Intelligent agricultural environment monitoring method and light supplement adjusting system

Abstract

The invention discloses an intelligent agricultural environment monitoring method and a light supplement adjusting system, and particularly relates to the technical field of agricultural monitoring. An edge computing device is utilized to extract photosynthesis efficiency irregular features and illumination intensity change rate fluctuation features in a data processing module, and a delay degree evaluation module evaluates the delay abnormal degree between real-time environment data and a light supplement adjustment decision. The light supplementing control module is used for deeply analyzing delay data in a fixed time period and dynamically adjusting the illumination intensity and spectral distribution of a light supplementing lamp, so that the problem of insufficient response caused by high delay and single parameter analysis in a traditional method is effectively solved; the accuracy and the real-time performance of agricultural environment monitoring are remarkably improved, the crop growth condition is optimized, and therefore the agricultural production efficiency and the crop quality are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural monitoring, and particularly to an intelligent agricultural environment monitoring method and a light supplement adjustment system. Background Art

[0002] With the rapid development of modern agriculture, intelligence and precision have become important directions for improving agricultural yield and efficiency. Traditional agricultural environment monitoring usually relies on manual inspections and simple manual control devices, making it difficult to achieve real-time and precise regulation of the crop growth environment. In recent years, the progress of Internet of Things (IoT), artificial intelligence (AI), and sensor technologies has enabled the wide application of data-driven agricultural environment monitoring and light supplement adjustment systems. These systems can collect environmental data such as temperature, humidity, light intensity, and carbon dioxide concentration through sensor networks, and optimize the growth conditions of crops by combining intelligent control algorithms. However, if there is an excessive delay between real-time environmental data and light supplement adjustment decisions, it may lead to failure to respond in a timely manner to dynamic environmental changes (such as sudden cloudy days or drastic climate changes), which may in turn affect the growth quality of crops. This is because existing systems mostly adopt a distributed data processing architecture, where data is collected from sensors and then transmitted through a local gateway to the cloud for processing. After the cloud analysis is completed, the light supplement adjustment instructions are then returned to the device. The high latency and network instability in this process make it difficult for the system to adapt to the rapidly changing agricultural environment. In addition, traditional light supplement decision algorithms are only based on historical data or threshold judgments of single parameters, lacking the ability to dynamically correlate and analyze real-time multi-dimensional data and make predictions, thus making it difficult to quickly formulate precise light supplement strategies during drastic environmental changes. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent agricultural environment monitoring method and a light supplement adjustment system to solve the deficiencies in the background art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: An intelligent agricultural environment monitoring method, comprising the following steps:

[0005] S1: Collect real-time multi-dimensional data of the agricultural environment through a sensor network, where the multi-dimensional data includes environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency;

[0006] S2: Use an edge computing device to preprocess the collected data, and extract the irregular features of photosynthesis efficiency and the fluctuating features of the light intensity change rate from the preprocessed collected data;

[0007] S3: According to the extracted irregular features of photosynthesis efficiency and the fluctuating features of the light intensity change rate, evaluate the degree of delay anomaly between the real-time environmental data and the light supplement adjustment decision.

[0008] S4: Dynamically optimize the supplementary lighting adjustment strategy according to the degree of latency anomaly, including adjusting the execution time window of the supplementary lighting adjustment to ensure real-time performance in the case of mild latency; and enabling a supplementary lighting prediction algorithm based on historical data and a prediction model to generate and issue supplementary lighting instructions in advance in the case of severe latency.

[0009] Preferably, in S2, after analyzing the irregular characteristics of the photosynthesis efficiency extracted, generate an irregular index of photosynthesis efficiency. The method for obtaining the irregular index of photosynthesis efficiency is as follows:

[0010] Use the collected photosynthesis efficiency time series X(s) as the input signal, and the time series is defined as: X = {x 1 , x 2 ,..., x N}; where N is the total number of data points, and x N is the photosynthesis efficiency value at the Nth moment. Select the wavelet basis function for analyzing the photosynthesis efficiency data, perform hierarchical decomposition on the signal X(s), and extract the low-frequency part A j and the high-frequency part D i , and the formula is: X(s) = A i + Select the decomposition level j according to the signal length and sampling frequency, extract the energy of all high-frequency components D i to measure the irregularity of the signal. The high-frequency energy E HighFreq calculation formula is: where D i [k] is the coefficient of the high-frequency signal at the i-th layer at the k-th sampling point, and N i is the number of signal points at the i-th layer; at the same time, calculate the total energy E Total of the complete signal, and the expression is: Calculate the irregular index of photosynthesis efficiency, In the formula, PEII is the irregular index of photosynthesis efficiency.

[0011] Preferably, in S2, after analyzing the fluctuation characteristics of the light intensity change rate extracted, generate a light intensity change rate fluctuation index. The method for obtaining the light intensity change rate fluctuation index is as follows:

[0012] The input is the light intensity time series L(t), which represents the change of the collected light intensity over time. The time series is defined as: L = {L 1 , L 2 ,..., L M}; where: L M is the light intensity value at the Mth moment, M is the total number of data points, and the first-order difference reflects the change rate of the light intensity between adjacent moments. The calculation formula is: △L t = L t-L t-1 where \(t = 2, 3,\cdots,M\); where \(\Delta L\) t is the change rate of light intensity at the \(t\)-th moment, \(L\) t and \(L\) t-1 are the light intensity values at the current moment and the previous moment respectively; the first-order difference sequence \(\Delta L=\{\Delta L\) 2 , \(\Delta L\) 3 ,\(\cdots,\Delta L\) N \}; the second-order difference reflects the fluctuation degree of the change rate of light intensity, and the calculation formula is: \(\Delta\) 2 L t =\(\Delta L\) t -\(\Delta L\) t-1 , where \(t = 3, 4,\cdots,M\); where: \(\Delta\) 2 L t is the change amount of the first-order difference at the \(t\)-th moment, \(\Delta L\) t and \(\Delta L\) t-1 are the first-order difference values at the current moment and the previous moment respectively, and the second-order difference sequence is \(\Delta\) 2 L = \{\Delta\) 2 L 3 , \(\Delta\) 2 L 4 ,\(\cdots,\Delta\) 2 L N \}; the fluctuation index is calculated according to the statistical characteristics of the second-order difference, and the expression is: In the formula, is the mean value of the second-order difference, and IRFI is the light intensity change rate fluctuation index.

[0013] Preferably, in S3, according to the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of the light intensity change rate, evaluate the delay anomaly degree between the real-time environmental data and the supplementary light adjustment decision;

[0014] Convert the photosynthesis efficiency irregularity index and the light intensity change rate fluctuation index into a comprehensive feature vector, use the comprehensive feature vector as the input of the machine learning model, use the machine learning model to predict the delay anomaly degree value label between the real-time environmental data and the supplementary light adjustment decision for each group of comprehensive feature vectors as the prediction target, and use minimizing the sum of the prediction errors of the delay anomaly degree value labels between all real-time environmental data and the supplementary light adjustment decision as the training target to train the machine learning model until the sum of the prediction errors reaches convergence and stop the model training, and determine the delay anomaly degree value between the real-time environmental data and the supplementary light adjustment decision according to the model output result, where the machine learning model is a polynomial regression model.

[0015] Preferably, in S4, dynamically optimize the supplementary light adjustment strategy according to the delay anomaly degree, specifically:

[0016] Compare the delay anomaly degree value between the obtained real-time environmental data and the light supplement adjustment decision with the reference threshold of the delay anomaly degree value preset under normal conditions according to historical data. If the delay anomaly degree value between the real-time environmental data and the light supplement adjustment decision is greater than or equal to the reference threshold of the delay anomaly degree value, it indicates that the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision is high. At this time, generate a warning signal, and classify the delay situation between the real-time environmental data and the light supplement adjustment decision as severe delay, and immediately enable the light supplement prediction algorithm based on historical data and prediction model to generate and issue the light supplement instruction in advance; if the delay anomaly degree value between the real-time environmental data and the light supplement adjustment decision is less than the reference threshold of the delay anomaly degree value, it indicates that the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision is low. At this time, do not generate a warning signal, and classify the delay situation between the real-time environmental data and the light supplement adjustment decision as mild delay. At this time, it is necessary to adjust the execution time window of the light supplement adjustment to ensure real-time performance.

[0017] The present invention also provides an intelligent agricultural environment light supplement adjustment system, including a sensor module, a data processing module, a delay degree evaluation module, an optimization module, and a light supplement control module;

[0018] Sensor module: Collect real-time multi-dimensional data of the agricultural environment through a sensor network. The multi-dimensional data includes environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency;

[0019] Data processing module: Use edge computing devices to preprocess the collected data, and extract the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of the light intensity change rate in the preprocessed collected data;

[0020] Delay degree evaluation module: Evaluate the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision according to the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of the light intensity change rate;

[0021] Optimization module: Dynamically optimize the light supplement adjustment strategy according to the delay anomaly degree, including adjusting the execution time window of the light supplement adjustment to ensure real-time performance in the case of mild delay; in the case of severe delay, enable the light supplement prediction algorithm based on historical data and prediction model to generate and issue the light supplement instruction in advance;

[0022] Light supplement control module: Used to deeply analyze the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision within a fixed time period according to the optimized adjustment strategy, and dynamically adjust the light intensity and spectral distribution of the light supplement lamp according to the analysis result.

[0023] Preferably, in the supplementary light control module, the delay anomaly degree between the real-time environmental data and the supplementary light adjustment decision within a fixed time period is deeply analyzed, and the light intensity and spectral distribution of the supplementary light are dynamically adjusted according to the analysis results. Specifically:

[0024] After dynamically optimizing the supplementary light adjustment strategy, collect the delay anomaly degree values between the real-time environmental data generated in subsequent fixed time periods and the supplementary light adjustment decision, establish a corresponding data set, calculate the mean and standard deviation of the data set, and after analyzing it, dynamically adjust the light intensity and spectral distribution of the supplementary light according to the analysis results.

[0025] Preferably, if the mean value of the delay anomaly degree values in the data set is greater than or equal to the reference threshold of the mean value of the delay anomaly degree values, and the standard deviation of the delay anomaly degree values is less than the reference threshold of the standard deviation of the delay anomaly degree values, it indicates that the delay anomaly degree is high and the fluctuation is small, indicating that the supplementary light adjustment strategy has a systematic lag. At this time, it is necessary to increase the light intensity of the supplementary light or adjust the spectral distribution to improve the photosynthesis efficiency of the crops and optimize the system response speed;

[0026] If the mean value of the delay anomaly degree values is greater than or equal to the reference threshold of the mean value of the delay anomaly degree values, and the standard deviation of the delay anomaly degree values is greater than or equal to the reference threshold of the standard deviation of the delay anomaly degree values, it indicates that the delay anomaly degree is high and the fluctuation is significant, indicating that the supplementary light adjustment strategy cannot adapt to the rapid change of environmental light. At this time, it is necessary to enable the supplementary light prediction algorithm, strengthen the real-time monitoring and rapid response capabilities, and reduce the delay and fluctuation;

[0027] If the mean value of the delay anomaly degree values is less than the reference threshold of the mean value of the delay anomaly degree values, and the standard deviation of the delay anomaly degree values is greater than or equal to the reference threshold of the standard deviation of the delay anomaly degree values, it indicates that the delay anomaly degree is low but the fluctuation is significant, and the adaptability to environmental changes is insufficient. At this time, it is necessary to dynamically optimize the spectral distribution, improve the light uniformity and the stability of crop growth, and reduce the system adjustment frequency;

[0028] If the mean value of the delay anomaly degree values is less than the reference threshold of the mean value of the delay anomaly degree values, and the standard deviation of the delay anomaly degree values is less than the reference threshold of the standard deviation of the delay anomaly degree values, it indicates that the delay anomaly degree is low and the fluctuation is small, indicating that the supplementary light adjustment strategy is efficient and stable. At this time, it is necessary to maintain the current supplementary light intensity and spectral distribution, and at the same time monitor regularly to ensure that the system continues to operate efficiently.

[0029] In the above technical solution, the technical effects and advantages provided by the present invention:

[0030] 1. The present invention significantly improves the real-time performance and accuracy of the agricultural environment monitoring and supplementary lighting adjustment system by introducing multi-dimensional data collection, edge computing processing, and dynamic optimization of the supplementary lighting adjustment strategy. Aiming at the problem of insufficient response caused by high latency and single-parameter analysis in traditional systems, the present invention extracts the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of the change rate of light intensity, uses a machine learning model to evaluate the degree of latency abnormality, and dynamically optimizes the supplementary lighting strategy according to real-time environmental data, so as to quickly adapt to environmental changes, accurately control the light intensity and spectral distribution, and optimize the growth conditions of crops.

[0031] 2. The present invention further improves the stability and adaptability of the supplementary lighting system by deeply analyzing the mean and standard deviation of the latency abnormality degree values. The prediction algorithm is enabled to enhance the system response ability in the case of high latency and significant fluctuations, and the high-efficiency operation state is maintained in the case of low latency and small fluctuations. This intelligent control mechanism effectively improves agricultural production efficiency, reduces resource waste, and significantly improves the photosynthesis efficiency and growth quality of crops, contributing to the high-yield, high-efficiency, and sustainable development of modern agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0033] Figure 1 It is a flowchart of the method of the present invention.

[0034] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0036] Example 1. Please refer to Figure 1 As shown, an intelligent agricultural environment monitoring method in this embodiment includes the following steps:

[0037] S1: Collect real-time multi-dimensional data of the agricultural environment through a sensor network. The multi-dimensional data includes environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency;

[0038] S2: Use edge computing devices to preprocess the collected data, and extract the irregular features of photosynthesis efficiency and the fluctuating features of the light intensity change rate from the preprocessed collected data;

[0039] S3: According to the extracted irregular features of photosynthesis efficiency and the fluctuating features of the light intensity change rate, evaluate the delay anomaly degree between the real-time environmental data and the supplementary lighting adjustment decision;

[0040] S4: Dynamically optimize the supplementary lighting adjustment strategy according to the delay anomaly degree, including adjusting the execution time window of the supplementary lighting adjustment to ensure real-time performance in the case of mild delay; in the case of severe delay, enable the supplementary lighting prediction algorithm based on historical data and prediction models to generate and issue supplementary lighting instructions in advance.

[0041] In S1, collect the real-time multi-dimensional data of the agricultural environment through the sensor network. The multi-dimensional data includes environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency. Specifically: The data source of environmental light intensity includes: light sensors (such as photodiodes, quantum sensors). Data content: Measure the light intensity (unit: lux or μmol / m2 / s) and spectral distribution (such as the ratio of red light to blue light). Application function: Evaluate in real time whether the lighting conditions meet the photosynthesis requirements of the crops; trigger the supplementary lighting system when insufficient light is monitored; optimize crop growth according to the spectral requirements (ratio of red light to blue light), such as blue light promoting leaf expansion and red light promoting fruit development.

[0042] The data source of temperature and humidity is: temperature and humidity sensors (such as DHT22, SHT31). Data content: Measure the air temperature (unit: °C) and humidity (unit: %RH). Application function: Provide the microclimate conditions suitable for crop growth, avoid stress on crops caused by high temperature or too low humidity; coordinate the linkage with the supplementary lighting system, such as reducing the light intensity or turning on the cooling device at high temperature; combine historical data analysis to predict the possible risk of pest and disease occurrence (such as high humidity environment is prone to mold).

[0043] The data source of carbon dioxide concentration is: carbon dioxide sensors (such as NDIR sensors). Data content: Measure the concentration of carbon dioxide in the air (unit: ppm). Application function: Carbon dioxide is a key raw material for photosynthesis. Ensure that it is within the appropriate range (generally between 300 - 1200 ppm) by monitoring the concentration; in a closed environment (such as a greenhouse), adjust the ventilation or carbon dioxide supplementation device in time to optimize the photosynthesis efficiency of the crops; the trend of carbon dioxide concentration change can be used as an indirect indicator of the environmental ventilation condition or the growth activity of the crops.

[0044] The data sources of crop photosynthesis efficiency are: chlorophyll fluorescence sensors, photosynthesis monitors (such as FMS or LI-COR). Data content: Relevant parameters for measuring crop photosynthesis: Photochemical efficiency (Fv / Fm): Reflects the potential efficiency of photosynthesis; Electron transport rate (ETR): Reflects the utilization efficiency of light energy during photosynthesis; Stomatal conductance and transpiration rate: Indirectly reflects the crop's response ability to environmental changes. Application: Dynamically monitor the physiological state of crops, evaluate whether environmental conditions meet crop requirements; Combine the photosynthesis efficiency with environmental parameters for analysis to provide real-time feedback for light supplementation adjustment; Detect whether crops are under environmental stress (such as excessive light, insufficient carbon dioxide), and take optimization measures in advance.

[0045] Sensors are evenly distributed in the planting area, and the layout density is adjusted according to the spatial layout of crop growth to ensure the representativeness of data. Sensors collect data in real time (the collection interval is usually 1–10 seconds) to ensure the timeliness of environmental parameters. Data transmission methods include: Short distance: Transmitted to the gateway device through wireless protocols (such as Zigbee, LoRa, Wi-Fi); Long distance: Uploaded to the cloud or local server for processing through cellular networks (such as 4G / 5G).

[0046] S2: Use edge computing devices to preprocess the collected data, and extract the irregular features of photosynthesis efficiency and the fluctuation features of the light intensity change rate from the preprocessed collected data.

[0047] The goal of data preprocessing is to improve data quality and extract core information, laying a foundation for subsequent feature analysis. Remove noise and outliers in the collected data to ensure the authenticity and accuracy of the data. Use a moving average filter to smooth the light intensity and photosynthesis efficiency data; Detect and remove outliers based on the box plot method; Use wavelet denoising method to remove high-frequency interference in the collected signal. The timestamps of data collected by different sensors may be asynchronous, and the time axis needs to be unified. Use linear interpolation method to align the multi-dimensional data at different times; For data with inconsistent sampling intervals, resample at a unified frequency (such as 1 second). Eliminate the influence of measurement units and magnitudes of different sensors for unified analysis. Normalize the data of each dimension (such as Min-Max normalization or Z-Score normalization).

[0048] Abnormal or irregular changes in photosynthesis efficiency reflect the response changes of crops to environmental conditions. Extracting these features helps monitor the health status of crops.

[0049] Extract the photosynthesis efficiency from the chlorophyll fluorescence parameters (such as Fv / Fm) or electron transport rate (ETR) data obtained from sensors. The change of photosynthesis efficiency may be affected by various factors such as light intensity, temperature and humidity, and correlation analysis needs to be combined with environmental parameters.

[0050] Fluctuation feature extraction: Calculate the short-term standard deviation of photosynthesis efficiency data to detect the degree of fluctuation; use discrete wavelet transform to decompose the data, extract the characteristic frequency components, and identify potential abnormal fluctuations. Abnormal pattern recognition: Construct a baseline model based on historical data to detect abnormal points beyond the baseline; use the autoregressive integrated moving average (ARIMA) model or long short-term memory network (LSTM) to predict photosynthesis efficiency, and use the deviation between the predicted value and the actual value as an abnormal indicator.

[0051] The fluctuation of the light intensity change rate reflects the dynamic characteristics of the ambient light and is an important reference for the adjustment of the supplementary lighting system. According to the real-time data of the light sensor, calculate the rate of change of light intensity over time (unit: lux / second or μmol / m2 / s 2 ). Sudden changes (such as cloud cover or artificial intervention) will cause abnormalities in the light intensity fluctuation characteristics. Calculate the mean and standard deviation of the light intensity rate change to evaluate the fluctuation amplitude; extract the change frequency characteristics through fast Fourier transform (FFT) to judge the periodicity of the light fluctuation. Mark the time periods with a rate fluctuation amplitude exceeding the set threshold as abnormal; use clustering algorithms (such as K-Means) to classify the light intensity rate change patterns into normal and abnormal categories.

[0052] After analyzing the extracted irregular characteristics of photosynthesis efficiency, generate an irregular index of photosynthesis efficiency. The method for obtaining the irregular index of photosynthesis efficiency is as follows:

[0053] Take the collected photosynthesis efficiency time series X(s) as the input signal. The time series is defined as: X = {x 1 , x 2 ,..., x N}; where N is the total number of data points, and x N is the photosynthesis efficiency value at the Nth moment. Select a wavelet basis function suitable for photosynthesis efficiency data analysis (such as Daubechies wavelet db4 or Symlet wavelet sym4). Perform hierarchical decomposition on the signal X(s) to extract the low-frequency part A j and the high-frequency part D i , and the formula is: Select the decomposition layer j according to the signal length and sampling frequency, and extract the energy of all high-frequency components D i to measure the irregularity of the signal. The high-frequency energy E HighFreq calculation formula is: where D i [k] is the coefficient of the high-frequency signal of the i-th layer at the k-th sampling point, and N i is the number of signal points of the i-th layer; at the same time, calculate the total energy E Total of the complete signal, and the expression is: Calculate the irregular index of photosynthesis efficiency, where PEII is the irregular index of photosynthesis efficiency.

[0054] When the irregular index of photosynthesis efficiency (PEII) is large, it indicates that the high-frequency fluctuation component occupies a large proportion in the photosynthesis efficiency data, and the signal shows violent and frequent changes. This situation usually reflects an obvious delay anomaly between real-time environmental data and the supplementary lighting adjustment decision. Specifically, when the external environment (such as light intensity, temperature and humidity, carbon dioxide concentration) changes rapidly, the supplementary lighting adjustment system fails to respond in time and optimize the lighting conditions, resulting in violent fluctuations in the photosynthesis efficiency of crops in a short period of time. This delay may be due to the lag of sensor data transmission, insufficient edge computing processing speed or slow execution speed of the supplementary lighting system. If the delay problem is not solved, it will have a negative impact on the physiological state of crops, which may lead to a decrease in energy utilization efficiency and abnormal growth.

[0055] When the irregular index of photosynthesis efficiency (PEII) is small, it indicates that the high-frequency fluctuation component is small, and the photosynthesis efficiency signal is relatively stable and has a consistent change trend. This indicates that the degree of delay anomaly between real-time environmental data and the supplementary lighting adjustment decision is low. The system can quickly sense environmental changes and adjust the supplementary lighting strategy within a short time to maintain the stability of the photosynthesis efficiency of crops. For example, when the light intensity weakens, the system quickly turns on the supplementary lights or optimizes the spectral distribution to provide continuous and stable lighting conditions for the crops. At this time, the low-delay environmental response ability reflects the high efficiency of the supplementary lighting adjustment system, which helps to ensure the growth quality and resource utilization rate of crops.

[0056] After analyzing the extracted fluctuation characteristics of the light intensity change rate, a light intensity change rate fluctuation index is generated. The acquisition method of the light intensity change rate fluctuation index is as follows:

[0057] The input is the light intensity time series L(t), which represents the change of the collected light intensity over time. The time series is defined as: L = {L 1 , L 2 ,..., L M}; where: L M is the light intensity value at the Mth moment, M is the total number of data points, and the first-order difference reflects the change rate of the light intensity between adjacent moments. The calculation formula is: △L t = L t - L t-1 , t = 2, 3,..., M; where, △L t is the light intensity change rate at the tth moment, L t and L t-1They are the light intensity values at the current moment and the previous moment respectively; the first-order difference sequence △L = {△L 2 , △L 3 , ..., △L N}; the second-order difference reflects the degree of fluctuation of the light intensity change rate (the change of the rate change), and the calculation formula is: △ 2 L t = △L t - △L t-1 , t = 3, 4, ..., M; where: △ 2 L t is the change amount of the first-order difference at the t-th moment, △L t and △L t-1 are the first-order difference values at the current moment and the previous moment respectively, and the second-order difference sequence is △ 2 L = {△ 2 L 3 , △ 2 L 4 , ..., △ 2 L N}.

[0058] Calculate the fluctuation index according to the statistical characteristics of the second-order difference. Usually, the variance or mean square value of the second-order difference is used to quantify the fluctuation, and the expression is: In the formula, is the mean of the second-order difference, and IRFI is the light intensity change rate fluctuation index.

[0059] When the light intensity change rate fluctuation index (IRFI) is large, it indicates that the change rate of the light intensity per unit time has significant fluctuations, manifested as drastic changes or frequent instabilities in the light conditions. This situation usually reflects a large delay anomaly between the real-time environmental data and the supplementary lighting adjustment decision: the supplementary lighting system fails to promptly perceive the rapid change of the light intensity and make appropriate adjustments. Due to the delay problem, after the environmental light suddenly changes, the switching or spectral adjustment of the supplementary lighting lamp lags behind, resulting in the light conditions not being able to quickly return to balance. A large fluctuation index indicates that the system needs to optimize the acquisition frequency of the sensor, the data processing speed, or the response ability of the supplementary lighting equipment to reduce the negative impact of the delay on crop photosynthesis.

[0060] When the fluctuation index of the light intensity change rate (IRFI) is small, it indicates that the change rate of the light intensity tends to be stable, and a good dynamic balance is maintained between the ambient light and the supplementary light adjustment. This shows that the acquisition, transmission of real-time environmental data, and the execution of the supplementary light adjustment decision can quickly respond to the external light change, thereby controlling the fluctuation within a lower range. A smaller fluctuation index reflects a lower degree of delay anomaly, and the supplementary light system operates efficiently, being able to quickly correct the change in ambient light (such as the weakening or strengthening of natural light intensity), ensuring that the crops receive continuous and stable light support, thereby optimizing the photosynthesis efficiency and growth conditions of the crops.

[0061] S3: According to the irregular characteristics of the photosynthesis efficiency and the fluctuation characteristics of the light intensity change rate extracted, evaluate the degree of delay anomaly between the real-time environmental data and the supplementary light adjustment decision;

[0062] Convert the irregular index of the photosynthesis efficiency and the fluctuation index of the light intensity change rate into a comprehensive feature vector. Use the comprehensive feature vector as the input of the machine learning model. The machine learning model takes predicting the label of the degree of delay anomaly value between the real-time environmental data and the supplementary light adjustment decision for each group of comprehensive feature vectors as the prediction target, and takes minimizing the sum of the prediction errors of the degree of delay anomaly values between all real-time environmental data and the supplementary light adjustment decision as the training target, and trains the machine learning model until the sum of the prediction errors reaches convergence and then stops the model training. Determine the degree of delay anomaly value between the real-time environmental data and the supplementary light adjustment decision according to the model output result. Among them, the machine learning model is a polynomial regression model.

[0063] The method for obtaining the degree of delay anomaly value between the real-time environmental data and the supplementary light adjustment decision is: obtain the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: LR = F(PEII, IRFI); where F is the output function of the model, PEII is the irregular index of the photosynthesis efficiency, IRFI is the fluctuation index of the light intensity change rate, and LR is the degree of delay anomaly value between the real-time environmental data and the supplementary light adjustment decision.

[0064] S4: Dynamically optimize the supplementary light adjustment strategy according to the degree of delay anomaly, including adjusting the execution time window of the supplementary light adjustment to ensure real-time performance in the case of mild delay; in the case of severe delay, enable the supplementary light prediction algorithm based on historical data and prediction model to generate and issue supplementary light instructions in advance.

[0065] Compare the delay anomaly degree value between the acquired real-time environmental data and the light supplement adjustment decision with the reference threshold of the delay anomaly degree value preset under normal conditions according to historical data. If the delay anomaly degree value between the real-time environmental data and the light supplement adjustment decision is greater than or equal to the reference threshold of the delay anomaly degree value, it indicates that the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision is high. At this time, generate a warning signal, and classify the delay situation between the real-time environmental data and the light supplement adjustment decision as severe delay, and immediately enable the light supplement prediction algorithm based on historical data and prediction models to generate and issue light supplement instructions in advance; if the delay anomaly degree value between the real-time environmental data and the light supplement adjustment decision is less than the reference threshold of the delay anomaly degree value, it indicates that the delay anomaly degree between the real-time environmental data and the light supplement adjustment decision is low. At this time, do not generate a warning signal, and classify the delay situation between the real-time environmental data and the light supplement adjustment decision as mild delay. At this time, it is necessary to adjust the execution time window of the light supplement adjustment to ensure real-time performance.

[0066] In the case of mild delay, there is a certain delay in the response between the real-time environmental data and the light supplement adjustment decision, but the impact is small, and the system still has a certain degree of real-time performance. By adjusting the execution time window of the light supplement adjustment, the response speed is further improved to ensure the dynamic balance of the lighting conditions. Adjust the execution time window: shorten the response delay of the supplementary light, that is, reduce the interval time from calculation to actual supplementary light execution; increase the sensor data acquisition frequency so that the system can capture the lighting change trend faster; use edge computing devices to perform real-time analysis on the data to avoid relying entirely on cloud processing for data. If the ambient light change rate is low (such as stable sunlight during the day), appropriately extend the execution window of the light supplement adjustment to avoid frequent switching; if the ambient light change rate is high (such as shadow occlusion), immediately trigger the light supplement instruction. Monitor the light supplement effect in real time, and verify the adjustment effect by collecting crop photosynthesis efficiency data (such as PEII); fine-tune the intensity or spectral distribution of the supplementary light according to the feedback signal after adjustment.

[0067] In the case of severe delay, the response time between the real-time environmental data and the light supplement adjustment decision exceeds the normal threshold, and the system is difficult to adapt to environmental changes in a timely manner, which may have a significant impact on crop photosynthesis. Through the light supplement prediction algorithm based on historical data and prediction models, predict the lighting change trend in advance, and generate and issue light supplement instructions.

[0068] Extract the rules from the historical data of ambient light and crop photosynthesis efficiency, and identify typical lighting change patterns (such as peak sunshine periods, cloud occlusion rules, etc.). Use time series prediction models (such as LSTM, ARIMA) to predict the change trend of light intensity in advance. Based on the time series prediction model of light intensity and PEII, calculate the change trend of future light intensity The change trend is expressed as: Where, Δt is the predicted time step, L(t) is the current light intensity, L(t - Δt) is the past light intensity, and PEII(t) is the irregular index of photosynthesis efficiency at time t. An early light supplement instruction is generated according to the prediction result, so that the supplementary light lamp can supplement light in time before the actual light intensity drops. f is the prediction model function. If it is predicted that the light intensity will drop significantly after Tp time, turn on the supplementary light lamp in advance by Ta (safety time window);

[0069] The intensity of the supplementary light lamp can be dynamically adjusted according to the predicted light change rate, and the expression is: I_target is the target light intensity. Combining the historical data of light supplementation and the fluctuation index of light intensity change rate (IRFI), the prediction model is further calibrated to improve the prediction accuracy. After the light supplementation is executed, the photosynthesis efficiency data (PEII) and light intensity data of the crops are monitored in real time to evaluate the effect of the prediction algorithm; if the deviation is large, the parameters of the prediction model are automatically adjusted to improve the accuracy.

[0070] In this embodiment, real-time multi-dimensional data of the agricultural environment are collected through a sensor network, including environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency. And the edge computing device is used to preprocess the data, and extract the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate. According to these characteristics, the delay anomaly degree between the real-time environmental data and the light supplementation adjustment decision is evaluated, and then the light supplementation adjustment strategy is dynamically optimized: in the case of mild delay, adjust the execution time window of the light supplementation adjustment to ensure real-time performance; in the case of severe delay, based on the light supplementation prediction algorithm of historical data and prediction model, generate and issue a light supplementation instruction in advance, so as to realize the precise control of the agricultural environment light and improve the crop growth efficiency.

[0071] Embodiment 2, please refer to Figure 2 As shown, an intelligent agricultural environment light supplementation adjustment system in this embodiment includes a sensor module, a data processing module, a delay degree evaluation module, an optimization module, and a light supplementation control module;

[0072] Sensor module: Collect real-time multi-dimensional data of the agricultural environment through a sensor network, and the multi-dimensional data includes environmental light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency;

[0073] Data processing module: Use the edge computing device to preprocess the collected data, and extract the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate in the preprocessed collected data;

[0074] Delay degree evaluation module: Evaluate the delay anomaly degree between the real-time environmental data and the light supplementation adjustment decision according to the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate;

[0075] Optimization module: Dynamically optimize the supplementary lighting adjustment strategy according to the degree of latency anomaly, including adjusting the execution time window of the supplementary lighting adjustment to ensure real-time performance in the case of mild latency; enabling a supplementary lighting prediction algorithm based on historical data and a prediction model to generate and issue supplementary lighting instructions in advance in the case of severe latency.

[0076] Supplementary lighting control module: Used to deeply analyze the degree of latency anomaly between real-time environmental data and supplementary lighting adjustment decisions within a fixed time period according to the optimized adjustment strategy, and dynamically adjust the light intensity and spectral distribution of the supplementary light according to the analysis results.

[0077] In the supplementary lighting control module, after dynamically optimizing the supplementary lighting adjustment strategy, collect the degree of latency anomaly values between the real-time environmental data generated within subsequent fixed time periods and the supplementary lighting adjustment decisions, establish a corresponding data set, calculate the mean and standard deviation of the data set, and after analyzing it, dynamically adjust the light intensity and spectral distribution of the supplementary light according to the analysis results.

[0078] If the mean value of the degree of latency anomaly in the data set is greater than or equal to the reference threshold of the mean value of the degree of latency anomaly, and the standard deviation of the degree of latency anomaly is less than the reference threshold of the standard deviation of the degree of latency anomaly, it indicates that the degree of latency anomaly is high and the fluctuation is small, indicating that the supplementary lighting adjustment strategy has a systematic lag. At this time, it is necessary to increase the light intensity of the supplementary light or adjust the spectral distribution to improve the photosynthesis efficiency of the crops and optimize the system response speed.

[0079] If the mean value of the degree of latency anomaly is greater than or equal to the reference threshold of the mean value of the degree of latency anomaly, and the standard deviation of the degree of latency anomaly is greater than or equal to the reference threshold of the standard deviation of the degree of latency anomaly, it indicates that the degree of latency anomaly is high and the fluctuation is significant, indicating that the supplementary lighting adjustment strategy cannot adapt to the rapid change of environmental light. At this time, it is necessary to enable a more refined supplementary lighting prediction algorithm, strengthen the real-time monitoring and rapid response capabilities, and reduce latency and fluctuations.

[0080] If the mean value of the degree of latency anomaly is less than the reference threshold of the mean value of the degree of latency anomaly, and the standard deviation of the degree of latency anomaly is greater than or equal to the reference threshold of the standard deviation of the degree of latency anomaly, it indicates that the degree of latency anomaly is low but the fluctuation is significant, indicating that the overall adjustment is relatively timely, but the adaptability to environmental changes is insufficient. At this time, it is necessary to dynamically optimize the spectral distribution, improve the lighting uniformity and the stability of crop growth, and reduce the system adjustment frequency.

[0081] If the mean value of the delay anomaly degree is less than the reference threshold of the mean value of the delay anomaly degree, and the standard deviation of the delay anomaly degree is less than the reference threshold of the standard deviation of the delay anomaly degree, it indicates that the delay anomaly degree is low and the fluctuation is small, suggesting that the supplementary lighting adjustment strategy is efficient and stable. At this time, it is necessary to maintain the current supplementary lighting intensity and spectral distribution, and regularly monitor to ensure the continuous efficient operation of the system.

[0082] In this embodiment, the supplementary lighting control module deeply analyzes the degree of delay anomaly between the real-time environmental data and the supplementary lighting adjustment decision within a fixed time period, and dynamically adjusts the lighting intensity and spectral distribution of the supplementary light according to the mean value and the standard deviation. If the mean value of the delay anomaly degree is high and the fluctuation is small, it indicates that the supplementary lighting strategy lags systematically, and it is necessary to increase the lighting intensity or adjust the spectral distribution to improve the response speed; if the mean value is high and the fluctuation is significant, it means that the supplementary lighting strategy is difficult to adapt to rapid changes, and a more refined prediction algorithm needs to be enabled to enhance the real-time monitoring ability; if the mean value is low but the fluctuation is significant, it indicates that the adjustment is relatively timely but the adaptability to environmental changes is insufficient, and the spectral distribution needs to be optimized to improve the uniformity and stability; if the mean value is low and the fluctuation is small, it indicates that the supplementary lighting adjustment strategy is efficient and stable. At this time, the current strategy should be maintained and regularly monitored to ensure the continuous efficient operation of the system.

[0083] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula that is closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0084] It should be understood that the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship. Specifically, it can be understood by referring to the context before and after.

[0085] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0086] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application.

Claims

1. A smart agricultural environment monitoring method, characterized in that: The following steps are involved: S1: Collecting real-time multi-dimensional data of the agricultural environment through a sensor network, the multi-dimensional data including ambient light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency; S2: Use edge computing devices to preprocess the collected data and extract the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate from the preprocessed collected data; S3: Based on the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate, the abnormal degree of delay between real-time environmental data and supplementary light adjustment decision is evaluated; S4: Dynamically optimize the fill light adjustment strategy according to the degree of delay abnormality, including adjusting the execution time window of fill light adjustment to ensure real-time performance in the case of mild delay; in the case of severe delay, enable the fill light prediction algorithm based on historical data and prediction model to generate and issue fill light instructions in advance.

2. The intelligent agricultural environment monitoring method according to claim 1, characterized in that: In S2, the extracted irregular features of photosynthesis efficiency are analyzed to generate the irregular index of photosynthesis efficiency. The method for obtaining the irregular index of photosynthesis efficiency is as follows: The collected photosynthesis efficiency time series X(s) is used as the input signal, and the time series is defined as: X = {x1, x2, …, x N }; where N is the total number of data points, x N is the photosynthesis efficiency value at the Nth moment. The wavelet basis function for photosynthesis efficiency data analysis is selected to perform hierarchical decomposition on the signal X(s) and extract the low-frequency part A j and high frequency part D i , the formula is: Select the decomposition layer j according to the signal length and sampling frequency to extract all high-frequency components D i Energy, used to measure the irregularity of the signal, high-frequency energy E HighFreq The calculation formula is: Among them, D i [k] is the coefficient of the high-frequency signal of the i-th layer at the k-th sampling point, N i is the number of signal points in the i-th layer; at the same time, the total energy E of the complete signal is calculated Total , the expression is: Calculate the irregularity index of photosynthesis efficiency, Where PEII is the photosynthesis efficiency irregularity index.

3. The intelligent agricultural environment monitoring method according to claim 2, characterized in that: In S2, the extracted light intensity change rate fluctuation characteristics are analyzed to generate a light intensity change rate fluctuation index. The light intensity change rate fluctuation index is obtained as follows: The input is the light intensity time series L(t), which represents the change of the collected light intensity over time. The time series definition is: L = {L1, L2, ..., L M }; where: L M is the light intensity value at the Mth moment, M is the total number of data points, and the first-order difference reflects the rate of change of light intensity between adjacent moments. The calculation formula is: ΔL t =L t -L t-1 ,t=2,3,…,M; where ΔL t is the rate of change of light intensity at time t, L t and L t-1 are the light intensity values ​​at the current moment and the previous moment respectively; the first-order difference sequence ΔL={ΔL2,ΔL3,…,ΔL N }; The second-order difference reflects the fluctuation degree of the rate of change of light intensity, and the calculation formula is: Δ 2 L t =ΔL t -ΔL t-1 ,t=3,4,…,M;where: Δ 2 L t is the change in the first-order difference at time t, ΔL t and ΔL t-1 are the first-order difference values ​​at the current moment and the previous moment respectively, and the second-order difference sequence is Δ 2 L = {Δ 2 L3,Δ 2 L4,…,Δ 2 L N }; The volatility index is calculated based on the statistical characteristics of the second-order difference, and the expression is: In the formula, is the mean of the second-order difference, and IRFI is the light intensity change rate fluctuation index.

4. The intelligent agricultural environment monitoring method according to claim 3 is characterized in that: In S3, the abnormal degree of delay between the real-time environmental data and the supplementary light adjustment decision is evaluated based on the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of the light intensity change rate; The irregular index of photosynthesis efficiency and the fluctuation index of light intensity change rate are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the delay anomaly value label between the real-time environmental data and the fill light adjustment decision as the prediction target, and takes minimizing the sum of prediction errors of the delay anomaly value labels between all real-time environmental data and the fill light adjustment decision as the training target. The machine learning model is trained until the sum of prediction errors converges, and the model training is stopped. The delay anomaly value between the real-time environmental data and the fill light adjustment decision is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. The intelligent agricultural environment monitoring method according to claim 4 is characterized in that: In S4, the fill light adjustment strategy is dynamically optimized according to the degree of delay anomaly, specifically: The acquired delay abnormality value between the real-time environmental data and the fill light adjustment decision is compared with the delay abnormality value reference threshold preset under normal circumstances according to historical data. If the delay abnormality value between the real-time environmental data and the fill light adjustment decision is greater than or equal to the delay abnormality value reference threshold, it means that the delay abnormality between the real-time environmental data and the fill light adjustment decision is high. At this time, an early warning signal is generated, and the delay between the real-time environmental data and the fill light adjustment decision is classified as severe delay. The fill light prediction algorithm based on historical data and prediction model is immediately enabled, and the fill light instruction is generated and issued in advance. If the delay abnormality value between the real-time environmental data and the fill light adjustment decision is less than the delay abnormality value reference threshold, it means that the delay abnormality between the real-time environmental data and the fill light adjustment decision is low. At this time, no early warning signal is generated, and the delay between the real-time environmental data and the fill light adjustment decision is classified as mild delay. At this time, the execution time window of the fill light adjustment needs to be adjusted to ensure real-time performance.

6. An intelligent agricultural environment supplementary light adjustment system, used to implement an intelligent agricultural environment monitoring method according to any one of claims 1 to 5, characterized in that: It includes a sensor module, a data processing module, a delay degree evaluation module, an optimization module and a fill light control module; Sensor module: collects real-time multi-dimensional data of the agricultural environment through a sensor network, including ambient light intensity, temperature and humidity, carbon dioxide concentration, and crop photosynthesis efficiency; Data processing module: Use edge computing devices to preprocess the collected data and extract the irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate from the preprocessed collected data; Delay evaluation module: Based on the extracted irregular characteristics of photosynthesis efficiency and the fluctuation characteristics of light intensity change rate, the abnormal delay between real-time environmental data and supplementary light adjustment decision is evaluated; Optimization module: Dynamically optimizes the fill light adjustment strategy according to the degree of delay abnormality, including adjusting the execution time window of fill light adjustment to ensure real-time performance in the case of mild delay; in the case of severe delay, enables the fill light prediction algorithm based on historical data and prediction models to generate and issue fill light instructions in advance; Fill light control module: It is used to conduct in-depth analysis of the abnormal degree of delay between real-time environmental data and fill light adjustment decisions within a fixed time period according to the optimized adjustment strategy, and dynamically adjust the light intensity and spectral distribution of the fill light according to the analysis results.

7. The intelligent agricultural environment supplementary light adjustment system according to claim 6, characterized in that: In the fill light control module, the delay anomaly between the real-time environmental data and the fill light adjustment decision within a fixed time period is deeply analyzed, and the light intensity and spectral distribution of the fill light are dynamically adjusted according to the analysis results. Specifically: After dynamically optimizing the fill light adjustment strategy, the delay anomaly values ​​between the real-time environmental data generated in the subsequent fixed time period and the fill light adjustment decision are collected, and a corresponding data set is established. The mean and standard deviation of the data set are calculated. After analyzing it, the light intensity and spectral distribution of the fill light are dynamically adjusted according to the analysis results.

8. The intelligent agricultural environment supplementary light adjustment system according to claim 7, characterized in that: If the mean value of the delay anomaly degree in the data set is greater than or equal to the reference threshold value of the delay anomaly degree, and the standard deviation of the delay anomaly degree is less than the reference threshold value of the delay anomaly degree, it means that the delay anomaly degree is high and the fluctuation is small, indicating that the fill light adjustment strategy is systematically lagging. At this time, it is necessary to increase the fill light intensity or adjust the spectral distribution to improve the photosynthesis efficiency of the crop and optimize the system response speed; If the mean value of the delay abnormality is greater than or equal to the reference threshold of the mean value of the delay abnormality, and the standard deviation of the delay abnormality is greater than or equal to the reference threshold of the standard deviation of the delay abnormality, it means that the delay abnormality is high and fluctuates significantly, indicating that the fill light adjustment strategy cannot adapt to the rapid changes in ambient light. At this time, it is necessary to enable the fill light prediction algorithm to strengthen real-time monitoring and rapid response capabilities to reduce delays and fluctuations; If the mean value of the delay anomaly degree is less than the reference threshold value of the delay anomaly degree, and the standard deviation of the delay anomaly degree is greater than or equal to the reference threshold value of the delay anomaly degree, it means that the delay anomaly degree is low but fluctuates significantly, and the adaptability to environmental changes is insufficient. At this time, it is necessary to dynamically optimize the spectral distribution, improve the uniformity of light and the stability of crop growth, and reduce the frequency of system adjustment; If the mean value of the delay abnormality degree is less than the reference threshold of the mean value of the delay abnormality degree, and the standard deviation of the delay abnormality degree is less than the reference threshold of the standard deviation of the delay abnormality degree, it means that the delay abnormality degree is low and the fluctuation is small, indicating that the fill light adjustment strategy is efficient and stable. At this time, it is necessary to maintain the current fill light intensity and spectral distribution, and monitor regularly to ensure that the system continues to operate efficiently.

Images