Environment control method based on Internet of Things and meteorological soil moisture content monitoring

Abstract

The invention discloses an environment control method based on Internet of Things and meteorological soil moisture monitoring, and belongs to the technical field of environment intelligent control. Meteorological element data, soil moisture content time sequence data and environment execution equipment state data of a target area are collected through Internet of Things monitoring nodes, and data time sequence alignment and preliminary noise reduction processing are completed; calculating a soil moisture content time sequence disturbance entropy value by adopting a permutation entropy algorithm and dividing environment disturbance grades; a multi-target fitness function containing multi-dimensional constraints is constructed, and environment control parameter optimization is completed based on an improved sparrow search algorithm; and issuing a control instruction to the execution terminal, and synchronously collecting the regulated data to complete disturbance entropy updating and closed-loop regulation. According to the invention, the dynamic perception and accurate regulation and control of the environment state are realized, the scene adaptability of the control strategy is improved, the multi-dimensional regulation and control requirements are considered, and the comprehensive operation efficiency of environment control is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent environmental control technology, and in particular to an environmental control method based on the Internet of Things and meteorological moisture monitoring. Background Technology

[0002] With the deep integration of IoT sensing technology, intelligent data processing technology, and modern agricultural environmental management technology, intelligent environmental control based on meteorological and soil moisture monitoring has become a core technology application direction in smart agriculture, facility planting, and ecological vegetation management. Currently, the deployment and application of IoT monitoring nodes have achieved large-scale popularization within the industry, enabling real-time collection and transmission of multi-dimensional environmental data such as meteorological elements, soil moisture, and equipment operating status. The coverage, transmission efficiency, and collection dimensions of data collection have all been significantly improved. Simultaneously, time-series data analysis methods and swarm intelligence optimization algorithms have been gradually applied to the decision-making process of environmental control, enabling the analysis and processing of collected data and the optimization of environmental control parameters. This replaces traditional manual management and fixed threshold control modes, driving the continuous development of environmental control technology from automation to intelligence. The application scenarios of related technologies are constantly expanding, and the technical system is continuously improving through practical application.

[0003] In the practical application of existing technologies, multi-source heterogeneous monitoring data lacks standardized time-series alignment and preprocessing procedures after acquisition. Insufficient consistency in the temporal dimensions of different data types leads to data misalignment and noise interference, failing to provide high-quality basic data support for subsequent control decisions. Simultaneously, existing methods cannot effectively quantify the temporal disturbances caused by changes in meteorological elements on soil moisture, making it difficult to accurately classify actual environmental fluctuations. This results in environmental control decisions lacking constraints aligned with real-time environmental conditions, and the control strategies are insufficiently adapted to actual environmental needs. Furthermore, the optimization process for existing environmental control parameters struggles to simultaneously address the multi-dimensional requirements of control precision, energy consumption, and response speed. Conventional optimization algorithms have limitations in multi-objective solutions and cannot achieve real-time data feedback and dynamic updates of control strategies after control execution, easily leading to problems such as control lag and over-adjustment, making it difficult to achieve closed-loop, dynamic, and precise control of environmental conditions. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide an environmental control method based on the Internet of Things and meteorological moisture monitoring.

[0005] The objective of this invention is achieved through the following technical solution: An environmental control method based on the Internet of Things and meteorological soil moisture monitoring is provided, the method comprising the following steps: S1. Collect meteorological element data, soil moisture time series data and environmental execution equipment status data of the target area through IoT monitoring nodes, perform time series alignment processing on the meteorological element data, soil moisture time series data and environmental execution equipment status data, and perform preliminary noise reduction processing on the time series aligned data; S2. Perform permutation entropy calculation on the meteorological element data and soil moisture time series data after preliminary noise reduction processing to obtain the soil moisture time series disturbance entropy value, and classify the environmental disturbance level according to the soil moisture time series disturbance entropy value; S3. Construct a multi-objective fitness function, which includes control accuracy constraints, energy consumption constraints, and response speed constraints. Process the multi-objective fitness function based on the improved sparrow search algorithm. The improved sparrow search algorithm performs chaotic initialization operation, adaptive weight adjustment operation, and boundary constraint correction operation. S4. Using the environmental disturbance level as the input constraint for the improved sparrow search algorithm, the combination of environmental control parameters is determined by improving the sparrow search algorithm, the combination of environmental control parameters is converted into control commands, and the control commands are sent to the environmental regulation execution terminal through the Internet of Things communication link. The soil moisture time series data after regulation is collected and the soil moisture time series disturbance entropy value is updated.

[0006] Furthermore, in step S1, step S1.1. performs abnormal value identification processing on the raw meteorological element data, raw soil moisture time series data and raw environmental execution device status data collected by the IoT monitoring node, identifies abnormal data that deviates from the normal data range and performs removal processing. S1.2. Perform time axis matching processing on the meteorological element data, soil moisture time series data, and environmental execution equipment status data after removing abnormal data to complete the time series alignment processing. Perform smoothing filtering processing on the time series aligned data to complete the preliminary noise reduction processing and remove interference information generated during data transmission. Abnormal value identification processing, time axis matching processing, and smoothing filtering processing are performed in sequence. The data processed in the previous step is directly used as the input data for the next step. The processed meteorological element data, soil moisture time series data, and environmental execution equipment status data directly enter the permutation entropy calculation processing flow.

[0007] Furthermore, in step S2, step S2.1. Select the soil moisture time series data after preliminary noise reduction to construct a time series sequence, perform phase space reconstruction processing on the time series sequence, and perform symbolic sorting processing on the data after phase space reconstruction according to the numerical size to form a corresponding sorting feature sequence; S2.2. Perform probabilistic statistical processing on the sorted feature sequence, calculate the permutation entropy based on the probabilistic statistical processing results, compare the soil moisture temporal disturbance entropy value with the preset division conditions, and classify the environmental disturbance level of the target area based on the comparison results. Phase space reconstruction processing, symbolic sorting processing and probabilistic statistical processing are sequentially connected. The processed data is directly used for environmental disturbance level classification. The classified environmental disturbance level is directly input into the improved sparrow search algorithm as a constraint condition.

[0008] Furthermore, in step S3, step S3.1: Select a chaotic mapping sequence to generate initial population data, and use the chaotic mapping sequence to perform the population chaos initialization operation of the improved sparrow search algorithm, so that the initial population presents a uniform distribution state in the search space; S3.2. During the iterative optimization process of the improved sparrow search algorithm, an adaptive inertia weight adjustment operation is performed to dynamically adjust the search step size of the improved sparrow search algorithm. For parameters that exceed the preset parameter range during the iterative calculation of the improved sparrow search algorithm, a boundary constraint correction operation is performed to correct the out-of-bounds parameters to within the preset parameter range. The chaos initialization operation, the adaptive inertia weight adjustment operation, and the boundary constraint correction operation are executed in sequence, and the operation results are directly applied to the solution process of the multi-objective fitness function.

[0009] Furthermore, in step S4, step S4.1, the divided environmental disturbance levels are input into the improved sparrow search algorithm. The improved sparrow search algorithm matches the corresponding iteration mode according to the environmental disturbance level and determines the combination of environmental control parameters through iterative optimization calculation. S4.2. The determined combination of environmental control parameters is converted into standardized control commands according to the IoT communication protocol format. The standardized control commands are transmitted to the environmental regulation execution terminal through the IoT communication link. The environmental regulation execution terminal executes environmental regulation actions according to the standardized control commands, collects the regulated soil moisture time series data and transmits it back to the data processing terminal. Iterative optimization calculation, command format conversion and command transmission operations are executed in sequence. The transmitted soil moisture time series data is directly used for updating the soil moisture time series disturbance entropy value.

[0010] Furthermore, in step S1, the meteorological element data, soil moisture time series data, and environmental execution equipment status data collected by the IoT monitoring nodes are subjected to data synchronization processing. The collection timestamps and transmission frequencies of the meteorological element data, soil moisture time series data, and environmental execution equipment status data are unified, so that different types of monitoring data are summarized in the same time dimension. Data synchronization processing, abnormal data removal operation, and time series alignment processing are performed simultaneously. The meteorological element data, soil moisture time series data, and environmental execution equipment status data after synchronization processing are uniformly entered into the smoothing filtering process. The data after data synchronization processing directly participates in subsequent time series alignment operations and preliminary noise reduction operations.

[0011] Furthermore, in step S2, a time-series trend feature extraction operation is performed on the meteorological element data and soil moisture time-series data after preliminary noise reduction processing to extract the meteorological element change trend features and soil moisture time-series fluctuation features. The extracted time-series trend features are then integrated into the permutation entropy calculation process. Before the permutation entropy calculation, feature data screening and feature data integration operations are performed to remove redundant feature data that is unrelated to time-series disturbances. The permutation entropy calculation is completed based on the screened and integrated feature data. The time-series trend feature extraction operation is performed after the preliminary noise reduction processing, and the extracted and integrated feature data is directly used as the input data for the permutation entropy calculation.

[0012] Furthermore, in step S3, after the improved sparrow search algorithm completes a single iteration, a neighborhood perturbation update operation is performed on the position data corresponding to the leader bird in the improved sparrow search algorithm. New position data is generated within a preset interval around the leader bird's position, and the original leader bird position data is replaced by the newly generated position data. The neighborhood perturbation update operation is performed after the single iteration is completed, and the updated leader bird position data directly participates in the next round of iteration calculation of the improved sparrow search algorithm. The neighborhood perturbation update operation is performed in conjunction with the chaotic initialization operation and the adaptive inertia weight adjustment operation to jointly complete the solution of the multi-objective fitness function.

[0013] Furthermore, in step S3, the ratio of discoverer individuals to follower individuals in the improved sparrow search algorithm is adjusted according to the divided environmental disturbance level. The allocation rules of discoverers and followers are updated synchronously when the environmental disturbance level changes. During the iteration process of the improved sparrow search algorithm, population roles are allocated according to the adjusted ratio. Discoverers perform global search operations, and followers perform local search operations. The population role ratio adjustment operation is performed after the leader bird position data is updated. The adjusted role allocation rules are directly applied to the search operations in the subsequent iteration process of the improved sparrow search algorithm.

[0014] Furthermore, in step S4, soil moisture time-series data is collected after the environmental control execution terminal has been running. Based on the soil moisture time-series data, the permutation entropy calculation is re-executed to obtain the updated soil moisture time-series disturbance entropy value. Based on the updated soil moisture time-series disturbance entropy value, the environmental disturbance level is reclassified. The reclassified environmental disturbance level is input into the improved sparrow search algorithm that has completed the population role ratio adjustment. The iteration number and search range of the improved sparrow search algorithm are adjusted to match the iteration strategy corresponding to the current environmental state. The reclassified environmental disturbance level is directly used to adjust the running parameters of the improved sparrow search algorithm.

[0015] The beneficial effects of this invention are: (1) Through the complete process of multi-source data acquisition and preprocessing, time-series disturbance quantitative analysis, multi-objective optimization and closed-loop control, dynamic perception and precise control of environmental status are achieved, ensuring the continuity of the entire environmental control process and the effectiveness of data flow. (2) The perturbation characteristics of time series data are quantitatively analyzed by permutation entropy algorithm, and the environmental disturbance level is accurately classified. This provides constraints that fit the actual environmental state for optimization and improves the pertinence and adaptability of control strategies. (3) By optimizing and improving the sparrow search algorithm in multiple dimensions, and combining it with the multi-objective fitness function, the global optimization of control parameters is completed, taking into account the multi-dimensional requirements of control accuracy, energy consumption and response speed, and improving the overall operating efficiency of environmental control. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the steps of an environmental control method based on the Internet of Things and meteorological moisture monitoring; Figure 2 The following is a flowchart illustrating the specific steps of an environmental control method based on the Internet of Things and meteorological moisture monitoring, provided as an example. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 See Figure 1 This embodiment provides an environmental control method based on the Internet of Things and meteorological soil moisture monitoring, which includes the following steps: S1. Collect meteorological element data, soil moisture time series data and environmental execution equipment status data of the target area through IoT monitoring nodes, perform time series alignment processing on the meteorological element data, soil moisture time series data and environmental execution equipment status data, and perform preliminary noise reduction processing on the time series aligned data; S2. Perform permutation entropy calculation on the meteorological element data and soil moisture time series data after preliminary noise reduction processing to obtain the soil moisture time series disturbance entropy value, and classify the environmental disturbance level according to the soil moisture time series disturbance entropy value; S3. Construct a multi-objective fitness function, which includes control accuracy constraints, energy consumption constraints, and response speed constraints. Process the multi-objective fitness function based on the improved sparrow search algorithm. The improved sparrow search algorithm performs chaotic initialization operation, adaptive weight adjustment operation, and boundary constraint correction operation. S4. Using the environmental disturbance level as the input constraint for the improved sparrow search algorithm, the combination of environmental control parameters is determined by improving the sparrow search algorithm, the combination of environmental control parameters is converted into control commands, and the control commands are sent to the environmental regulation execution terminal through the Internet of Things communication link. The soil moisture time series data after regulation is collected and the soil moisture time series disturbance entropy value is updated.

[0019] In some embodiments, in step S1, step S1.1. performs abnormal value identification processing on the raw meteorological element data, raw soil moisture time series data and raw environmental execution device status data collected by the IoT monitoring node, identifies abnormal data that deviates from the normal data range and performs removal processing. S1.2. Perform time axis matching processing on the meteorological element data, soil moisture time series data, and environmental execution equipment status data after removing abnormal data to complete the time series alignment processing. Perform smoothing filtering processing on the time series aligned data to complete the preliminary noise reduction processing and remove interference information generated during data transmission. Abnormal value identification processing, time axis matching processing, and smoothing filtering processing are performed in sequence. The data processed in the previous step is directly used as the input data for the next step. The processed meteorological element data, soil moisture time series data, and environmental execution equipment status data directly enter the permutation entropy calculation processing flow.

[0020] In some embodiments, in step S2, step S2.1. Select the soil moisture time series data after preliminary noise reduction processing to construct a time series sequence, perform phase space reconstruction processing on the time series sequence, and perform symbolic sorting processing on the data after phase space reconstruction according to the numerical size to form a corresponding sorting feature sequence; S2.2. Perform probabilistic statistical processing on the sorted feature sequence, calculate the permutation entropy based on the probabilistic statistical processing results, compare the soil moisture temporal disturbance entropy value with the preset division conditions, and classify the environmental disturbance level of the target area based on the comparison results. Phase space reconstruction processing, symbolic sorting processing and probabilistic statistical processing are sequentially connected. The processed data is directly used for environmental disturbance level classification. The classified environmental disturbance level is directly input into the improved sparrow search algorithm as a constraint condition.

[0021] In some embodiments, in step S3, step S3.1. Select a chaotic mapping sequence to generate initial population data, and use the chaotic mapping sequence to perform a population chaos initialization operation of the improved sparrow search algorithm so that the initial population presents a uniform distribution state in the search space. S3.2. During the iterative optimization process of the improved sparrow search algorithm, an adaptive inertia weight adjustment operation is performed to dynamically adjust the search step size of the improved sparrow search algorithm. For parameters that exceed the preset parameter range during the iterative calculation of the improved sparrow search algorithm, a boundary constraint correction operation is performed to correct the out-of-bounds parameters to within the preset parameter range. The chaos initialization operation, the adaptive inertia weight adjustment operation, and the boundary constraint correction operation are executed in sequence, and the operation results are directly applied to the solution process of the multi-objective fitness function.

[0022] In some embodiments, in step S4, step S4.1, the divided environmental disturbance levels are input into the improved sparrow search algorithm, and the improved sparrow search algorithm matches the corresponding iteration mode according to the environmental disturbance level, and determines the combination of environmental control parameters through iterative optimization calculation; S4.2. The determined combination of environmental control parameters is converted into standardized control commands according to the IoT communication protocol format. The standardized control commands are transmitted to the environmental regulation execution terminal through the IoT communication link. The environmental regulation execution terminal executes environmental regulation actions according to the standardized control commands, collects the regulated soil moisture time series data and transmits it back to the data processing terminal. Iterative optimization calculation, command format conversion and command transmission operations are executed in sequence. The transmitted soil moisture time series data is directly used for updating the soil moisture time series disturbance entropy value.

[0023] In some embodiments, in step S1, data synchronization processing is performed on the meteorological element data, soil moisture time series data, and environmental execution device status data collected by the IoT monitoring node. The collection timestamps and transmission frequencies of the meteorological element data, soil moisture time series data, and environmental execution device status data are unified, so that different types of monitoring data are summarized in the same time dimension. Data synchronization processing, abnormal data removal operation, and time series alignment processing are performed simultaneously. The meteorological element data, soil moisture time series data, and environmental execution device status data after synchronization processing are uniformly entered into the smoothing filtering process. The data after data synchronization processing directly participates in the subsequent time series alignment operation and preliminary noise reduction operation.

[0024] In some embodiments, in step S2, a time-series trend feature extraction operation is performed on the meteorological element data and soil moisture time-series data after preliminary noise reduction processing to extract the meteorological element change trend features and soil moisture time-series fluctuation features. The extracted time-series trend features are then integrated into the permutation entropy calculation process. Before the permutation entropy calculation, a feature data screening operation and a feature data integration operation are performed to remove redundant feature data that is unrelated to time-series disturbances. The permutation entropy calculation is completed based on the screened and integrated feature data. The time-series trend feature extraction operation is performed after the preliminary noise reduction processing, and the extracted and integrated feature data is directly used as the input data for the permutation entropy calculation.

[0025] In some embodiments, in step S3, after the improved sparrow search algorithm completes a single iteration, a neighborhood perturbation update operation is performed on the position data corresponding to the leader bird in the improved sparrow search algorithm. New position data is generated within a preset interval around the leader bird's position, and the original leader bird position data is replaced by the newly generated position data. The neighborhood perturbation update operation is performed after the single iteration is completed, and the updated leader bird position data directly participates in the next round of iteration calculation of the improved sparrow search algorithm. The neighborhood perturbation update operation is performed in conjunction with the chaotic initialization operation and the adaptive inertia weight adjustment operation to jointly complete the solution of the multi-objective fitness function.

[0026] In some embodiments, in step S3, the ratio of discoverer individuals to follower individuals in the improved sparrow search algorithm is adjusted according to the divided environmental disturbance level. The allocation rules of discoverers and followers are updated synchronously when the environmental disturbance level changes. During the iteration process of the improved sparrow search algorithm, population roles are allocated according to the adjusted ratio. Discoverers perform global search operations, and followers perform local search operations. The population role ratio adjustment operation is performed after the leader bird position data is updated. The adjusted role allocation rules are directly applied to the search operations in the subsequent iteration process of the improved sparrow search algorithm.

[0027] In some embodiments, in step S4, soil moisture time-series data after the environmental regulation execution terminal is running is collected, permutation entropy calculation is re-executed based on the soil moisture time-series data to obtain updated soil moisture time-series disturbance entropy values, environmental disturbance levels are reclassified based on the updated soil moisture time-series disturbance entropy values, the reclassified environmental disturbance levels are input into the improved sparrow search algorithm that completes the population role ratio adjustment, the iteration number and search range of the improved sparrow search algorithm are adjusted, and the iteration strategy corresponding to the current environmental state is matched. The reclassified environmental disturbance levels are directly used to adjust the running parameters of the improved sparrow search algorithm.

[0028] Example 2 This embodiment proposes a specific implementation process for an environmental control method based on the Internet of Things (IoT) and meteorological soil moisture monitoring. This process relies on the real-time data acquisition capabilities of the IoT, combined with the temporal disturbance analysis characteristics of the permutation entropy algorithm and the optimization capabilities of the improved sparrow search algorithm, to achieve dynamic regulation of the target area's environment. The entire implementation process proceeds sequentially through data acquisition and preprocessing, temporal disturbance analysis, algorithm optimization, regulation execution, and closed-loop update. Each step is closely linked, with the processing results of the previous step directly serving as the input for the next, ensuring the continuity and effectiveness of the regulation process. Figure 2 As shown, the specific implementation process is as follows: S1. Data Acquisition and Preprocessing: This step is the foundation of the entire environmental control method. Its core function is to collect, filter, and process relevant data of the target area, providing a high-quality data foundation for subsequent time series analysis and algorithm optimization. The IoT monitoring node is the core data collection carrier in this step. This node is a sensing device used to collect environmental data. It can collect different types of environmental data through various sensing modules. In this embodiment, multiple types of data are collected through this type of node, and operations such as anomaly removal, time series alignment, and synchronization processing are performed in sequence. Each sub-step is carried out in order, and the result of the previous step is directly used as the input of the next step.

[0029] S1.1. Raw Data Acquisition and Anomaly Removal: Meteorological data, soil moisture time-series data, and environmental control equipment status data of the target area are collected through IoT monitoring nodes. These three types of data serve as the foundation for environmental control. Meteorological data reflects the meteorological conditions of the target area, soil moisture time-series data is collected in a time series, and environmental control equipment status data reflects the operational status of environmental control equipment. Anomaly identification processing is performed on the collected raw meteorological data, raw soil moisture time-series data, and raw environmental control equipment status data. This anomaly identification process identifies invalid data by using preset data ranges. In this embodiment, this method identifies and removes anomaly data that deviates from the normal data range. Removed anomaly data refers to invalid data that is outside the normal data distribution range, preventing such data from interfering with subsequent processing and ensuring the validity of the data foundation for subsequent steps. This allows subsequent data analysis and calculations to be based on real and valid environmental data.

[0030] S1.2. Data time-series alignment and preliminary noise reduction: Time axis matching processing is performed on meteorological element data, soil moisture time series data, and environmental execution equipment status data after removing abnormal data to complete time series alignment. Time axis matching processing is a processing method to keep different types of data consistent in the time dimension. This operation eliminates the time misalignment problem caused by the difference in the collection frequency of IoT monitoring nodes, allowing different types of data to be analyzed and utilized under the same time reference.

[0031] After time alignment, smoothing filtering is performed on the aligned data to complete preliminary noise reduction. Smoothing filtering is a conventional digital signal processing method whose core function is to eliminate random interference noise in the data. In this embodiment, this method removes interference information generated during data transmission, making the data more consistent with the actual environmental conditions and reducing the impact of data noise on subsequent calculation results. Anomaly identification, time axis matching, and smoothing filtering are executed sequentially. The data processed in the previous step is directly used as the input data for the next step. The processed meteorological element data, soil moisture time series data, and environmental execution equipment status data directly enter the permutation entropy calculation process.

[0032] S1.3. Multi-source data synchronization processing: Data synchronization processing is performed on meteorological element data, soil moisture time-series data, and environmental execution device status data collected by IoT monitoring nodes. This processing is a supplementary optimization operation to the time-series alignment processing, and its core purpose is to achieve consistency of multi-source data at the acquisition and transmission levels. In this implementation, this operation unifies the acquisition timestamps and transmission frequencies of meteorological element data, soil moisture time-series data, and environmental execution device status data, enabling different types of monitoring data to be aggregated in the same time dimension. Data synchronization processing, abnormal data removal, and time-series alignment processing are performed simultaneously. The synchronized meteorological element data, soil moisture time-series data, and environmental execution device status data are uniformly entered into a smoothing and filtering process. The synchronized data directly participates in subsequent time-series alignment and preliminary noise reduction operations, further ensuring the consistency of multi-source data in the time dimension, avoiding subsequent calculation deviations caused by asynchronous data acquisition and transmission, and improving the stability of the data foundation.

[0033] In some embodiments, data synchronization processing can be achieved by timestamp calibration. The collection time of various types of data is accurately calibrated by a unified network time reference. At the same time, the transmission frequency is adjusted according to the data collection characteristics, so that high-frequency data is downsampled according to preset rules and low-frequency data is supplemented according to preset rules, ensuring the synchronization of different types of data at the collection and transmission levels. Moreover, the rules for supplementing collection and downsampling can be flexibly set according to the data type and characteristics.

[0034] In some specific implementations, to address the issues of asynchronous multi-source data acquisition and low accuracy in anomaly data identification, specific numerical parameters were set for various operations of S1. The frequency of meteorological element data acquisition by the IoT monitoring node was 10 minutes / time, soil moisture time series data acquisition was 5 minutes / time, and environmental execution equipment status data acquisition was 2 minutes / time. During data synchronization processing, the acquisition timestamps of the three types of data were uniformly calibrated to the millisecond level to eliminate time deviations caused by different acquisition frequencies. Anomaly value identification and processing adopted the 3σ principle to set the normal data range, that is, calculating the mean and standard deviation of a single type of data, and judging values ​​that deviate from the mean ± 3 times the standard deviation as anomaly data and removing them. This numerical setting can balance the accuracy and comprehensiveness of anomaly data identification, avoiding the omission of anomaly data due to an overly wide threshold, and avoiding the mistaken deletion of valid data due to an overly narrow threshold.

[0035] After time-series alignment, a 5th-order moving average filter is used for initial noise reduction. The average of five consecutive data points is selected as the filtered data. This setting effectively eliminates random impulse interference in data transmission while preserving the temporal variation characteristics of the data itself, preventing data distortion due to excessively high filter orders. After all processing is complete, the data is integrated at a 1-minute time granularity to form a standardized time-series dataset. This time granularity balances the level of data detail with the workload of subsequent calculations, providing regular and efficient input data for subsequent permutation entropy calculations.

[0036] S2. Temporal Disturbance Analysis and Classification: This step, based on the high-quality data processed by S1, combines the permutation entropy algorithm to perform a quantitative analysis of soil moisture temporal disturbances and classify the corresponding environmental disturbance levels, providing constraints that fit the actual environmental conditions for subsequent algorithm optimization. Permutation entropy is the core algorithm of this step; it is a quantitative indicator used to analyze the complexity and degree of disturbance in time series data. By calculating this indicator, the fluctuation characteristics and changing patterns of time series data can be accurately reflected. In this embodiment, the soil moisture temporal disturbance entropy value is calculated using this algorithm to quantify the impact of meteorological element changes on soil moisture. Simultaneously, existing methods such as phase space reconstruction and feature extraction are combined to improve the accuracy of entropy value calculation. Each sub-step is carried out sequentially, and the processing results are directly used for classifying the environmental disturbance levels.

[0037] S2.1. Time Series Construction and Feature Processing: A time-series data set of soil moisture after preliminary denoising in S1 is selected to construct a time-series sequence. This time-series sequence, arranged chronologically, forms the basis for conducting time-series perturbation analysis. Phase space reconstruction is then performed on the constructed time-series sequence. Phase space reconstruction is a classic data processing method that maps one-dimensional time-series data to a higher-dimensional space. Its core function is to uncover the hidden intrinsic temporal characteristics and variation patterns within the one-dimensional time-series data. In this embodiment, this processing allows for a more comprehensive representation of the characteristics of the soil moisture time-series data, laying the foundation for subsequent symbolic sorting and entropy calculation.

[0038] After the phase space reconstruction is completed, the reconstructed data is subjected to symbolic sorting according to its numerical value. Symbolic sorting is a process that converts continuous numerical data into discrete sorting features. Through this process, a corresponding sorting feature sequence is formed, which transforms complex numerical changes into intuitive sorting features, facilitating subsequent probability statistics and permutation entropy calculations.

[0039] S2.2. Probability Statistics and Permutation Entropy Calculation: The sorting feature sequence obtained in S2.1 is subjected to probabilistic statistical processing. Probabilistic statistical processing is a conventional mathematical processing method for statistically analyzing the probability distribution of different features. In this embodiment, this processing is used to statistically analyze the probability distribution of different sorting features in the sequence, forming a complete probability distribution result. Based on the probabilistic statistical processing result, permutation entropy is calculated. According to the inherent calculation logic of the permutation entropy algorithm, combined with the probability distribution result, the temporal perturbation entropy value of soil moisture is calculated. The magnitude of this entropy value is directly related to the degree of soil moisture perturbation. The change in entropy value can accurately reflect the degree of influence of meteorological element changes on soil moisture. The entropy value calculation result is the core basis for subsequent environmental perturbation level classification.

[0040] S2.3. Temporal Feature Extraction and Data Filtering: The meteorological element data and soil moisture time-series data after preliminary noise reduction in S1 are subjected to time-series trend feature extraction. Time-series trend feature extraction is a method for mining the changing trends of data in the time dimension. In this embodiment, this operation extracts the changing trend features of meteorological elements and the time-series fluctuation features of soil moisture. These two types of features reflect the changing patterns of meteorological elements and soil moisture in the time dimension, respectively. The extracted time-series trend features are integrated into the permutation entropy calculation process. Before the permutation entropy calculation, feature data filtering and feature data integration operations are performed. Feature data filtering is a processing method to remove redundant features, and feature data integration is a processing method to fuse effective features. These two operations remove redundant feature data that is unrelated to time-series disturbances, reducing the interference of invalid data on the calculation results. The permutation entropy calculation is completed based on the filtered and integrated feature data. The time-series trend feature extraction operation is performed after the initial noise reduction processing. The extracted and integrated feature data is directly used as the input data for the permutation entropy calculation. Through feature extraction and filtering, not only can the amount of computation of redundant data be reduced, but the permutation entropy calculation can also be more focused on the core features related to time-series perturbations, thereby improving the accuracy of the permutation entropy calculation results.

[0041] S2.4. Classification of Environmental Disturbance Levels: The soil moisture temporal disturbance entropy values ​​obtained in S2.2 are compared with preset classification conditions. These conditions are entropy value interval division rules set according to environmental regulation needs. Based on the comparison results, the environmental disturbance level of the target area is classified. The environmental disturbance level characterizes the degree of temporal fluctuation of soil moisture caused by changes in meteorological elements; different levels correspond to different fluctuation states of soil moisture. Phase space reconstruction processing, symbolic sorting processing, and probabilistic statistical processing are sequentially performed. The processed data is directly used for environmental disturbance level classification. The classified environmental disturbance levels are directly input into the subsequent improved sparrow search algorithm as constraints, providing corresponding environmental state basis for the algorithm's optimization and ensuring that the algorithm's optimization process closely matches the actual environmental disturbance situation.

[0042] In some embodiments, the preset classification conditions for environmental disturbance levels can be set according to the historical environmental data of the target area and the actual control needs. The classified levels can be divided into multiple different fluctuation levels. Each level corresponds to a set of constraint parameters for optimization by the improved sparrow search algorithm, so that the algorithm can match the corresponding optimization strategy according to different disturbance levels and improve the targeting of the algorithm optimization.

[0043] In some specific implementations, to address issues such as insufficient feature mining during phase space reconstruction, low accuracy in permutation entropy calculation, and lack of clear standards for perturbation level classification, specific numerical parameters were set for the core operations of S2. During phase space reconstruction, the CC method was used to determine an embedding dimension of 6 and a delay time of 3. This numerical combination effectively mines the high-dimensional features of soil moisture time-series data, avoiding both excessively high computational load due to an excessively high embedding dimension and incomplete feature mining due to an excessively low embedding dimension. During symbolic sorting, the phase space-reconstructed data was divided into 8 equidistant numerical intervals, mapping each data point to a symbol identifier within the corresponding interval, forming a sorting feature sequence of length 100. The setting of 8 intervals balances data discriminability and computational complexity, while the sequence length of 100 ensures the effectiveness of probability statistics. During probability statistics processing, 50 consecutive data points were selected as a sample window, with a sliding step size of 10 data points. Sliding window probability statistics were performed on the sorting feature sequence to calculate the probability value of each symbol combination. The permutation entropy calculation is based on this probability value, and the resulting soil moisture temporal disturbance entropy value ranges from 0 to 1. Subsequently, the entropy value range is divided into 4 environmental disturbance levels: 0-0.3 represents steady state, 0.3-0.6 represents slight disturbance, 0.6-0.9 represents moderate disturbance, and above 0.9 represents severe disturbance. The numerical setting of this level division can accurately distinguish different degrees of soil moisture fluctuations, providing clear and explicit constraints for subsequent algorithm optimization, allowing the algorithm to match differentiated optimization strategies according to different levels.

[0044] S3. Algorithm Construction and Parameter Optimization: This step is the core optimization step of the entire environmental control method. By constructing a multi-objective fitness function and combining it with an improved sparrow search algorithm, the environmental control parameters are optimized, providing specific parameter basis for subsequent environmental regulation. The sparrow search algorithm is a swarm intelligence optimization algorithm that simulates the foraging and anti-predation behavior of sparrows. The improved sparrow search algorithm is an optimized and improved algorithm based on the original sparrow search algorithm. In this embodiment, by performing optimization operations such as chaotic initialization, weight adjustment, and leader bird update on the algorithm, the algorithm's optimization ability and convergence efficiency are improved. At the same time, a multi-objective fitness function is constructed as an evaluation criterion for algorithm optimization, so that the optimization results meet the multifaceted needs of environmental regulation.

[0045] S3.1. Algorithm Population Chaos Initialization: Chaotic mapping sequences are numerical sequences with randomness and ergodicity. The uniform numerical distribution of such sequences effectively enhances data diversity. This sequence is selected to generate the initial population data for the improved sparrow search algorithm. The chaotic mapping sequence is then used to perform chaotic initialization of the population in the improved sparrow search algorithm, ensuring a uniform distribution of the initial population within the search space. The population is the foundation for the sparrow search algorithm's optimization; the distribution of the initial population directly affects the algorithm's optimization performance. Chaotic initialization effectively improves the diversity of the initial population, preventing the algorithm from getting trapped in local optima in the early stages of optimization, and laying a solid population foundation for the algorithm's global optimization. The result of this operation is directly applied to the subsequent iterative optimization process of the algorithm.

[0046] S3.2. Dynamic adjustment and correction of algorithm parameters: In the iterative optimization process of the improved sparrow search algorithm, an adaptive inertia weight adjustment operation is performed. The adaptive inertia weight is a weight coefficient that dynamically changes according to the algorithm's iteration progress. The magnitude of this coefficient directly affects the algorithm's search step size. In this embodiment, this operation dynamically adjusts the search step size of the improved sparrow search algorithm, allowing the algorithm to maintain a larger search step size in the early stages of iteration to achieve a global search, and to reduce the search step size in the later stages of iteration to achieve a precise search in a local range, making the algorithm's search strategy more aligned with the iteration process. Simultaneously, a boundary constraint correction operation is performed on parameters that exceed the preset parameter range during the iterative calculation of the improved sparrow search algorithm. This corrects out-of-bounds parameters to within the preset parameter range, preventing algorithm iteration interruption or distortion of optimization results due to parameter out-of-bounds errors. The chaos initialization operation, adaptive inertia weight adjustment operation, and boundary constraint correction operation are executed sequentially. The operation results are directly applied to the solution process of the multi-objective fitness function, ensuring the parameter stability of the algorithm's iterative process.

[0047] In some embodiments, the boundary constraint correction operation of the improved sparrow search algorithm can be implemented by truncation correction, which directly corrects parameters that exceed the preset parameter range to the boundary value of the range; or by reflection correction, which maps out-of-bounds parameters back to the parameter range according to preset reflection rules. Both methods can effectively correct parameters and ensure stable iteration of the algorithm. The appropriate method can be flexibly selected according to the actual algorithm optimization needs.

[0048] S3.3. Update the neighborhood perturbation of the leader bird's position: After the improved sparrow search algorithm completes a single iteration, a neighborhood perturbation update operation is performed on the position data corresponding to the leader bird in the algorithm. The leader bird is the individual in the sparrow search algorithm responsible for the global search, and its position data corresponds to the candidate optimal solution in the algorithm's optimization process. The position state of the leader bird directly affects the algorithm's global optimization capability. In this embodiment, new position data is generated within a preset interval around the leader bird's position, and the newly generated position data replaces the original leader bird position data. The neighborhood perturbation update operation is performed after the single iteration is completed, and the updated leader bird position data directly participates in the next round of iteration calculation of the improved sparrow search algorithm.

[0049] The neighborhood perturbation update operation, along with the chaotic initialization operation and the adaptive inertia weight adjustment operation, are executed in conjunction to solve the multi-objective fitness function. This operation can effectively prevent the leader bird position from getting stuck in a local optimum, expand the global search range of the algorithm, and increase the probability of the algorithm finding the global optimum.

[0050] S3.4. Algorithm Population Role Ratio Adjustment: Based on the environmental disturbance levels divided in S2, the ratio of discoverer individuals to follower individuals in the improved sparrow search algorithm is adjusted. Discoverer individuals are those that perform global search in the algorithm, while follower individuals are those that perform local search in the algorithm. The ratio of the two types of individuals directly affects the search efficiency and optimization effect of the algorithm.

[0051] In this embodiment, the allocation rules for discoverers and followers are updated synchronously when the level of environmental disturbance changes. During the iteration process of the improved sparrow search algorithm, population roles are allocated according to the adjusted ratio. Discoverers perform global search operations, while followers perform local search operations. The population role ratio adjustment operation is performed after the leader bird position data is updated. The adjusted role allocation rules are directly applied to the search operations in subsequent iterations of the improved sparrow search algorithm. By dynamically adjusting the population role ratio, the algorithm's search strategy is adapted to the current level of environmental disturbance. When the soil moisture disturbance level is high, the proportion of discoverers is increased to improve global search capability; when the disturbance level is low, the proportion of followers is increased to improve local precision search capability, thereby improving the overall optimization efficiency of the algorithm.

[0052] S3.5. Construction and solution of multi-objective fitness function: A multi-objective fitness function is constructed. This function is the core function used to evaluate the optimization results of the improved sparrow search algorithm. It serves as the evaluation standard and constraint basis for algorithm optimization. In this embodiment, the multi-objective fitness function includes control accuracy constraints, energy consumption constraints, and response speed constraints. These three types of constraints correspond to the three aspects of environmental control requirements: accuracy, energy consumption cost, and response efficiency, respectively, so that the optimization results of the algorithm can take into account multiple objectives of environmental control.

[0053] The environmental disturbance levels defined in S2 are used as input constraints for the improved sparrow search algorithm. The multi-objective fitness function is processed based on the improved sparrow search algorithm. The function is solved by iterative optimization calculation of the algorithm. All optimization operations of the algorithm revolve around solving the function until the algorithm meets the preset convergence conditions. The optimal position of the population obtained at this time is the combination of environmental control parameters that satisfies all constraints, providing specific and accurate parameter basis for subsequent environmental regulation.

[0054] In some embodiments, the various constraints of the multi-objective fitness function can be set with different weight ratios according to the actual environmental control needs. The core control needs can be prioritized by adjusting the weights. At the same time, the number of iterations of the improved sparrow search algorithm can be dynamically set according to the level of environmental disturbance. Under high disturbance levels, the number of iterations can be appropriately increased to improve the accuracy of the optimization results, while under low disturbance levels, the number of iterations can be appropriately reduced to reduce the computational load of the algorithm.

[0055] In some specific implementations, to address the issues of low optimization efficiency, susceptibility to local optima, and lack of clear basis for adjusting population role ratios in the improved sparrow search algorithm, specific numerical values ​​were set for the S3 algorithm parameters. The population size of the improved sparrow search algorithm was set to 40, and the maximum number of iterations was set to 100. This combination of values ​​balances the algorithm's optimization accuracy and computational time. The population size of 40 ensures population diversity, and the maximum number of iterations of 100 provides sufficient optimization space. Chaotic initialization uses a logistic chaotic mapping to generate initial population data, with the mapping parameter set to 3.9. This parameter ensures good traversal of the chaotic sequence, allowing the initial population to be evenly distributed within the search space of [-1,1]. The adaptive inertia weight uses a linear decreasing method, with an initial value of 0.9, linearly decreasing to 0.1 when the maximum number of iterations is reached. This numerical change allows the algorithm to perform large-step global search in the early stages and small-step local precise search in the later stages. When updating the neighborhood of the leader bird's position, new position data is randomly generated within a ±0.05 interval around the leader bird's current position. This interval setting avoids sudden changes in the leader bird's position and effectively escapes local optima. Boundary constraint correction limits the parameter interval to [0,1], and parameters outside this interval are directly corrected to the interval boundary values. The population role ratio is dynamically adjusted according to the level of environmental disturbance. Under steady state, the ratio of discoverers to followers is 1:4, under slight disturbances it is 1:3, under moderate disturbances it is 1:2, and under severe disturbances it is 2:1. This ratio setting allows the algorithm to match the optimal search strategy under different levels of disturbance. It improves the global search capability under high disturbances and improves the local optimization accuracy under low disturbances, making the solution of the multi-objective fitness function more in line with actual control needs.

[0056] S4. Regulation Execution and Entropy Update: This step is the execution and closed-loop step of the entire environmental control method. The combination of environmental control parameters obtained in S3 is converted into executable control instructions, which are sent to the execution terminal through the Internet of Things and the control actions are executed. At the same time, the environmental data after the control is collected, the soil moisture temporal disturbance entropy value is updated, and the closed-loop management of environmental control is realized. This allows the environmental control to be dynamically adjusted according to the changes in the actual environmental state. Each sub-step is carried out in sequence to form a complete control execution and feedback process.

[0057] S4.1. Algorithm Iterative Optimization and Parameter Determination: The environmental disturbance levels defined in S2 are input into the improved sparrow search algorithm. The improved sparrow search algorithm matches the corresponding iteration mode based on the environmental disturbance level. Different disturbance levels correspond to different algorithm iteration parameters and search strategies. The combination of environmental control parameters is determined through iterative optimization. The iterative optimization process involves the algorithm continuously updating the population position and solving the multi-objective fitness function. During the iteration process, the algorithm continuously optimizes the population position based on the evaluation results of the multi-objective fitness function until the algorithm meets the preset convergence condition. The optimal population position obtained at this point is the required combination of environmental control parameters. This parameter combination serves as the specific basis for subsequent environmental regulation, conforming to the current environmental disturbance state and regulation requirements.

[0058] S4.2. Control command conversion and transmission: The environmental control parameter combination determined in S3 is converted into standardized control commands according to the IoT communication protocol format. The IoT communication protocol is a set of rules for data transmission and command interaction between IoT devices. Different IoT devices follow a unified protocol to achieve interconnection. In this embodiment, the commands converted according to this rule can be accurately identified by the environmental control execution terminal. The standardized control commands are transmitted to the environmental control execution terminal through the IoT communication link. The environmental control execution terminal executes the corresponding environmental control actions according to the standardized control commands. Iterative optimization calculation, command format conversion, and command transmission operations are executed sequentially to ensure that the control commands can be accurately and quickly transmitted to the execution terminal and effectively executed, so that the control parameters obtained by the algorithm optimization can be effectively implemented into environmental control actions.

[0059] S4.3. Data Acquisition and Transmission after Adjustment: After the environmental control execution terminal completes the control actions, IoT monitoring nodes collect time-series data on soil moisture after the control. This data reflects the actual effect of the control actions and directly demonstrates the degree of impact of the control actions on soil moisture. The collected time-series data on soil moisture after the control is transmitted back to the data processing terminal, which is a terminal device that performs various data calculations and analyses. The transmitted soil moisture time-series data is directly used to update the entropy value of subsequent soil moisture time-series disturbances. By collecting and transmitting this data, the actual impact of the control actions on the environment can be grasped in a timely manner, providing real and effective data support for subsequent closed-loop control, allowing environmental control to be adjusted according to the actual effects.

[0060] S4.4. Entropy Update and Algorithm Strategy Adjustment: Based on the returned soil moisture time-series data after regulation, the permutation entropy calculation is re-executed. Following the calculation logic of permutation entropy in S2, the updated soil moisture time-series disturbance entropy value is obtained. This entropy value is the core indicator reflecting the soil moisture disturbance state after regulation. The environmental disturbance level is reclassified based on the updated soil moisture time-series disturbance entropy value, with the classification rules consistent with the preset classification conditions in S2. The reclassified environmental disturbance level is input into the improved sparrow search algorithm for adjusting population role ratios. The iteration count and search range of the improved sparrow search algorithm are adjusted according to the new environmental disturbance level, matching the iteration strategy corresponding to the current environmental state. The recalculated disturbance entropy value is directly used as the basis for reclassifying the environmental disturbance level, and the reclassified level data is directly used to adjust the operating parameters of the improved sparrow search algorithm. This step achieves a closed-loop update of the regulation process, enabling environmental regulation to dynamically adjust with changes in the actual environmental state, allowing environmental control methods to continuously adapt to the constantly changing environmental state.

[0061] In some embodiments, the frequency of entropy updates can be set according to the level of environmental disturbance. Under high disturbance levels, the frequency of entropy updates can be increased to capture subtle changes in the environmental state in a timely manner, allowing the control strategy to respond quickly. Under low disturbance levels, the update frequency can be appropriately reduced to reduce the workload of data processing and algorithm calculation, thereby reducing the overall operating cost of the method while ensuring the control effect.

[0062] In some specific implementations, to address issues such as high transmission latency of control commands, lack of differentiation in closed-loop update frequency, and poor adaptability of algorithm iteration counts to disturbance levels, specific numerical parameters were set for the S4 operation. The MQTT IoT communication protocol was used for control command conversion, with the baud rate set to 115200bps. This parameter ensures the speed and stability of command transmission, guaranteeing that the latency from the data processing end to the environmental control execution terminal does not exceed 2 seconds, meeting the real-time requirements of control. Upon receiving the command, the environmental control execution terminal initiates the corresponding control action within 5 seconds, with the accuracy error controlled within ±2%, ensuring that the control action accurately matches the control parameter requirements. Post-control data acquisition begins 10 seconds after the action is completed to avoid environmental data fluctuations affecting the acquisition results. The acquired soil moisture time-series data is transmitted back via the IoT communication link, with the data packet size limited to 128 bytes to ensure transmission efficiency. The entropy update frequency is set differently based on the level of environmental disturbance. Under steady state, the soil moisture temporal disturbance entropy is updated once every 30 minutes; under slight disturbance, once every 15 minutes; under moderate disturbance, once every 5 minutes; and under severe disturbance, once every 1 minute. This frequency setting allows the closed-loop update to control computational costs while ensuring responsiveness. The number of iterations of the improved sparrow search algorithm is also dynamically adjusted according to the level of disturbance: 20 iterations under steady state, 40 iterations under slight disturbance, 70 iterations under moderate disturbance, and 100 iterations under severe disturbance. This numerical setting allows the algorithm to obtain precise control parameter combinations with optimal computational cost under different environmental conditions, achieving a balance between control effect and computational cost.

[0063] This embodiment achieves dynamic environmental control based on the Internet of Things and meteorological moisture monitoring through a complete process of data acquisition and preprocessing, time-series disturbance analysis, algorithm optimization, control execution, and closed-loop update. The entire technical solution relies on existing mature algorithms and data processing methods, with each step closely linked and data flow smoothly, effectively ensuring the consistency and accuracy of environmental control. First, this embodiment performs a series of preprocessing operations on the collected multi-source data, such as anomaly removal, time-series alignment, and synchronization processing, effectively filtering out invalid data and interference information, improving the quality of basic data, and providing reliable data support for subsequent time-series disturbance analysis and algorithm optimization. This effect can be directly achieved through the basic steps of data acquisition and preprocessing, ensuring that all subsequent calculations and analyses are based on high-quality data, reducing calculation bias caused by data problems. Secondly, by performing perturbation analysis on soil moisture time-series data using the permutation entropy algorithm, combined with phase space reconstruction and feature extraction, the degree of disturbance to soil moisture caused by meteorological element changes was accurately quantified, and corresponding environmental disturbance levels were classified. This provided constraints that fit the actual environmental conditions for algorithm optimization, ensuring that the optimization process was not divorced from the actual environmental situation and improving the targeting of algorithm optimization. Furthermore, the improved sparrow search algorithm, through multiple optimization operations such as chaotic initialization, leader bird position update, and population role ratio adjustment, effectively improved the problem of the original algorithm easily getting trapped in local optima, enhanced the algorithm's global optimization ability and convergence efficiency, and could quickly solve for environmental control parameter combinations that meet multi-objective constraints within a reasonable computational load. This made the control parameters more in line with actual environmental control needs, balancing control accuracy, energy consumption, and response speed. Meanwhile, this embodiment constructs a complete closed-loop control process. By collecting soil moisture data after control, the entropy value is recalculated, the disturbance level is classified, and the algorithm's iteration strategy is adjusted accordingly. This allows environmental control to dynamically adjust with changes in environmental conditions, improving the adaptability and flexibility of control and enabling the environmental control method to continuously adapt to different environmental disturbance states. Furthermore, the synchronous processing of multi-source data and the accurate extraction of time-series features further reduce computational bias and improve the stability of the entire control process. This ensures that various environmental control operations are accurately implemented, improving the effectiveness of environmental control while also reasonably controlling energy consumption and response time. Under multi-objective constraints, environmental control achieves comprehensive improvement in effectiveness. The various steps of the entire method cooperate with each other to form a complete and efficient environmental control system, effectively realizing intelligent environmental control based on the Internet of Things and meteorological moisture monitoring.

[0064] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. An environmental control method based on the Internet of Things and meteorological soil moisture monitoring, characterized in that, Includes the following steps: S1. Collect meteorological element data, soil moisture time series data and environmental execution equipment status data of the target area through IoT monitoring nodes, perform time series alignment processing on the meteorological element data, soil moisture time series data and environmental execution equipment status data, and perform preliminary noise reduction processing on the time series aligned data; S2. Perform permutation entropy calculation on the meteorological element data and soil moisture time series data after preliminary noise reduction processing to obtain the soil moisture time series disturbance entropy value, and classify the environmental disturbance level according to the soil moisture time series disturbance entropy value; S3. Construct a multi-objective fitness function, which includes control accuracy constraints, energy consumption constraints, and response speed constraints. Process the multi-objective fitness function based on the improved sparrow search algorithm. The improved sparrow search algorithm performs chaotic initialization operation, adaptive weight adjustment operation, and boundary constraint correction operation. S4. Using the environmental disturbance level as the input constraint for the improved sparrow search algorithm, the combination of environmental control parameters is determined by improving the sparrow search algorithm, the combination of environmental control parameters is converted into control commands, and the control commands are sent to the environmental regulation execution terminal through the Internet of Things communication link. The soil moisture time series data after regulation is collected and the soil moisture time series disturbance entropy value is updated.

2. The method according to claim 1, characterized in that, In step S1, step S1.

1. Perform abnormal value identification processing on the raw meteorological element data, raw soil moisture time series data and raw environmental execution device status data collected by the IoT monitoring node, identify abnormal data that deviates from the normal data range and perform removal processing. S1.

2. Perform time axis matching processing on the meteorological element data, soil moisture time series data, and environmental execution equipment status data after removing abnormal data to complete the time series alignment processing. Perform smoothing filtering processing on the time series aligned data to complete the preliminary noise reduction processing and remove interference information generated during data transmission. Abnormal value identification processing, time axis matching processing, and smoothing filtering processing are performed in sequence. The data processed in the previous step is directly used as the input data for the next step. The processed meteorological element data, soil moisture time series data, and environmental execution equipment status data directly enter the permutation entropy calculation processing flow.

3. The method according to claim 1, characterized in that, In step S2, step S2.1: Select the soil moisture time series data after preliminary noise reduction to construct a time series sequence, perform phase space reconstruction processing on the time series sequence, and perform symbolic sorting processing on the data after phase space reconstruction according to the numerical size to form a corresponding sorted feature sequence; S2.

2. Perform probabilistic statistical processing on the sorted feature sequence, calculate the permutation entropy based on the probabilistic statistical processing results, compare the soil moisture temporal disturbance entropy value with the preset division conditions, and classify the environmental disturbance level of the target area based on the comparison results. Phase space reconstruction processing, symbolic sorting processing and probabilistic statistical processing are sequentially connected. The processed data is directly used for environmental disturbance level classification. The classified environmental disturbance level is directly input into the improved sparrow search algorithm as a constraint condition.

4. The method according to claim 1, characterized in that, In step S3, step S3.1: Select a chaotic mapping sequence to generate initial population data, and use the chaotic mapping sequence to perform the population chaos initialization operation of the improved sparrow search algorithm so that the initial population presents a uniform distribution state in the search space; S3.

2. During the iterative optimization process of the improved sparrow search algorithm, an adaptive inertia weight adjustment operation is performed to dynamically adjust the search step size of the improved sparrow search algorithm. For parameters that exceed the preset parameter range during the iterative calculation of the improved sparrow search algorithm, a boundary constraint correction operation is performed to correct the out-of-bounds parameters to within the preset parameter range. The chaos initialization operation, the adaptive inertia weight adjustment operation, and the boundary constraint correction operation are executed in sequence, and the operation results are directly applied to the solution process of the multi-objective fitness function.

5. The method according to claim 1, characterized in that, In step S4, step S4.1, the environmental disturbance levels are input into the improved sparrow search algorithm. The improved sparrow search algorithm matches the corresponding iteration mode according to the environmental disturbance levels and determines the combination of environmental control parameters through iterative optimization calculation. S4.

2. The determined combination of environmental control parameters is converted into standardized control commands according to the IoT communication protocol format. The standardized control commands are transmitted to the environmental regulation execution terminal through the IoT communication link. The environmental regulation execution terminal executes environmental regulation actions according to the standardized control commands, collects the regulated soil moisture time series data and transmits it back to the data processing terminal. Iterative optimization calculation, command format conversion and command transmission operations are executed in sequence. The transmitted soil moisture time series data is directly used for updating the soil moisture time series disturbance entropy value.

6. The method according to claim 2, characterized in that, In step S1, the meteorological element data, soil moisture time series data, and environmental execution equipment status data collected by the IoT monitoring nodes are processed for data synchronization. The collection timestamps and transmission frequencies of the meteorological element data, soil moisture time series data, and environmental execution equipment status data are unified, so that different types of monitoring data are summarized in the same time dimension. Data synchronization processing, abnormal data removal, and time series alignment processing are performed simultaneously. The meteorological element data, soil moisture time series data, and environmental execution equipment status data after synchronization processing are uniformly entered into the smoothing filtering process. The data after data synchronization processing directly participates in subsequent time series alignment and preliminary noise reduction operations.

7. The method according to claim 3, characterized in that, In step S2, a time-series trend feature extraction operation is performed on the meteorological element data and soil moisture time-series data after preliminary noise reduction processing to extract the change trend features of meteorological elements and the time-series fluctuation features of soil moisture. The extracted time-series trend features are then integrated into the permutation entropy calculation process. Before the permutation entropy calculation, feature data screening and feature data integration operations are performed to remove redundant feature data that is not related to time-series disturbances. The permutation entropy calculation is completed based on the screened and integrated feature data. The time-series trend feature extraction operation is performed after the preliminary noise reduction processing, and the extracted and integrated feature data is directly used as the input data for the permutation entropy calculation.

8. The method according to claim 4, characterized in that, In step S3, after the improved sparrow search algorithm completes a single iteration, a neighborhood perturbation update operation is performed on the position data corresponding to the leader bird in the improved sparrow search algorithm. New position data is generated within a preset interval around the leader bird's position, and the original leader bird position data is replaced by the newly generated position data. The neighborhood perturbation update operation is performed after the single iteration is completed. The updated leader bird position data directly participates in the next round of iteration calculation of the improved sparrow search algorithm. The neighborhood perturbation update operation is performed in conjunction with the chaotic initialization operation and the adaptive inertia weight adjustment operation to jointly complete the solution of the multi-objective fitness function.

9. The method according to claim 8, characterized in that, In step S3, the ratio of discoverer individuals to follower individuals in the improved sparrow search algorithm is adjusted according to the environmental disturbance level after division. The allocation rules of discoverers and followers are updated synchronously when the environmental disturbance level changes. During the iteration process of the improved sparrow search algorithm, population roles are allocated according to the adjusted ratio. Discoverers perform global search operations, and followers perform local search operations. The population role ratio adjustment operation is performed after the leader bird position data is updated. The adjusted role allocation rules are directly applied to the search operations in the subsequent iteration process of the improved sparrow search algorithm.

10. The method according to claim 9, characterized in that, In step S4, soil moisture time-series data is collected after the environmental control execution terminal is running. Based on the soil moisture time-series data, the permutation entropy is recalculated to obtain the updated soil moisture time-series disturbance entropy value. Based on the updated soil moisture time-series disturbance entropy value, the environmental disturbance level is reclassified. The reclassified environmental disturbance level is input into the improved sparrow search algorithm to complete the population role ratio adjustment. The iteration number and search range of the improved sparrow search algorithm are adjusted to match the iteration strategy corresponding to the current environmental state. The reclassified environmental disturbance level is directly used to adjust the running parameters of the improved sparrow search algorithm.

Images