a time - series integration problem. For the first observation time window, the vegetation biomass is initialized to a known initial value corresponding to the vegetation species, which can be dynamically set according to historical growth data, species characteristics, and the ecological characteristics of the vegetation community. Subsequently, within each time window, the system dynamically corrects by calculating the product of photosynthesis efficiency and water - use efficiency, combined with multiple factors such as temperature changes, soil moisture, and salt stress, to obtain the vegetation biomass growth increment within the current time window. This increment is accumulated through non - linear time - series modeling to the vegetation biomass of the previous time window, thereby obtaining the vegetation growth amount in the current time window. The system calculates the biomass change amount between different time windows through an integral - form biomass dynamic update equation to achieve the gradual dynamic update of biomass.

[0047] This dynamic update mechanism can not only accurately reflect the change trend of vegetation biomass at different growth stages but also perform continuous state prediction in the time dimension. The system adopts a sliding - window prediction mechanism to predict the vegetation growth state in future observation time windows. Based on the biomass data of the current observation time window and state vectors such as environmental parameters, photosynthesis efficiency, and water - use efficiency, the system dynamically predicts the biomass of the next observation time window through an autoregressive integrated model (ARIM). This prediction process combines non - linear time - series regression, temperature - response modeling, and transpiration - photosynthesis coupling mechanisms, thus achieving an accurate prediction of the future vegetation growth state. The biomass of each observation time window is not only affected by the growth characteristics and environmental parameters within the current window but also dynamically corrected by the biomass accumulation and respiration consumption factors in the previous time window. Therefore, the system adopts a biomass update mechanism based on temperature sensitivity and respiration metabolism correction, and non - linearly adjusts the biomass accumulation rate by introducing a basic respiration inhibition term, so that the prediction results can accurately reflect the true state of vegetation growth.

[0048] Furthermore, the growth characteristic parameters include: the leaf transpiration rate Tr(t), with the unit of mmolH2O / m 2 ·s, the vegetation light - saturation constant K P , with the unit of μmol / m 2 ·s; the maximum photosynthetic rate P max , with the unit of μmolCO2 / m 2 ·s; the lowest temperature T min for vegetation growth, with the unit of °C; the optimum temperature T opt for vegetation growth, with the unit of °C; the highest temperature T max for vegetation growth, with the unit of °C; the lowest pH value pH min for vegetation growth; the optimum pH value pHopt ; among them, the leaf transpiration rate Tr(t) is measured by a porometer; the light saturation constant K of the vegetation P is determined by a light response curve, and its value range is from 200 to 500; the maximum photosynthetic rate P of the vegetation max is measured by a photosynthesis meter; t represents the t-th observation time window; the length of each observation time window is equal.

[0049] The leaf transpiration rate Tr(t) is first introduced into the system as one of the key growth characteristic parameters, with the unit of mmolH2O / m 2 ·s, and is measured by a porometer. The leaf transpiration rate is an important indicator reflecting the water use efficiency of vegetation and the coupling relationship between transpiration and photosynthesis. It directly affects the accumulation of photosynthetic products and the water evaporation rate. When photosynthesis is active, the stomata open more, and water is lost through transpiration, thus affecting the leaf water potential and photosynthesis intensity. Therefore, within each observation time window, the system dynamically monitors Tr(t), and combines the changes in vapor pressure deficit (VPD) and ambient relative humidity to establish a transpiration rate - photosynthesis feedback model. This model can simulate the dynamic regulation mechanism of transpiration rate on photosynthesis under different environmental stresses, providing accurate dynamic inputs for the subsequent growth state assessment and water regulation strategy of the system.

[0050] The light saturation constant K of the vegetation P is one of the important parameters for the system to evaluate the photosynthesis efficiency, with the unit of μmol / m 2 ·s, and its value range is from 200 to 500, and is dynamically obtained by the light response curve determination method. K P represents the light level at which the photosynthesis rate of the vegetation tends to saturate after the light intensity reaches a certain threshold. The system measures the photosynthetically active radiation (PAR) through a quantum sensor, and combines the light response model (LightResponseModel) to calculate the change trend of K P . When the light intensity gradually increases, the photosynthesis rate shows a non-linear growth and tends to be stable after reaching the light saturation point K P . Therefore, the change of K P can reflect the adaptability of the vegetation to light conditions. The system uses K P as a photosynthesis potential evaluation factor, and combines parameters such as temperature and soil moisture to dynamically correct the photosynthesis efficiency evaluation model, so as to accurately evaluate the growth state of the vegetation under different light conditions.

[0051] The maximum photosynthetic rate P of the vegetation max reflects the maximum photosynthesis rate that the vegetation can reach under the optimal environmental conditions, with the unit of μmolCO2 / m 2·s, measured by a photosynthesis meter. Within each observation time window, the system measures the photosynthesis rate of vegetation under different light intensities, fits a light response curve, and extracts the maximum photosynthesis rate P from it. max P max The change of P reflects the physiological activity level and environmental adaptability of vegetation. Under different growth stages or environmental stress conditions, its value will be dynamically adjusted with the changes of factors such as light, temperature, and moisture. Therefore, the system uses a non-linear photosynthesis-environment coupling model to take P max as an important input variable for predicting the growth state, and combines historical growth data to dynamically correct the future growth trend.

[0052] The growth temperature thresholds of vegetation include the minimum growth temperature T min , the optimum growth temperature T opt , and the maximum growth temperature T max , with the unit of °C. These parameters together determine the growth state and photosynthesis potential of vegetation under different temperature conditions. The system uses temperature and humidity sensors to collect environmental temperature data in real time, and combines a Gaussian temperature response model to model the relationship between the photosynthesis rate and temperature changes. When the environmental temperature is lower than T min , the photosynthesis rate approaches zero, and when the temperature reaches T opt , the photosynthesis rate reaches the maximum value. Subsequently, as the temperature rises to T max , the photosynthesis rate gradually decays. The system models this process through a non-linear temperature response function and dynamically corrects it by combining factors such as photosynthetically active radiation and soil moisture to improve the accuracy of growth state prediction under different temperature conditions.

[0053] In addition, the system also introduces the growth adaptability pH range of vegetation as an important reference parameter for growth state evaluation, where the minimum pH value pH min and the optimum pH value pH opt are measured by a soil pH sensor. The pH value directly affects the ability of vegetation to absorb soil nutrients, thereby indirectly affecting the photosynthesis efficiency and biomass growth rate. When the soil pH value is lower than pH min or higher than pH max , the growth rate of vegetation will be inhibited, while near pH opt , the nutrient absorption ability and photosynthesis efficiency of vegetation are both in the best state. The system dynamically monitors the change of soil acidity and alkalinity through a soil pH adjustment model, and combines the growth state of vegetation for real-time feedback adjustment, so as to optimize the growth potential and water use efficiency of vegetation under different soil conditions.

[0054] Furthermore, the environmental parameters include: photosynthetically active radiation PAR(t), with the unit of μmol / m2 ·s, obtained by measuring with a quantum sensor; soil water content SM(t), obtained by measuring with a soil moisture sensor; soil conductivity EC(t), in dS / m, obtained by measuring with a conductivity sensor; soil nitrogen content N(t), in mg / kg, obtained by measuring with a soil nutrient sensor; photosynthetic photon flux density PPFD(t), in μmol / m 2 ·s, obtained by measuring with a quantum sensor; soil pH value pH(t), obtained by measuring with a pH sensor; vapor pressure deficit VPD(t), in kPa, obtained by calculation after measuring with a temperature and humidity sensor; relative humidity RH(t), obtained by measuring with a humidity sensor; current temperature T(t), in °C, obtained by measuring with a temperature sensor.

[0055] First, the system uses a quantum sensor to collect photosynthetically active radiation (PAR), in μmol / m 2 ·s, which reflects the quantum energy available for plants to carry out photosynthesis. PAR is the direct source of the driving force for photosynthesis. Under different light conditions, its changes will significantly affect the photosynthesis rate and the change trend of growth potential. The system densely arranges a quantum sensor array in different vegetation areas to collect quantum flux data at different heights and in different directions in real time, and combines photosynthetic photon flux density (PPFD) for spatial interpolation to construct a multi-dimensional photosynthetic radiation distribution model. As a key subset of PAR, PPFD is also in μmol / m 2 ·s, obtained by measuring with a quantum sensor, and reflects the number of effective photons received by the vegetation within a specific time window. The change of PPFD is closely related to the photosynthesis rate, stomatal opening degree and transpiration of plants. Therefore, the system realizes the accurate evaluation of photosynthesis potential under different light conditions by dynamically coupling and modeling PAR and PPFD.

[0056] Soil water content (SM) is a key moisture factor affecting vegetation growth and photosynthesis efficiency, in m 3 / m 3 , obtained by measuring with a soil moisture sensor. SM reflects the saturation of water in the soil. The system obtains soil water content data at different depths through a soil moisture sensor in each observation time window, and dynamically corrects the water use efficiency in combination with the field capacity (FC). When the soil water content approaches FC, both the transpiration rate and photosynthesis rate of the vegetation are in the best state. When SM is lower than FC, the water use efficiency of the vegetation is inhibited, leading to a decrease in the photosynthesis rate. The system dynamically adjusts the calculation result of photosynthetic efficiency through a non-linear water-photosynthesis coupling model, and combines VPD and RH for water stress correction to optimize the growth state evaluation.

[0057] Soil electrical conductivity (EC) reflects the concentration level of dissolved salts in the soil, with the unit of dS / m, and is obtained by measuring with an electrical conductivity sensor. The change of EC has a significant impact on the water absorption capacity and photosynthesis rate of vegetation. Especially in a high-salt environment, EC will cause salt stress phenomenon, thus inhibiting the normal growth state of vegetation. The system calculates the dynamic inhibition effect of salt stress on the growth state through the conductivity-photosynthesis stress model, and introduces a salt inhibition factor to correct the photosynthesis efficiency, so as to achieve the adaptive optimization of the growth potential under different soil electrical conductivity conditions.

[0058] Soil nitrogen content (N) is an essential nutrient element in the process of plant growth, with the unit of mg / kg, and is obtained by measuring with a soil nutrient sensor. The change of soil nitrogen content directly affects the photosynthesis potential and biomass accumulation rate of vegetation. The system calculates the change trend of photosynthesis efficiency of vegetation at different nitrogen content levels through the nitrogen nutrient response model, and combines the photosynthesis model to dynamically predict the growth state of the future observation window. When the nitrogen content is close to the optimal level, the growth potential of vegetation is the largest. When the nitrogen content is insufficient, the system automatically reduces the evaluation result of photosynthesis potential and triggers the soil nitrogen nutrient supplement strategy through the fertilization feedback mechanism, so as to ensure the continuous growth state of vegetation.

[0059] Soil pH value (pH) is an important parameter that affects the absorption of nutrients and water by the roots of vegetation, and is dynamically measured by a pH sensor. The change of pH directly affects the activity of root microorganisms and the availability of soil nutrients. The system calculates the photosynthesis rate of vegetation under different pH conditions through the pH-photosynthesis regulation model, and combines the nonlinear pH response mechanism to correct the growth state. When the pH is lower or higher than the optimal growth range of vegetation, the system automatically reduces the evaluation values of growth potential and photosynthetic potential, and makes timely adjustment through the soil improvement feedback module to ensure that the vegetation is always in the optimal growth conditions.

[0060] Atmospheric vapor pressure deficit (VPD) is obtained by calculating with a temperature and humidity sensor, with the unit of kPa. VPD reflects the balance state between the water evaporation potential in the atmosphere and the transpiration rate of plants, and is an important feedback parameter for the water use efficiency of vegetation. When VPD is too high, the transpiration rate of vegetation increases, but the water use efficiency decreases. The system calculates the inhibition effect of VPD on the water use of vegetation through the transpiration-photosynthesis dynamic feedback mechanism, and combines RH and T to correct the transpiration rate, so as to optimize the dynamic prediction of growth potential and water use efficiency.

[0061] Relative humidity (RH) and current temperature (T) are also important environmental parameters dynamically monitored by the system, and are measured in real time through humidity sensors and temperature sensors. Changes in RH and T directly affect the degree of stomatal opening and transpiration rate, and jointly determine the water use efficiency of vegetation with VPD. The system calculates the transpiration efficiency and water use efficiency under different temperature and humidity conditions through a temperature-humidity-transpiration coupling model, and couples it with the photosynthesis efficiency to correct the growth state. When RH is too low or T is too high, the system automatically reduces the predicted growth potential result and dynamically adjusts through the irrigation and temperature control feedback mechanism to keep the vegetation in the best growth state.

[0062] Furthermore, the vegetation photosynthetic efficiency evaluation part evaluates the growth potential of the vegetation in each observation time window as:

[0063]

[0064] where G(t) represents the growth potential of the vegetation in the t-th observation time window, and its value is a real number greater than 1 and less than 0; FC is the field capacity, and its value range is different according to the different soil types where the vegetation is located; α is the salinity stress coefficient, with the unit of m / dS, and its value range is from 0.2 to 0.4.

[0065] In the garden vegetation growth state monitoring system based on nonlinear time series prediction of the present invention, the vegetation photosynthetic efficiency evaluation part adopts a multi-factor coupling model to dynamically evaluate the growth potential of the vegetation in each observation time window. The growth potential G(t) reflects the comprehensive growth potential of the vegetation in the current observation time window, and is a real number between 0 and 1. When G(t) approaches 1, the vegetation is in the best growth state, and both the photosynthesis efficiency and the water use efficiency reach the optimum; when G(t) is close to 0, it indicates that the growth of the vegetation is severely inhibited, and the photosynthesis rate and the biomass accumulation rate are significantly reduced. The model combines the dynamic changes of various environmental factors such as light, temperature, water, and soil conductivity through a nonlinear multi-parameter coupling mechanism, so as to achieve an accurate evaluation of the growth potential of the vegetation in different time windows.

[0066] Photosynthetically active radiation PAR(t), as the primary factor for calculating the growth potential, represents the effective light quanta available for photosynthesis received by the vegetation in the t-th observation time window, with the unit of μmol / m 2 ·s. Photosynthetically active radiation directly affects the rate of photosynthesis and the biomass accumulation rate, and its change trend is positively correlated with the photosynthesis efficiency of the vegetation. The system collects PAR(t) data in different observation time windows through a quantum sensor, and combines the photosynthetic photon flux density PPFD for dynamic light response correction. When PAR(t) increases, the photosynthesis rate increases accordingly, and the system will automatically increase the evaluation value of the growth potential. While at the light saturation constant KP When reaching the saturation point nearby, the increasing trend of the growth potential slows down, and the system automatically introduces a light saturation inhibition term to correct the saturation effect of the photosynthesis potential.

[0067] The temperature factor is modeled through a non-linear temperature response function in the calculation of the growth potential. This function comprehensively considers the minimum growth temperature T min , the optimum growth temperature T opt and the maximum growth temperature T max . Among the temperature impact factors, describes the linear promotion effect of temperature on the photosynthesis efficiency between T min and T opt . When the temperature is lower than T min , photosynthesis stops and the growth potential tends to zero; while when the temperature gradually rises to T opt , the photosynthesis rate gradually approaches the maximum value. The term in the formula is a temperature decay function based on the Gaussian distribution. When the temperature exceeds T opt , this term shows an exponential decline, simulating the non-linear decay characteristic of the photosynthesis rate after the temperature exceeds the optimum range.

[0068] The soil moisture content SM(t) is a key factor affecting the water use efficiency and growth state of vegetation. Its value is obtained by measuring with a soil moisture sensor and is calculated as a ratio with the field capacity FC to form a water availability correction term When the soil moisture approaches the field capacity, the water availability reaches the maximum value and the photosynthesis potential of the vegetation is fully exerted; while when the soil moisture is insufficient, this term gradually decreases, thus inhibiting the photosynthesis rate and leading to a decline in the growth potential. The field capacity FC is dynamically set according to different soil types. Among them, the FC value range of sandy soil is from 0.05 to 0.12, that of loam is from 0.20 to 0.35, that of clay loam is from 0.30 to 0.45, and that of clay is from 0.35 to 0.50. The system automatically selects the corresponding FC value range according to the soil type in different vegetation growth areas and dynamically corrects it in combination with the change of water content to ensure that the influence of water availability on the growth potential is accurately reflected.

[0069] The soil conductivity EC(t) is an important parameter affecting the salt stress of vegetation, with the unit of dS / m and is obtained by measuring with a conductivity sensor. Excessive conductivity will lead to the accumulation of soil salts, thus inhibiting the water absorption and photosynthesis rate of vegetation. The salt inhibition term The growth potential is corrected by dynamically adjusting the photosynthesis efficiency of vegetation under different salinity environments. The value range of the salinity stress coefficient α is from 0.2 to 0.4. Different vegetation species have different sensitivities to salinity stress. Therefore, within each observation time window, the system dynamically adjusts the value range of α according to the vegetation type, so as to achieve adaptive correction of the salinity inhibition effect. When EC(t) increases, the value of this term gradually decreases, thus significantly reducing the photosynthesis efficiency and causing the growth potential G(t) to decline.

[0070] Furthermore, if the soil type where the vegetation is located is sandy soil, the value range of FC is from 0.05 to 0.12; if the soil type where the vegetation is located is loam, the value range of FC is from 0.20 to 0.35; if the soil type where the vegetation is located is clay loam, the value range of FC is from 0.30 to 0.45; if the soil type where the vegetation is located is clay, the value range of FC is from 0.35 to 0.50.

[0071] In the garden vegetation growth state monitoring system based on nonlinear time series prediction of the present invention, the field capacity (FC) is one of the crucial parameters in the dynamic evaluation of the photosynthesis efficiency and water use efficiency of vegetation. The field capacity FC represents the maximum amount of water that the soil can hold after excluding gravitational water and is an important factor affecting soil water availability and vegetation growth state. Since there are significant differences in the water holding capacity of different soil types, the system needs to dynamically adjust the value range of FC according to the soil type where the vegetation is located, so as to accurately correct the soil water effect in the growth potential calculation model.

[0072] When the system models the vegetation growth potential G(t), a water availability correction term is introduced to simulate the dynamic changes in the photosynthesis rate of vegetation under different water conditions. When the soil moisture SM(t) approaches FC, the water use efficiency of the vegetation reaches the optimal state and the photosynthesis potential is the largest; while when the soil moisture is lower than FC, the vegetation enters the water stress state, the growth potential G(t) drops rapidly, and the water use efficiency also decreases significantly. Therefore, the system needs to dynamically assign values to FC according to different soil types within each observation time window to ensure the accuracy of the growth potential calculation results.

[0073] When the soil type where the vegetation is located is sandy soil, the value range of the field capacity (FC) is from 0.05 to 0.12. Sandy soil has relatively large particle sizes and low porosity, resulting in poor water retention capacity. After precipitation or irrigation, water quickly penetrates into the deep soil. Therefore, under sandy soil conditions, the vegetation's water acquisition ability is weak, and the soil moisture SM(t) often fluctuates significantly. In this case, the system dynamically corrects the water effect by setting FC between 0.05 and 0.12, and optimizes the growth potential assessment results by combining the changes in the water transpiration rate and photosynthesis potential.

[0074] When the soil type where the vegetation is located is loam, the value range of the field capacity (FC) is from 0.20 to 0.35. Loam is a soil type with relatively uniform texture, having good water retention capacity and air permeability. Its pore structure helps the uniform distribution of water, enabling the vegetation to continuously absorb water from the soil and maintain photosynthesis efficiency. In the loam environment, the system uses the range from 0.20 to 0.35 as the dynamic value range of FC, and adjusts the water use correction factor in the growth potential calculation model by combining the changes in soil water content and vapor pressure deficit (VPD), thereby achieving an accurate assessment of the photosynthesis potential.

[0075] When the soil type where the vegetation is located is clay loam, the value range of the field capacity (FC) is from 0.30 to 0.45. Clay loam has a finer texture and better water retention ability than loam. However, due to its smaller pores, its water permeability is poor, and it is prone to form water stress in a wet state, inhibiting root respiration and photosynthesis efficiency. Under clay loam conditions, the system sets FC between 0.30 and 0.45, dynamically adjusts the water availability correction term in the growth potential assessment model to simulate the non-linear change trend of the vegetation growth state under different water conditions. At the same time, the system combines the dynamic changes of soil conductivity EC(t) and relative humidity RH(t) to optimize the water - photosynthesis coupling mechanism, ensuring the accuracy of the growth potential assessment results.

[0076] When the soil type where the vegetation is located is clay, the value range of the field capacity (FC) is from 0.35 to 0.50. Due to its fine particles and low porosity, clay has extremely strong water retention ability. However, due to its poor water permeability, it is prone to form a situation of insufficient oxygen supply in a high water state, thus inhibiting root respiration and photosynthesis efficiency of the vegetation. Under clay conditions, the system sets FC between 0.35 and 0.50, dynamically models by combining multiple parameters such as soil water content, soil conductivity, and temperature, and adjusts the water use factor in the growth potential assessment model through a non-linear water stress response mechanism to ensure the accuracy of growth potential prediction under different water conditions.

[0077] The process of the system dynamically setting FC under different soil types is achieved through a non-linear multi-parameter optimization mechanism and an adaptive time window adjustment strategy. Within each observation time window, the system measures SM(t) through soil moisture sensors and combines environmental parameters such as soil conductivity EC(t), temperature and humidity T(t), and RH(t) to automatically identify the soil type where the vegetation is located and dynamically assign the value range of FC according to the soil type. When the system identifies a change in soil type, it automatically adjusts the moisture correction term in the growth potential assessment model to ensure the dynamic adaptability of growth state assessment.

[0078] Furthermore, the vegetation photosynthetic efficiency assessment part assesses the photosynthesis efficiency of the vegetation in each observation time window as follows:

[0079]

[0080] where P(t) represents the photosynthesis rate in the t-th observation time window, with the unit of μmolCO2 / m 2 ·s; K N represents the nitrogen semi-saturation constant, with the unit of mg / kg, and the value range is from 25 to 40; β represents the pH deviation sensitivity coefficient, and the value range is from 0.3 to 0.6.

[0081] First, This part is the dynamic modeling of photosynthesis based on the light response curve. The photosynthetic photon flux density PPFD(t) is a measure of the effective photon flux received by the plant leaves in the t-th observation time window, with the unit of μmol / m 2 ·s. P max represents the maximum photosynthesis rate of the vegetation when the light intensity reaches saturation, with the unit of μmolCO2 / m 2 ·s. This part of the model is based on the Michaelis-Menten light response model. When PPFD(t) increases, the photosynthesis rate also increases, but when PPFD(t) approaches the light saturation constant K P , the photosynthesis rate gradually tends to saturation. K P represents the light saturation constant, and the value range is from 200 to 500 μmol / m 2 ·s, which is obtained by measuring the light response curve. As PPFD(t) increases, it shows an increasing trend, but when PPFD(t) is much larger than K P , the photosynthesis efficiency tends to saturation and the growth rate reaches the maximum.

[0082] Secondly, This part is the dynamic correction term of soil nitrogen nutrients on the photosynthesis rate. N(t) represents the soil nitrogen content, with the unit of mg / kg, which is measured by a soil nitrogen content sensor. Nitrogen is an important nutrient for vegetation to carry out photosynthesis and protein synthesis. When the soil nitrogen content increases, the photosynthesis rate correspondingly increases. However, when the nitrogen content exceeds a certain level, the photosynthesis rate tends to be stable. K N represents the nitrogen half-saturation constant, with a value range of 25 to 40 mg / kg, which is the nitrogen concentration required for vegetation to reach the half-saturation photosynthesis rate. When N(t) is much smaller than K N , the nitrogen supply becomes the bottleneck restricting photosynthesis, approaches zero, thus significantly reducing the photosynthesis rate. When N(t) gradually increases, the photosynthesis rate approaches the saturation state, and the nitrogen supply no longer becomes a limiting factor.

[0083] This part is the linear correction term of soil pH change on the photosynthesis rate. pH(t) represents the acidity and alkalinity of the soil within the t-th observation time window, which is measured in real time by a pH sensor. pH min and pH opt respectively represent the minimum pH and the optimal pH range for vegetation growth. When pH(t) is near pH opt , the photosynthesis rate is at the best level. When pH(t) approaches or is lower than pH min , the nutrient absorption and root growth of vegetation are inhibited, thus significantly reducing the photosynthesis efficiency. This term simulates the change trend of vegetation photosynthesis under different pH conditions through a linear proportional relationship, thereby achieving a basic correction of pH changes.

[0084] This term is the non-linear inhibition term of photosynthesis after the soil pH deviates from the optimal range. As the deviation between pH(t) and pH opt increases, the photosynthesis rate shows an exponential decline. β represents the pH deviation sensitivity coefficient, with a value range of 0.3 to 0.6, which is obtained by fitting the pH response curve. For vegetation with high pH sensitivity, the value of β is relatively high, meaning that the photosynthesis rate is extremely sensitive to pH deviation changes. For vegetation with strong acid and alkali tolerance, the value of β is relatively low, indicating that the photosynthesis has stronger tolerance to pH deviation. When the deviation of pH(t) from pH opt increases, the exponential term rapidly decreases, thus significantly reducing the photosynthesis rate.

[0085] Furthermore, the vegetation photosynthetic efficiency evaluation part evaluates the water use efficiency of vegetation in each observation time window as:

[0086]

[0087] Among them, W(t) represents the water use efficiency of the t-th observation time window, with the unit of mmolCO2 / molH2O; γ represents the water stress response coefficient, and its value range is from 2.5 to 4.0; δ represents the temperature deviation sensitivity coefficient, with the unit of 1 / ℃, and its value range is from 0.08 to 0.12.

[0088] is the transpiration-vapor pressure difference feedback term of water use efficiency, where Tr(t) represents the leaf transpiration rate, with the unit of mmolH2O / m 2 ·s, which is measured in real time by a porometer. The transpiration rate reflects the amount of water lost by the vegetation through leaf transpiration and is one of the key parameters affecting water use efficiency. VPD(t) represents the atmospheric vapor pressure difference, with the unit of kPa, which is the difference between the air saturation vapor pressure and the current vapor pressure and is calculated after collecting temperature and humidity data by a temperature and humidity sensor. When VPD(t) increases, the transpiration rate accelerates, but the water use efficiency decreases, resulting in a reduction in the water use efficiency of the vegetation. Therefore, this term reflects the dynamic balance relationship between the transpiration of the vegetation and water consumption. When VPD(t) is too large, the water loss increases, leading to a decrease in W(t), while when VPD(t) is moderate, the water loss rate slows down, thereby improving the water use efficiency.

[0089] This part is the water use correction term based on soil water stress. FC represents the field capacity, and its value range is dynamically adjusted according to different soil types. The value range for sandy soil is from 0.05 to 0.12, for loam soil is from 0.20 to 0.35, for clay loam soil is from 0.30 to 0.45, and for clay soil is from 0.35 to 0.50. SM(t) represents the soil water content within the t-th observation time window, which is measured by a soil moisture sensor, with the unit of m 3 / m 3 . When the soil water SM(t) is much lower than FC, the vegetation is in a water stress state, and the photosynthesis rate and water use efficiency decrease significantly. The water stress response coefficient γ has a value range of 2.5 to 4.0, indicating the sensitivity of soil water stress to the water use efficiency of the vegetation. As SM(t) decreases, the value of gradually increases, resulting in a rapid decrease, thereby inhibiting the increase of W(t). When SM(t) approaches FC, this term approaches 1, the water stress effect is minimal, and the water use efficiency of the vegetation reaches the optimum.

[0090] is the relative humidity correction term. RH(t) represents the relative air humidity in the t-th observation time window, which is measured by a humidity sensor. Relative humidity has a direct impact on the vegetation transpiration rate and water use efficiency. When RH(t) approaches 100%, the leaf transpiration slows down and the vegetation water use efficiency increases; while when RH(t) is low, the transpiration rate accelerates and the water use efficiency decreases. Therefore, this term simulates the change trend of water use efficiency under different humidity conditions and makes dynamic corrections when the humidity deviates from the optimal range.

[0091] This term is the water use inhibition factor based on temperature deviation. T(t) represents the ambient temperature in the t-th observation time window, in °C, which is measured by a temperature sensor. T opt is the optimal growth temperature of the vegetation. The value of T opt varies for different vegetation types and is usually between 20°C and 30°C. The temperature deviation sensitivity coefficient δ ranges from 0.08 to 0.12 and is used to describe the degree of inhibition of water use efficiency when the temperature deviates from T opt . When T(t) deviates from T opt , this term decreases exponentially, simulating the inhibitory effect of temperature deviation from the optimal range on water use efficiency. As the temperature deviation increases, decreases rapidly, thus reducing the value of W(t), reflecting the non-linear attenuation characteristics of water use efficiency under temperature stress conditions.

[0092] Furthermore, the vegetation growth status assessment part assesses the vegetation biomass in the t-th observation time window as:

[0093]

[0094] where B(t) is the vegetation biomass in the t-th observation time window; B(t - 1) is the vegetation biomass in the (t - 1)-th observation time window; s is the time integration variable; R0 represents the basic respiration rate of the vegetation, with a value range of 0.01 to 0.02; Q 10 is the respiration temperature sensitivity coefficient, with a value of 2.0.

[0095] In the monitoring system for the growth status of garden vegetation based on non-linear time series prediction proposed by the present invention, in order to accurately reveal the law of the evolution of vegetation biomass over time, an integration model that comprehensively considers the photosynthesis increment and the basic respiratory metabolism loss within each observation time window is specifically designed. Starting from the biomass of the previous time window, this model uses the product of the growth potential, photosynthesis rate, and water use efficiency to describe the contribution of photosynthesis to the accumulation of organic carbon within the current window, and then subtracts the additional metabolic consumption triggered by basic respiration and temperature increase from this accumulated value, so as to obtain a comprehensive result reflecting the difference between the net increment of photosynthesis and the negative increment of respiratory metabolism.

[0096] In the specific implementation of the system, this difference is superimposed on the biomass value of the previous window through time integration, and the biomass level of the current window can be calculated. If it is roughly expressed in the form of an integral expression, it can be understood as the biomass at the moment of the previous observation window plus the positive increment of photosynthesis from the previous moment to the current moment minus the inhibitory effect of temperature-sensitive respiratory metabolism on biomass. Here, the growth potential usually takes values between 0 and 1. The closer it is to 1, the greater the growth potential that the vegetation can exert in the environment it is in. If factors such as soil moisture, salt content, and temperature are all in the optimal range, this potential value will approach its upper limit, thus driving the rapid accumulation of organic carbon; on the contrary, if facing high salinity or extreme temperature, this value will decrease accordingly, resulting in a slowdown in the growth rate. The photosynthesis rate is jointly determined by the photon flux, soil nitrogen content, pH deviation degree, etc. Once the light saturation is reached and the nitrogen supply is sufficient, the upper limit of photosynthesis can be fully stimulated. If the soil nitrogen level is insufficient or the pH seriously deviates from the optimal range, it will quickly cause inhibition and lead to a decrease in the net production rate. The water use efficiency is also a key item affecting the photosynthesis increment. The system combines the ratio of the leaf transpiration rate to the vapor pressure deficit of the atmosphere and corrects the actual contribution of this ratio through the soil water stress response coefficient. If the soil water content is much lower than the field capacity at this time, the water use efficiency will be significantly reduced, making photosynthesis unable to continue efficiently.

[0097] When the curve obtained by multiplying these three elements of photosynthesis is relatively ideal, the carbon accumulation of vegetation shows a strong positive contribution within this time window. However, the system does not only consider positive accumulation, but also introduces a negative regulation mechanism of basal respiration to deduct the growth amount. The basal respiration rate is often selected within a relatively small numerical range, and the exponential amplification effect of the temperature sensitivity coefficient is used to reflect the stimulation effect of high temperature on respiratory metabolism. Whenever the environmental temperature is higher than the standard reference temperature, the corresponding exponential term will increase the overall respiration rate, which means that leaves and other organs will consume more accumulated organic matter during the maintenance of life activities, thus inhibiting the growth of biomass. When this respiratory demand becomes too strong, no matter how high the photosynthesis increment is, it is difficult to completely compensate for the metabolic loss, making the biomass curve may show a slowdown or even a temporary decline in growth.

[0098] During the calculation process, the system will conduct dynamic tracking at the integral level for temperature sensitivity, that is, estimate the magnification of the temperature deviation on respiration amplification in real time throughout the entire time period, accumulate the net changes in all micro time periods through an integral form from the starting moment to the current moment, and then add it to the biomass of the previous window at the end to obtain a new biomass estimate. If the soil moisture content also continues to be low during this stage, both the growth potential and water use efficiency will be inhibited, further increasing the relative weight of the respiration term in the integral process, resulting in a significant reduction in the actual increase in biomass. Such a strategy of "positive integral photosynthetic accumulation and negative integral respiratory metabolism" can map the dynamic changes of various environmental factors onto the biomass curve and continuously iterate in the time series prediction algorithm, enabling the system to take into account the coupling relationships in different growth stages. When there is sufficient data support, the system can also predict in advance the possible trend of biomass in the next few days and take irrigation or shading strategies in a timely manner when threats such as high temperature and drought are detected to maintain the healthy growth of vegetation to the greatest extent. This time integral model of vegetation biomass not only reflects the synergistic or antagonistic relationship between photosynthesis efficiency, water use efficiency, and temperature-sensitive respiration function, but also ensures that in a complex environment with the superposition of multiple interaction factors, the growth trend of vegetation can be truly reflected and intervened in a timely manner.

[0099] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. A monitoring system for the growth state of garden vegetation based on non-linear time series prediction, characterized in that, Including: A data acquisition part, which is used to acquire the growth characteristic parameters of each type of vegetation in the garden in each observation time window, and acquire the environmental parameters of the environment where the vegetation is located in each observation time window; A vegetation photosynthetic efficiency evaluation part, which is used to evaluate the vegetation growth potential, photosynthesis efficiency and water use efficiency of the vegetation in each observation time window according to the growth characteristic parameters and environmental parameters; A vegetation growth state evaluation part, which is used to evaluate the vegetation biomass increment of the current observation time window compared with the previous observation time window according to the vegetation growth potential, photosynthesis efficiency and water use efficiency, and accumulate the vegetation biomass of the previous observation time window to obtain the vegetation growth amount of the current time window and predict the vegetation biomass of the next observation time window.

2. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction according to claim 1, characterized in that, In the first observation time window, the biomass of each type of vegetation is set to an initial value corresponding to that type of vegetation; the vegetation growth amount of the second observation time window is obtained by accumulating the vegetation growth amount of the first observation time window and the vegetation biomass increment between the first observation time window and the second observation time window, and so on, to obtain the vegetation growth amount of each observation time window and predict the vegetation growth amount of the next observation time window of the current observation time window.

3. The monitoring system for the growth state of garden vegetation based on non - linear time - series prediction as claimed in claim 2, wherein, Growth characteristic parameters include: leaf transpiration rate Tr(t), in mmolH2O / m 2 ·s, vegetation light saturation constant K P , in μmol / m 2 ·s; vegetation maximum photosynthetic rate P max , in μmolCO2 / m 2 ·s; vegetation minimum growth temperature T min , in °C; vegetation optimum growth temperature T opt , in °C; vegetation maximum growth temperature T max , in °C; vegetation minimum growth pH value pH min ; vegetation optimum growth pH value pH opt ; among them, the leaf transpiration rate Tr(t) is obtained by measuring with a porometer; the vegetation light saturation constant K P is determined by a light response curve, and the value range is from 200 to 500; the vegetation maximum photosynthetic rate P max is determined by a photosynthesis meter; t represents the t-th observation time window; the length of each observation time window is equal.

4. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction as claimed in claim 3, wherein Environmental parameters include: Photosynthetically Active Radiation PAR(t), with the unit of μmol / m 2 ·s, which is measured by a quantum sensor; Soil Moisture SM(t), which is measured by a soil moisture sensor; Soil Electrical Conductivity EC(t), with the unit of dS / m, which is measured by a conductivity sensor; Soil Nitrogen Content N(t), with the unit of mg / kg, which is measured by a soil nutrient sensor; Photosynthetic Photon Flux Density PPFD(t), with the unit of μmol / m 2 ·s, which is measured by a quantum sensor; Soil pH value pH(t), which is measured by a pH sensor; Vapor Pressure Deficit VPD(t), with the unit of kPa, which is calculated after being measured by a temperature and humidity sensor; Relative Humidity RH(t), which is measured by a humidity sensor; Current Temperature T(t), with the unit of °C, which is measured by a temperature sensor.

5. The monitoring system for the growth state of garden vegetation based on non - linear time - series prediction according to claim 4, characterized in that, The vegetation photosynthetic efficiency evaluation part evaluates the vegetation growth potential of the vegetation in each observation time window as: Wherein, G(t) represents the vegetation growth potential of the t-th observation time window, and the value is a real number greater than 1 and less than 0; FC is the field water holding capacity, and its value range is different according to the different soil types where the vegetation is located; α is the salt stress coefficient, with the unit of m / dS, and the value range is from 0.2 to 0.

4.

6. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction according to claim 5, characterized in that If the soil type where the vegetation is located is sandy soil, the value range of FC is from 0.05 to 0.12; if the soil type where the vegetation is located is loam, the value range of FC is from 0.20 to 0.35; if the soil type where the vegetation is located is clay loam, the value range of FC is from 0.30 to 0.45; if the soil type where the vegetation is located is clay, the value range of FC is from 0.35 to 0.

50.

7. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction according to claim 6, wherein The vegetation photosynthetic efficiency evaluation part evaluates the photosynthesis efficiency of the vegetation in each observation time window as: Among them, P(t) represents the photosynthesis rate at the t-th observation time window, with the unit of μmolCO2 / m 2 ·s; K N represents the nitrogen semi-saturation constant, with the unit of mg / kg and the value range from 25 to 40; β represents the pH deviation sensitivity coefficient, and the value range is from 0.3 to 0.

6.

8. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction according to claim 7, wherein, The vegetation photosynthetic efficiency evaluation part evaluates the water use efficiency of the vegetation in each observation time window as: Wherein, W(t) represents the water use efficiency of the t-th observation time window, with the unit of mmolCO2 / molH2O; γ represents the water stress response coefficient, and the value range is from 2.5 to 4.0; δ represents the temperature deviation sensitivity coefficient, with the unit of 1 / ℃, and the value range is from 0.08 to 0.

12.

9. The monitoring system for the growth state of garden vegetation based on non-linear time series prediction according to claim 8, characterized in that, The vegetation growth state evaluation part evaluates the vegetation biomass of the t-th observation time window as: Among them, B(t) is the vegetation biomass of the t-th observation time window; B(t - 1) is the vegetation biomass of the (t - 1)-th observation time window; s is the time integration variable; R0 represents the basal respiration rate of vegetation, and its value range is from 0.01 to 0.02; Q 10 is the respiration temperature sensitivity coefficient, and its value is 2.0.

Images