Selenium sand melon growth monitoring management method and system based on Internet of Things

Abstract

The invention discloses a selenium sand melon growth monitoring management method and system based on Internet of Things, and particularly relates to the technical field of data management. A planting area is divided into a plurality of monitoring areas, sensor data are collected in real time, the matching condition of the data and a historical range and the overlapping degree of data of multiple land parcels are analyzed, and the possible problem of equipment binding errors or serial number confusion is recognized; when the data overlapping degree is high, an operation instruction sent by the management system and an equipment feedback signal are recorded, and the fluctuation condition of the equipment response success rate is analyzed; according to the method, data overlapping and response success rate fluctuation are integrated, the accuracy of system operation execution is evaluated, when the accuracy is low, control logic and decision threshold values are dynamically adjusted by predicting an accuracy coefficient in a fixed time period, so that large-area crop loss caused by decision errors is effectively avoided, the accuracy and reliability of system operation are improved, and the system performance is improved. And high efficiency and stability of agricultural production are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for monitoring and managing the growth of selenium sand melon based on the Internet of Things. Background Art

[0002] The IoT-based monitoring and management of selenium sand melon growth refers to the use of IoT technology to conduct real-time monitoring and intelligent management of key factors such as the environment, soil, and meteorology during the growth of selenium sand melon. Through the sensor network, the monitoring equipment can collect data including soil moisture, temperature, light, air humidity, and selenium content, which are uploaded to the cloud platform via a wireless network. Growers can view this data in real time through their mobile phones or computers to understand the growth status and required environmental conditions of the selenium sand melon. In addition, the management system can also provide planting recommendations through big data analysis and artificial intelligence technology, combined with historical data and expert planting models. For example, the system will automatically adjust the irrigation or fertilization strategy based on the current data to ensure the best environment for the growth of selenium sand melon. This technology not only improves planting efficiency and yield, but also reduces resource waste, helping growers manage crops in a scientific way and achieve high-quality selenium sand melon production.

[0003] The prior art has the following deficiencies:

[0004] When a management system manages multiple plots, if there are errors in plot numbering or equipment binding, it may lead to confusion in decision-making, such as applying the irrigation demand of plot X to the fertilization operation of plot Y, resulting in over-irrigation or over-fertilization of a certain plot. Such errors may not only cause waterlogging and fertilizer damage in a single plot, but in serious cases, they may spread through the regional management chain, affecting multiple plots, and ultimately leading to large-scale crop yield reduction or even crop failure, with disastrous consequences for agricultural production. Summary of the invention

[0005] The purpose of the present invention is to provide a selenium sand melon growth monitoring and management method and system based on the Internet of Things to solve the shortcomings of the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a method for monitoring and managing the growth of selenium sand melon based on the Internet of Things, comprising the following steps:

[0007] S1: Divide the selenium sand melon growing area into several monitoring areas, and collect real-time data streams from plot sensors in each monitoring area, including selenium sand melon growth environment data and fertilization operation data;

[0008] S2: Check the matching of the selenium sand melon growth environment data collected by the plot sensors with the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time;

[0009] S3: If the overlap is high, record each operation command sent from the management system to the device and the feedback signal after the monitoring device executes it, and analyze the fluctuation of the response success rate of the device correctly executing the command and returning the confirmation signal within the specified time;

[0010] S4: Evaluate the accuracy of management system operation execution based on the overlap of selenium sand melon growth environment data collected by multiple plot sensors at the same time and the fluctuation of the response success rate of the equipment correctly executing instructions and returning confirmation signals within the specified time;

[0011] S5: When the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

[0012] Preferably, in S2, the growth environment data of selenium sand melon collected by the plot sensor and the matching situation of the historical data range of the plot are analyzed to generate a growth environment data matching anomaly index, and the method for obtaining the growth environment data matching anomaly index is:

[0013] The historical growth environment data obtained from the plot sensors, the historical data set is denoted as D = {x1, x2, ..., x n}; where x n Represents the nth data point. Use the kernel density estimation method to estimate the probability density function P(x) of historical data. The kernel density estimation formula is: n is the number of historical data points, h is the bandwidth of the kernel function, which controls the degree of smoothness, and K is the kernel function; for real-time data points x new Input the model and calculate its probability density value P(x new ), the formula is: The growth environment data matching anomaly index is calculated based on the probability value. The calculation expression is: AI = -log(P(x new )); where AI is the growth environment data matching anomaly index.

[0014] Preferably, in S3, after analyzing the fluctuation of the response success rate of the device correctly executing the instruction and returning the confirmation signal within the specified time, a response success rate fluctuation value is generated, and the method for obtaining the response success rate fluctuation value is:

[0015] Collect the sequence data of the device's response success rate over time and record the success rate R in each time period t = {R1, R2, …, R n}, where R t is the response success rate in the tth time period; select the time window length n as the basic time period for calculating the fluctuation value, and calculate the average value μ of the response success rate in the time window R , the expression is: Calculate the standard deviation of the success rate σ R Indicates the fluctuation range: Calculate the response success rate fluctuation value, which is the ratio of the standard deviation to the mean, to measure the intensity of the fluctuation: Where AS is the response success rate fluctuation value.

[0016] Preferably, in S4, the accuracy of the management system operation execution is evaluated according to the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time and the fluctuation of the response success rate of the device correctly executing the instruction and returning the confirmation signal within the specified time, specifically:

[0017] The growth environment data matching anomaly index and the response success rate fluctuation value are normalized, and the accuracy coefficient of the management system operation execution is calculated based on the normalized growth environment data matching anomaly index and the response success rate fluctuation value.

[0018] Preferably, the acquired accuracy coefficient of the management system operation execution is compared with the accuracy coefficient reference threshold preset under normal conditions in historical data. If the accuracy coefficient of the management system operation execution is greater than or equal to the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is high, and no warning signal is generated at this time; if the accuracy coefficient of the management system operation execution is less than the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is low, and a warning signal is generated at this time.

[0019] Preferably, in S5, when the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction result, specifically:

[0020] Collect the accuracy coefficient A of the management system operation execution within a fixed time period n , the time series is expressed as: A={A1,A2,…,A n}; Among them, A n is the accuracy coefficient at the nth time point; the prediction formula based on ARIMA is: is the accuracy coefficient of the next time point predicted, φ i is the autoregression coefficient, which indicates the impact of past data on current predictions, θ j is the moving average coefficient, which indicates the influence of residual on prediction, ∈ t is the prediction error, p,q is the model order; based on the prediction results Dynamically adjust the control logic and decision threshold of the management system. The decision threshold adjustment formula is: Tnew Dynamically adjusted decision threshold, T base is the basic decision threshold, which indicates the standard threshold of the system under normal conditions. ref is the accuracy reference value, which indicates the accuracy coefficient of the system under normal conditions, and δ is the adjustment factor, which controls the adjustment range.

[0021] Preferably, according to the accuracy coefficient of the prediction Adjustments include: When the accuracy coefficient of the forecast If the error is less than the reference threshold of the predicted accuracy coefficient, an abnormal signal is triggered to prompt the operator to check the system operation status and adjust the system according to the predicted error. Dynamically optimize the prediction model and gradually improve the prediction accuracy.

[0022] The present invention also provides a selenium sand melon growth monitoring and management system based on the Internet of Things, including a real-time data acquisition module, an overlap analysis module, an equipment response analysis module, an accuracy evaluation module and a dynamic adjustment module:

[0023] The selenium sand melon growing area is divided into several monitoring areas, and the real-time data stream of the plot sensors in each monitoring area is collected in real time, and the real-time data stream includes the selenium sand melon growth environment data and fertilization operation data;

[0024] Overlap analysis module: Check the matching between the selenium sand melon growth environment data collected by the plot sensors and the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time;

[0025] Equipment response analysis module: If the overlap is high, record each operation instruction sent from the management system to the equipment and the feedback signal after the monitoring equipment executes it, and analyze the fluctuation of the response success rate of the equipment correctly executing the instruction and returning the confirmation signal within the specified time;

[0026] Accuracy evaluation module: evaluates the accuracy of management system operation execution based on the degree of overlap of selenium sand melon growth environment data collected by sensors in multiple plots at the same time and the fluctuation of the response success rate of the equipment to correctly execute instructions and return confirmation signals within the specified time;

[0027] Dynamic adjustment module: When the accuracy of management system operation execution is low, the accuracy of management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

[0028] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0029] 1. The present invention provides an efficient and intelligent solution for the growth management of selenium sand melon by introducing Internet of Things technology, combining real-time monitoring, anomaly detection and dynamic optimization. Through regional division and sensor data collection, the system monitors the plot environment and fertilization operation data in real time, and generates a growth environment data matching anomaly index based on kernel density estimation to quantify data deviation. At the same time, the system comprehensively evaluates the accuracy of operation execution by analyzing the fluctuation value of the response success rate of the device execution instruction. When the accuracy is lower than the reference threshold, the system improves the accuracy of the operation through time series prediction and dynamic adjustment of the decision threshold, avoids decision confusion caused by data anomalies or device binding errors, and ensures the reliability and stability of agricultural production.

[0030] 2. The present invention has significant advantages in improving the accuracy of system operation, reducing risks and optimizing resource allocation. Its core technology can not only quickly discover and solve the problems of device binding errors and data overlap in multi-plot management, but also continuously optimize system performance by dynamically adjusting the control logic and prediction model. Ultimately, this method significantly reduces the risk of crop yield reduction, improves the scientificity and efficiency of agricultural management, and provides important technical support for selenium sand melon cultivation. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0032] Figure 1 The figure is a flow chart of the method of the present invention.

[0033] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Example 1, please refer to Figure 1 and Figure 2 As shown, the selenium sand melon growth monitoring and management method based on the Internet of Things described in this embodiment includes the following steps:

[0036] S1: Divide the selenium sand melon growing area into several monitoring areas, and collect real-time data streams from plot sensors in each monitoring area, including selenium sand melon growth environment data and fertilization operation data;

[0037] S2: Check the matching of the selenium sand melon growth environment data collected by the plot sensors with the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time;

[0038] S3: If the overlap is high, record each operation command sent from the management system to the device and the feedback signal after the monitoring device executes it, and analyze the fluctuation of the response success rate of the device correctly executing the command and returning the confirmation signal within the specified time;

[0039] S4: Evaluate the accuracy of management system operation execution based on the overlap of selenium sand melon growth environment data collected by multiple plot sensors at the same time and the fluctuation of the response success rate of the equipment correctly executing instructions and returning confirmation signals within the specified time;

[0040] S5: When the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

[0041] In S1, the selenium sand melon growing area is divided into several monitoring areas, and the real-time data stream of the plot sensors in each monitoring area is collected in real time. The real-time data stream includes the selenium sand melon growth environment data and fertilization operation data, specifically:

[0042] The growing area of ​​selenium sand melon is divided into several monitoring areas, including:

[0043] Based on topography: The selenium sand melon planting area is divided into several monitoring areas according to natural features such as topography, soil type, drainage conditions, etc. For example, highlands and flat lands may require different management strategies.

[0044] Crop planting density: Determine the coverage area of ​​each monitoring area based on the planting density and growth scale of selenium sand melon to ensure uniform sensor monitoring coverage.

[0045] Land management requirements: Consider factors such as water source distribution and irrigation pipeline layout in the area to facilitate subsequent installation and maintenance of management equipment.

[0046] Each monitoring area is assigned a unique number to facilitate subsequent data management and instruction distribution. Each number is bound to a sensor device to form a plot-device mapping relationship.

[0047] Install the sensor network: Soil sensor: used to collect key data such as soil moisture, temperature, selenium content, nutrient concentration (such as nitrogen, phosphorus, potassium), etc. Meteorological sensor: monitors environmental conditions such as air temperature and humidity, light intensity, wind speed, rainfall, etc. Fertilization operation monitoring sensor: records the amount of fertilizer, frequency of fertilizer application and type of fertilizer, and detects the execution of fertilization in real time.

[0048] A sufficient number of sensors are deployed in each monitoring area to ensure the comprehensiveness and accuracy of data collection. The sensors are connected to the regional management system through wireless communication (such as LoRa, NB-IoT) to form an IoT data transmission network.

[0049] Real-time data collection: Selenium sand melon growth environment data: Soil moisture and temperature: used to determine irrigation needs and root health. Selenium content: ensure precise regulation of crop selenium absorption. Meteorological parameters: such as air humidity and light intensity, to evaluate whether the growth environment has reached the optimal conditions. Fertilization operation data: Actual fertilizer amount: verify whether it is consistent with the planned fertilizer amount. Fertilization completion feedback signal: confirm whether the equipment has completed the operation. Data collection frequency: Real-time collection of environmental data, it is recommended to update at a frequency of minutes or hours. For fertilization operation data, it is collected according to the time period of each operation process, and is generally uploaded immediately after the operation is completed. Data transmission and storage: Data is uploaded to the cloud or local server through wireless communication protocol and stored in the database for subsequent analysis. Real-time data streams are marked with timestamps and area numbers to ensure accurate positioning of data.

[0050] The user interface of the management system is constructed to display the real-time data of each monitoring area, such as soil moisture curve, selenium content changes, fertilization operation records, etc. If the environmental parameters of a certain area exceed the preset range (such as humidity is too low or selenium concentration is too high), the system sends a real-time alarm through the monitoring platform to prompt manual intervention.

[0051] S2: Check whether the selenium sand melon growth environment data collected by the plot sensors match the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time.

[0052] The historical environmental parameter ranges of each plot are extracted from the database, including: Soil moisture: the historical minimum and maximum humidity value range. Soil temperature: the temperature range for normal crop growth (such as 15℃-30℃). Selenium content: the average value and upper and lower floating range during the planting period.

[0053] Monitor whether the current real-time data falls within the historical range. For example, if the humidity data of a plot of land suddenly falls below the historical minimum or rises above the maximum, there may be data anomalies or equipment problems. If the selenium content exceeds the set range, fertilization operations may need to be stopped immediately. Mark the real-time data that fails to match and generate an alarm notification to prompt managers to further check the operating status of the sensor or equipment.

[0054] After analyzing the matching of the growth environment data of selenium sand melon collected by the plot sensor with the historical data range of the plot, the growth environment data matching anomaly index is generated. The method for obtaining the growth environment data matching anomaly index is as follows:

[0055] The historical growth environment data obtained from the plot sensors, such as humidity, temperature, selenium content, etc. The historical data set is denoted as D = {x1, x2, …, x n}; where x n Represents the nth data point. Use these data to build a data distribution model as a benchmark for anomaly detection.

[0056] Use the kernel density estimation method to estimate the probability density function P(x) of historical data. The kernel density estimation formula is: n is the number of historical data points, h is the bandwidth of the kernel function, which controls the degree of smoothness. The best value is usually selected through cross-validation. K is the kernel function. new Input the model and calculate its probability density value P(x new ), the formula is: If P(x new ) value is high, then x new is a common value; for example, P(x new ) value is low, then x new is an abnormal value. The growth environment data matching abnormal index is calculated based on the probability value. The calculation expression is: AI = -log(P(x new )); where AI is the growth environment data matching anomaly index.

[0057] The larger the growth environment data matching anomaly index is, the lower the probability density of real-time data is, that is, the higher the deviation of the current data point from the historical distribution is. If sensors in multiple plots collect data with large anomaly indexes at the same time, it may mean that the environmental conditions of these plots are abnormally similar (for example, humidity and temperature are highly consistent at the same time), but they do not match their respective historical data distributions. This situation usually reflects an abnormally high degree of data overlap, which may be caused by confusion in device binding, sensor failure, or cross-numbering.

[0058] On the contrary, the smaller the growth environment data matching anomaly index is, the higher the probability density of the current data point is, which is consistent with the historical distribution and is a typical or common situation. If the data anomaly indexes of multiple plots are very small at the same time, it means that the environmental conditions of these plots match their respective historical conditions, and the degree of data overlap between them is low, reflecting that the differences between the plots are normal and the system is in good operation. In this case, it means that the plot data collected by the sensor is highly independent and the decision credibility is high.

[0059] S3: If the degree of overlap is high, record each operation instruction sent from the management system to the device and the feedback signal after the monitoring device executes it, and analyze the fluctuation of the response success rate of the device correctly executing the instruction and returning the confirmation signal within the specified time.

[0060] Record each operation instruction sent from the management system, including the following information: Target plot number: Identifies the plot to which the instruction is directed. Specific equipment bound (such as irrigation equipment or fertilization equipment). Instruction type: irrigation start, fertilization ratio adjustment, etc. Instruction timestamp: Records the specific time when the instruction is sent. Ensure that the plot and equipment bound are correct when the instruction is issued.

[0061] Monitor the feedback signal after the device executes the command, and record: Device feedback status: such as completion confirmation, error code, etc. Actual execution time: The timestamp when the device completes the operation, used to calculate the response delay. Check whether the device executes the command correctly within the specified time.

[0062] After analyzing the fluctuation of the response success rate of the device correctly executing the command and returning the confirmation signal within the specified time, a response success rate fluctuation value is generated. The method for obtaining the response success rate fluctuation value is as follows:

[0063] Collect the time series data of the device's response success rate, and record the success rate R in each time period (such as every minute or every hour). t = {R1, R2, …, R n}, where R t is the response success rate in the tth time period. Select the time window length n (such as 1 hour, 1 day) as the basic time period for calculating the fluctuation value. Calculate the average value of the response success rate in the time window μ R , the expression is: Calculate the standard deviation of the success rate σ R Indicates the fluctuation range: Calculate the response success rate fluctuation value, which is the ratio of the standard deviation to the mean, to measure the relative strength of the fluctuation: Where AS is the response success rate fluctuation value.

[0064] The larger the fluctuation value of the response success rate, the more unstable and fluctuating the success rate of the device in executing the operation command is. This may mean that the management system has problems such as device binding errors, communication delays, and uneven command distribution during the operation execution, resulting in reduced operation reliability. Therefore, a larger fluctuation value reflects that the accuracy of the management system's operation execution is low, and the system's operating mechanism and exception handling capabilities need to be optimized.

[0065] On the contrary, the smaller the fluctuation value of the response success rate, the more stable the success rate of the equipment in executing the operation instruction is, and the fluctuation range is small. This shows that the management system manages the equipment and operation more accurately, the communication link is unobstructed, and the instruction distribution is reasonable, thus ensuring the high accuracy of the operation execution. The smaller fluctuation value reflects that the system is in good operation and can reliably complete the operation tasks of multiple plots.

[0066] S4: Evaluate the accuracy of management system operation execution based on the degree of overlap of selenium sand melon growth environment data collected by sensors in multiple plots at the same time and the fluctuation of the response success rate of the equipment to correctly execute instructions and return confirmation signals within the specified time.

[0067] The growth environment data matching anomaly index and the response success rate fluctuation value are normalized, and the accuracy coefficient of the management system operation execution is calculated based on the normalized growth environment data matching anomaly index and the response success rate fluctuation value.

[0068] For example, the present invention may use the following formula to calculate the accuracy coefficient of the management system operation execution, and the calculation expression is: Where A is the accuracy coefficient of management system operation execution, AI is the growth environment data matching anomaly index, AS is the response success rate fluctuation value, a1 and a2 are the proportional coefficients of the growth environment data matching anomaly index and the response success rate fluctuation value, and a2>a1>0.

[0069] The acquired accuracy coefficient of the management system operation execution is compared with the accuracy coefficient reference threshold preset under normal conditions in historical data. If the accuracy coefficient of the management system operation execution is greater than or equal to the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is high and the system operates normally. At this time, no warning signal is generated; if the accuracy coefficient of the management system operation execution is less than the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is low and there may be problems. At this time, a warning signal is generated.

[0070] S5: When the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

[0071] Collect the accuracy coefficient A of the management system operation execution within a fixed time period n , the time series is expressed as: A={A1,A2,…,A n}; Among them, A n is the accuracy coefficient at the nth time point.

[0072] Select the forecast model. You can choose the following models to forecast the accuracy coefficient for time series:

[0073] Simple Moving Average (SMA): used to smooth fluctuations, suitable for short-term forecasting. Autoregressive Moving Average (ARIMA): captures trends and seasonal changes in time series, suitable for medium- and long-term forecasting. Long Short-Term Memory (LSTM): a nonlinear forecasting model based on deep learning, suitable for complex dynamic data.

[0074] ARIMA-based forecasting formula: is the accuracy coefficient of the next time point predicted, φ i is the autoregression coefficient, which indicates the impact of past data on current predictions, θ j is the moving average coefficient, which indicates the influence of residual on prediction, ∈ t is the prediction error, p and q are the model orders, which need to be determined through data fitting.

[0075] Based on the prediction results Dynamically adjust the control logic and decision threshold of the management system. The decision threshold adjustment formula is: T new Dynamically adjusted decision threshold, T base is the basic decision threshold, which indicates the standard threshold of the system under normal conditions. ref is the accuracy reference value, which represents the accuracy coefficient of the system under normal conditions (such as 0.85), and δ is the adjustment factor, which controls the adjustment range.

[0076] According to the accuracy coefficient of the prediction Adjust the following logic:

[0077] Increase the frequency of data verification: If Lower, increase the frequency of real-time verification of plot numbers and equipment binding to reduce errors. Increase the weight of manual intervention, reduce the scope of automatic decision-making, and introduce manual review of key instructions.

[0078] When the prediction accuracy coefficient If the error is less than the reference threshold of the predicted accuracy coefficient, an abnormal signal is triggered to prompt the operator to check the system operation status. Dynamically optimize the prediction model and gradually improve the prediction accuracy.

[0079] In this embodiment, the selenium sand melon growth area is divided into multiple monitoring areas, the growth environment data and fertilization operation data of the sensors in each area are collected in real time, and the matching of these data with the historical data range of the plot is checked, and the overlap degree of multiple plot data at the same time is analyzed. When the data overlap degree is high, the operation instructions and equipment feedback signals sent from the management system are recorded, and the fluctuation of the response success rate of the equipment executing the instructions and returning the confirmation signal within the specified time is analyzed. The accuracy of the management system operation execution is evaluated by comprehensively considering the data overlap degree and the response success rate fluctuation. If the accuracy is low, the system operation within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted based on the prediction results to optimize the system performance.

[0080] Example 2: A selenium sand melon growth monitoring and management system based on the Internet of Things described in this example includes a real-time data acquisition module, an overlap analysis module, an equipment response analysis module, an accuracy assessment module, and a dynamic adjustment module:

[0081] Real-time data acquisition module: divide the selenium sand melon growth area into several monitoring areas, and collect real-time data streams from plot sensors in each monitoring area. The real-time data streams include selenium sand melon growth environment data and fertilization operation data;

[0082] Overlap analysis module: Check the matching between the selenium sand melon growth environment data collected by the plot sensors and the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time;

[0083] Equipment response analysis module: If the overlap is high, record each operation instruction sent from the management system to the equipment and the feedback signal after the monitoring equipment executes it, and analyze the fluctuation of the response success rate of the equipment correctly executing the instruction and returning the confirmation signal within the specified time;

[0084] Accuracy evaluation module: evaluates the accuracy of management system operation execution based on the degree of overlap of selenium sand melon growth environment data collected by sensors in multiple plots at the same time and the fluctuation of the response success rate of the equipment to correctly execute instructions and return confirmation signals within the specified time;

[0085] Dynamic adjustment module: When the accuracy of management system operation execution is low, the accuracy of management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

[0086] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0087] It should be understood that the term "and / or" in this article is only 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 at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0088] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0089] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A selenium sand melon growth monitoring and management method based on the Internet of Things, characterized in that: The following steps are involved: S1: Divide the selenium sand melon growing area into several monitoring areas, and collect real-time data streams from sensors in each monitoring area, including selenium sand melon growth environment data and fertilization operation data; S2: Check whether the selenium sand melon growth environment data collected by the plot sensor matches the historical data range of the plot, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time; S3: If the overlap is high, record each operation instruction sent from the management system to the device and the feedback signal after the monitoring device executes it, and analyze the fluctuation of the response success rate of the device correctly executing the instruction and returning the confirmation signal within the specified time; S4: Evaluate the accuracy of management system operation execution based on the overlap of selenium sand melon growth environment data collected by multiple plot sensors at the same time and the fluctuation of the response success rate of the equipment correctly executing instructions and returning confirmation signals within the specified time; S5: When the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

2. A method for monitoring and managing the growth of selenium sand melon based on the Internet of Things according to claim 1, characterized in that: In S2, the matching situation between the growth environment data of selenium sand melon collected by the plot sensor and the historical data range of the plot is analyzed to generate the growth environment data matching anomaly index. The method for obtaining the growth environment data matching anomaly index is: The historical growth environment data obtained from the plot sensors, the historical data set is denoted as D = {x1, x2, ..., x n }; where x n Represents the nth data point. Use the kernel density estimation method to estimate the probability density function P(x) of historical data. The kernel density estimation formula is: n is the number of historical data points, h is the bandwidth of the kernel function, which controls the degree of smoothness, and K is the kernel function; for real-time data points x new Input the model and calculate its probability density value P(x new ), the formula is: The growth environment data matching anomaly index is calculated based on the probability value. The calculation expression is: AI = -log(P(x new )); where AI is the growth environment data matching anomaly index.

3. A selenium sand melon growth monitoring and management method based on the Internet of Things according to claim 2, characterized in that: In S3, the response success rate fluctuation of the device correctly executing the instruction and returning the confirmation signal within the specified time is analyzed to generate a response success rate fluctuation value. The method for obtaining the response success rate fluctuation value is: Collect the sequence data of the device's response success rate over time and record the success rate R in each time period t = {R1, R2, …, R n }, where R t is the response success rate in the tth time period; select the time window length n as the basic time period for calculating the fluctuation value, and calculate the average value μ of the response success rate in the time window R , the expression is: Calculate the standard deviation of the success rate σ R Indicates the fluctuation range: Calculate the response success rate fluctuation value, which is the ratio of the standard deviation to the mean, to measure the intensity of the fluctuation: Where AS is the response success rate fluctuation value.

4. A method for monitoring and managing the growth of selenium sand melon based on the Internet of Things according to claim 3, characterized in that: In S4, the accuracy of the management system operation execution is evaluated based on the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time and the fluctuation of the response success rate of the equipment to correctly execute the command and return the confirmation signal within the specified time, specifically: The growth environment data matching anomaly index and the response success rate fluctuation value are normalized, and the accuracy coefficient of the management system operation execution is calculated based on the normalized growth environment data matching anomaly index and the response success rate fluctuation value.

5. A method for monitoring and managing the growth of selenium sand melon based on the Internet of Things according to claim 4, characterized in that: The acquired accuracy coefficient of the management system operation execution is compared with the accuracy coefficient reference threshold preset under normal conditions in historical data. If the accuracy coefficient of the management system operation execution is greater than or equal to the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is high, and no warning signal is generated at this time; If the accuracy coefficient of the management system operation execution is less than the accuracy coefficient reference threshold, it means that the accuracy of the management system operation execution is low, and an early warning signal is generated at this time.

6. A method for monitoring and managing the growth of selenium sand melon based on the Internet of Things according to claim 1, characterized in that: In S5, when the accuracy of the management system operation execution is low, the accuracy of the management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results, specifically: Collect the accuracy coefficient A of the management system operation execution within a fixed time period n , the time series is expressed as: A={A1,A2,…,A n }; Among them, A n is the accuracy coefficient at the nth time point; the prediction formula based on ARIMA is: is the accuracy coefficient of the next time point predicted, φ i is the autoregression coefficient, which indicates the impact of past data on current predictions, θ j is the moving average coefficient, which indicates the influence of residual on prediction, ∈ t is the prediction error, p,q is the model order; based on the prediction results Dynamically adjust the control logic and decision threshold of the management system. The decision threshold adjustment formula is: T new Dynamically adjusted decision threshold, T base is the basic decision threshold, which indicates the standard threshold of the system under normal conditions. ref is the accuracy reference value, which indicates the accuracy coefficient of the system under normal conditions, and δ is the adjustment factor, which controls the adjustment range.

7. A method for monitoring and managing the growth of selenium sand melon based on the Internet of Things according to claim 6, characterized in that: According to the accuracy coefficient of the prediction Adjustments include: When the accuracy coefficient of the forecast If the error is less than the reference threshold of the predicted accuracy coefficient, an abnormal signal is triggered to prompt the operator to check the system operation status and adjust the system according to the predicted error. Dynamically optimize the prediction model and gradually improve the prediction accuracy.

8. A selenium sand melon growth monitoring and management system based on the Internet of Things, used to implement the selenium sand melon growth monitoring and management method based on the Internet of Things according to any one of claims 1 to 7, characterized in that: It includes real-time data acquisition module, overlap analysis module, equipment response analysis module, accuracy assessment module and dynamic adjustment module: Real-time data acquisition module: divide the selenium sand melon growth area into several monitoring areas, and collect real-time data streams from plot sensors in each monitoring area. The real-time data streams include selenium sand melon growth environment data and fertilization operation data; Overlap analysis module: Check the matching between the selenium sand melon growth environment data collected by the plot sensors and the historical data range of the plots, and determine the degree of overlap of the selenium sand melon growth environment data collected by multiple plot sensors at the same time; Equipment response analysis module: If the overlap is high, record each operation instruction sent from the management system to the equipment and the feedback signal after the monitoring equipment executes it, and analyze the fluctuation of the response success rate of the equipment correctly executing the instruction and returning the confirmation signal within the specified time; Accuracy evaluation module: evaluates the accuracy of management system operation execution based on the degree of overlap of selenium sand melon growth environment data collected by sensors in multiple plots at the same time and the fluctuation of the response success rate of the equipment to correctly execute instructions and return confirmation signals within the specified time; Dynamic adjustment module: When the accuracy of management system operation execution is low, the accuracy of management system operation execution within a fixed time period is predicted, and the control logic and decision threshold of the management system are dynamically adjusted according to the prediction results.

Images