Intelligent gate remote control method and system based on remote sensing and internet of things technology

Abstract

The application discloses a remote control method and system for an intelligent gate based on remote sensing and Internet of Things technology, and belongs to the technical field of gate control. The method comprises the following steps: capturing spectral data of a target area by a remote sensing sensor carried by a UAV, and capturing environmental data of the target area by environmental sensors deployed at key nodes in the target area, so as to construct a multidimensional data set; after data preprocessing of the multidimensional data set, gradually screening and marking key indicator abnormal points to obtain an abnormal aggregation area, and fitting the operation parameters of the gate in the abnormal aggregation area; generating an execution instruction of the gate, monitoring the execution of the gate by a state sensor, performing abnormal early warning on the gate with execution abnormalities, and transmitting the execution result to a data processing module through an Internet of Things device to update the execution instruction; and realizing real-time monitoring and analysis of the key indicator abnormal points, so as to optimize the use of the gate and improve the effective utilization efficiency of water resources.

Description

Technical Field

[0001] This invention relates to the field of gate control technology, and more specifically to a method and system for remote control of intelligent gates based on remote sensing and Internet of Things (IoT) technologies. Background Technology

[0002] With global population growth and increasing agricultural water demand, the efficient management and utilization of water resources has become a crucial issue for sustainable agricultural development. Especially in arid and semi-arid regions, water scarcity and inefficient irrigation methods severely impact farmland water management, leading to water waste and soil degradation. Therefore, improving the efficiency of agricultural irrigation water use and reducing unnecessary water consumption are urgent needs for promoting green agricultural development.

[0003] Intelligent agricultural irrigation management systems are gradually becoming a research hotspot. The integration of remote sensing and Internet of Things (IoT) technologies provides efficient and precise solutions. Remote sensing technology acquires large-scale farmland moisture and key indicator information through satellites or drones, helping farmers to monitor soil moisture conditions and climate change in real time. IoT technology, through sensors, data acquisition devices, and communication networks, transmits this data to the control center in real time, supporting irrigation decision-making.

[0004] Patent application CN203455682U discloses a remote intelligent control system for irrigation canal gates, including a communication module, a power supply module, a remote control center, and a main controller. The main controller is connected to an infrared microwave alarm device, a video monitoring device, a flow measurement device, a gate drive module, and a power control module. This technical solution can achieve automatic gate control, anti-theft alarm, anti-theft monitoring video, and overlay display of gate operating status data. It features high power utilization, strong system scalability and versatility, reducing water resource waste and lowering management and maintenance costs. However, it still fails to solve the problem of accuracy in real-time dynamic adjustment to soil moisture changes under variable weather conditions, and lacks an effective feedback mechanism to handle irrigation needs in areas with abnormal concentrations of water. Therefore, to overcome these limitations, this invention proposes an intelligent gate remote control method and system based on remote sensing and Internet of Things (IoT) technologies. Summary of the Invention

[0005] In view of the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for remote control of intelligent gates based on remote sensing and Internet of Things technologies, which solves the problems of real-time monitoring and anomaly detection of key indicators in target areas, optimizes gate execution commands, and improves the efficiency of water resource utilization.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The intelligent gate remote control system based on remote sensing and Internet of Things technologies includes a data acquisition module, a data processing module, and an execution feedback module.

[0008] The data acquisition module uses an intelligent sampling strategy to capture spectral data of the target area through a drone equipped with a remote sensing sensor, and captures environmental data of the target area through environmental sensors deployed at key nodes in the target area. The spectral data and environmental data are then transmitted to a central data platform for aggregation via IoT devices to construct a multidimensional dataset.

[0009] The data processing module is used to preprocess the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the key indicator anomalies, identify abnormal cluster areas, and fit the operation parameters of the gate of abnormal cluster areas by establishing a key indicator change model.

[0010] The execution feedback module generates execution instructions for the gates based on the operating parameters of the gates in the abnormal cluster area, monitors the execution status of the gates through status sensors installed on the gates, issues abnormal warnings for gates with execution anomalies, and transmits the execution results to the data processing module through IoT devices to update the execution instructions.

[0011] Specifically, the data acquisition module includes a data acquisition unit and a data transmission unit;

[0012] The intelligent sampling strategy is configured within the data acquisition unit. The intelligent sampling strategy adjusts the sampling frequency and range according to the crop growth cycle and meteorological conditions of the target area. Through UAV flight planning and automatic sensor control, it collects spectral data and environmental data, and records the timestamps and location coordinates of the spectral data and environmental data collection.

[0013] The data transmission unit is equipped with an optimized transmission strategy. Based on network conditions, the optimized transmission strategy selects the wireless communication method through IoT devices to transmit spectral data and environmental data to the central data platform.

[0014] Specifically, the steps of the intelligent sampling strategy include:

[0015] Set the initial sampling frequency according to the crop's growth stage;

[0016] The system receives meteorological data and dynamically adjusts the sampling frequency based on the changes in meteorological data and extreme weather conditions. The changes in meteorological data include changes in temperature, precipitation, humidity, and wind speed. The changes in extreme weather conditions are obtained based on the extreme nature of the meteorological data, which is determined by the deviation between the received meteorological data and meteorological standard data.

[0017] The environmental sensor's acquisition time is adjusted according to the dynamically adjusted acquisition frequency to obtain environmental data, and the UAV's flight frequency is adjusted to meet the requirements of the dynamically adjusted acquisition frequency. The flight path is also recorded to ensure consistency in the path for acquiring spectral data.

[0018] Specifically, the data processing module includes a preprocessing unit, an anomaly detection unit, and a predictive modeling unit;

[0019] The anomaly detection unit is equipped with a multi-level identification strategy. The multi-level identification strategy is used to identify anomalies in key indicators in the multidimensional dataset after data preprocessing. Through multi-level identification and judgment, abnormal data points are gradually filtered and marked to identify areas of abnormal clusters.

[0020] The predictive modeling unit is equipped with a multidimensional analysis strategy. This strategy obtains gate usage information, combines it with historical meteorological data to establish a key indicator change model, and fits the model to estimate the impact coefficient. This allows the system to obtain the gate's operating parameters based on the key indicator requirements of the abnormal clustering area.

[0021] Specifically, the steps of the multi-level recognition strategy include:

[0022] Configure a time threshold to extract multidimensional data points within the time threshold range from the current timestamp from the multidimensional dataset, forming an anomaly identification dataset;

[0023] Based on the crop type in the target area, key indicators affecting crop growth are selected. Crop types include aquatic crops and non-aquatic crops. For aquatic crops, field liquid level is selected as the key indicator affecting crop growth, and for non-aquatic crops, soil moisture is selected as the key indicator affecting crop growth.

[0024] Based on the growth stage of crops in the target area, indicator thresholds are configured, including upper and lower thresholds. Preliminary anomaly detection is performed on the anomaly identification dataset. If the key indicators of multidimensional data points in the anomaly identification dataset are greater than the upper threshold or less than the lower threshold, the data point is marked as a potential anomaly.

[0025] Time series analysis is used to fit the long-term trend of key indicators in a multidimensional dataset, and the trend deviation between the key indicators containing potential outliers and the fitted trend values ​​is calculated.

[0026] Configure a trend deviation threshold to assess the reasons for the anomalies in key indicators of potential outliers. If the trend deviation of the key indicators of a potential outlier is greater than the trend deviation threshold, it is marked as a sudden change outlier; otherwise, it is marked as a trend outlier.

[0027] Specifically, the steps of the multi-level recognition strategy also include:

[0028] Configure a neighborhood threshold to divide the neighborhood region range for neighborhood analysis of mutation anomalies and trend anomalies, and calculate the pattern similarity between mutation anomalies and trend anomalies and the multidimensional data points in their neighborhood regions respectively.

[0029] Configure a similarity threshold, count the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold, and obtain the similarity frequency of mutation anomalies and trend anomalies in the neighborhood area by calculating the ratio of the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold to the total number of multidimensional data points in the neighborhood area.

[0030] Configure a frequency threshold. If the similar frequency in the neighborhood of a mutation anomaly is greater than the frequency domain threshold, then mark the neighborhood as a mutation anomaly cluster area; otherwise, issue a mutation warning for the mutation anomaly. If the similar frequency in the neighborhood of a trend anomaly is greater than the frequency domain threshold, then mark the neighborhood as a trend anomaly cluster area; otherwise, issue a trend warning for the trend anomaly.

[0031] Output all multidimensional data points within the abnormal clustering region, which includes abrupt change abnormal clustering region and trend abnormal clustering region.

[0032] Specifically, the steps of a multidimensional analysis strategy include:

[0033] Obtain usage information of gates within the target area, including location coordinates, opening timestamp, closing timestamp, opening duration, water flow velocity, and cumulative water flow.

[0034] Configure the impact area threshold and impact time threshold, and filter out multidimensional data points from the multidimensional dataset that are within the gate impact area threshold and within the impact time threshold from the gate closing timestamp to construct the impact dataset;

[0035] Based on the impact dataset and historical meteorological data, a key indicator change model is constructed. The key indicator change model is jointly determined by the key indicators within the threshold of the gate's impact area, the gate's opening duration, the gate's cumulative flow, the distance of multi-dimensional data points in the impact dataset from the gate, and meteorological factors.

[0036] The meteorological factors mentioned are the comprehensive impact values ​​of meteorological data within the time range, which are jointly determined by the average temperature, average humidity, average wind speed, and cumulative precipitation within the time range.

[0037] Based on the fitted key indicator change model of the impact dataset, estimate each impact coefficient, and based on the multidimensional data points of the abnormal cluster area and its required key indicators, including the abnormal cluster area of ​​sudden change and the abnormal cluster area of ​​trend, solve the gate opening time and cumulative flow in the abnormal cluster area.

[0038] Specifically, the execution feedback module includes an instruction generation unit, a status detection unit, and a data update unit;

[0039] The instruction generation unit is configured with an instruction optimization strategy. The instruction optimization strategy generates gate execution instructions based on the gate's operating parameters and transmits instructions to open, close, or adjust the flow rate through IoT devices.

[0040] The status detection unit is equipped with a real-time feedback strategy. The real-time feedback strategy collects key parameters of the gate through status sensors installed on the gate, detects execution deviations and judges anomalies, and automatically triggers anomaly warnings.

[0041] The data update unit is configured with a feedback adjustment strategy. This strategy is used to collect spectral and environmental data after the execution of the command and transmit them to the data processing module via IoT devices to update the water resource demand of the target area in order to obtain new gate execution commands.

[0042] Specifically, the steps of the instruction optimization strategy include:

[0043] Receive the opening duration and cumulative water flow of the gates within the abnormal cluster area, and obtain the location coordinates of the gates. Based on the location coordinates of the gates, obtain the crop growth stage in the multi-dimensional data points within the threshold of the gate's influence area.

[0044] The water flow velocity threshold is set according to the crop growth stage, and the average flow rate is calculated based on the gate opening time and cumulative flow.

[0045] Calculate the water flow velocity based on the default gate opening and the average flow rate;

[0046] If the water flow velocity at the default gate opening is greater than the water flow velocity threshold, the gate opening is adjusted by the ratio of the gate's average flow rate to the water flow velocity threshold, and the adjusted gate opening and water flow velocity threshold are output; otherwise, the default gate opening and water flow velocity are output.

[0047] The system generates execution commands based on the gate opening degree and water flow velocity, including the gate opening duration, cumulative flow volume, opening degree, and water flow velocity.

[0048] Specifically, the remote control method for smart gates based on remote sensing and Internet of Things technologies includes the following steps:

[0049] Step S1: Using an intelligent sampling strategy, the spectral data of the target area is captured by a drone equipped with a remote sensing sensor, and the environmental data of the target area is captured by environmental sensors deployed at key nodes in the target area. The spectral data and environmental data are transmitted to a central data platform through IoT devices for aggregation and construction of a multidimensional dataset.

[0050] Step S2: After preprocessing the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the outliers of the key indicators, identify the abnormal clustering areas, and fit the operation parameters of the gate of the abnormal clustering areas by establishing a key indicator change model.

[0051] Step S3: Generate gate execution instructions based on the operating parameters of the gates in the abnormal cluster area, monitor the gate execution status through status sensors installed on the gates, issue abnormal warnings for gates with execution abnormalities, and transmit the execution results to the data processing module through IoT devices to update the execution instructions.

[0052] The beneficial effects of this invention are:

[0053] 1. Spectral data of the target area is captured using remote sensing sensors mounted on drones, combined with environmental parameters collected by ground-based environmental sensors. This data is transmitted to a central data platform via IoT technology for aggregation, providing a comprehensive foundation for subsequent data analysis and decision-making. Simultaneously, sampling frequency and coverage are dynamically adjusted based on crop growth cycles and real-time weather conditions to ensure data timeliness and representativeness. During severe weather or critical crop growth periods, the system increases sampling density to more precisely capture soil and environmental changes, thereby improving the accuracy and timeliness of early warnings.

[0054] 2. A multi-level identification strategy was adopted to progressively screen out anomalies in key indicators. Through time series analysis and neighborhood similarity calculation, abrupt changes and trend anomalies were effectively distinguished, providing a scientific basis for formulating targeted water-saving measures. Based on the needs of areas with clustered anomalies, and combining historical meteorological data and gate usage records, optimal gate operation parameters, including opening duration and cumulative flow, were calculated. This not only improved irrigation efficiency but also reduced water waste.

[0055] 3. Based on the processed data analysis results, execution commands for the gate are generated, and the execution status is monitored in real time by status sensors installed on the gate. A closed-loop control mechanism ensures the effective execution of commands and can promptly detect and address execution deviations or equipment malfunctions. Execution data is fed back to the data processing module for updating key indicator models and adjusting future operational strategies. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the intelligent gate remote control system based on remote sensing and Internet of Things technology of the present invention;

[0057] Figure 2 This is a flowchart illustrating the specific steps of the intelligent sampling strategy of the present invention;

[0058] Figure 3This is a flowchart illustrating the specific steps of the multi-level identification strategy of the present invention;

[0059] Figure 4 This is an example diagram illustrating the application of the multi-level recognition strategy of the present invention;

[0060] Figure 5 This is a flowchart illustrating the specific steps of the multidimensional analysis strategy of the present invention;

[0061] Figure 6 This is a flowchart of the intelligent gate remote control method based on remote sensing and Internet of Things technology according to the present invention. Detailed Implementation

[0062] Example 1

[0063] Please see Figure 1 This embodiment introduces a remote control system for intelligent gates based on remote sensing and Internet of Things technologies, including a data acquisition module, a data processing module, and an execution feedback module;

[0064] The data acquisition module utilizes an intelligent sampling strategy to capture spectral data of the target area using a drone equipped with a remote sensing sensor, and captures environmental data of the target area using environmental sensors deployed at key nodes within the target area. The spectral data and environmental data are then transmitted to a central data platform via IoT devices for aggregation, constructing a multidimensional dataset to monitor crop growth and soil quality in the target area.

[0065] The spectral data include: crop chlorophyll content, crop moisture content, and crop growth index.

[0066] Environmental data include: air temperature, soil temperature, air humidity, soil moisture, light intensity, carbon dioxide concentration, and field liquid level;

[0067] In this embodiment, a drone is used as a platform, equipped with remote sensing sensors to periodically fly over the target area and collect surface spectral data. The remote sensing sensors include a multispectral sensor and a thermal infrared sensor. The multispectral sensor captures spectral reflectance information of the target area in multiple specific bands. As the drone flies along its flight path, the multispectral sensor takes pictures of the ground at regular time intervals. Each shot records raw image data in different bands, reflecting the radiation intensity of the target area in each band. During the data acquisition process, the shooting time, location, and sensor attitude information are recorded to provide necessary auxiliary information for subsequent data processing. The thermal infrared sensor measures the surface temperature distribution of the target area. During the drone's flight, the thermal infrared sensor continuously operates, acquiring real-time ground thermal radiation information. Similarly, the shooting time, location, and attitude are recorded synchronously to ensure accurate spatiotemporal matching with the multispectral data. The collected multispectral image data undergoes radiometric calibration. By using pre-determined radiometric calibration parameters, including sensor response functions and atmospheric correction coefficients, the digital quantization values ​​in the raw image data are converted into actual surface reflectance data to accurately reflect the spectral characteristics of the target area, providing a reliable basis for subsequent analysis.

[0068] Based on the converted reflectance data and the absorption characteristics of chlorophyll in specific wavelength bands (including red and near-infrared bands), inversion models are used to estimate the chlorophyll content of crops. For example, by utilizing the absorption peak intensity of spectral reflectance near 670 nm and based on field measurements, an empirical relationship between reflectance and chlorophyll content is established, and an empirical formula is developed to invert the chlorophyll content in crop leaves. Water-sensitive wavelengths, such as the short-wave infrared band, are used to calculate moisture indices, such as the normalized water index (NDI). This is achieved by calculating the ratio or difference of short-wave infrared reflectance to obtain a numerical value reflecting crop moisture content. The extraction of crop growth indices depends on the reflectance in the red and near-infrared bands. Using vegetation index models, such as the normalized vegetation index (NDI) and enhanced vegetation index (EDI), combined with the reflectance in the red and near-infrared bands from multispectral data, crop growth indices are calculated. These indices can intuitively reflect the growth status of crops, such as vegetation cover and growth vigor. The temperature data acquired by the thermal infrared sensor is preprocessed, including noise removal and bias correction. Then, the thermal infrared data is spatiotemporally synchronized with the radiation intensity of each band extracted from the original image data. Timestamps and location information ensure precise temporal and spatial correspondence between thermal infrared and multispectral data for the same area.

[0069] Environmental sensors are deployed at key nodes in the target area, including major irrigation points of the agricultural irrigation system, soil moisture monitoring points, and weather stations. These sensors include temperature and humidity sensors, soil moisture sensors, barometric pressure sensors, light sensors, and field liquid level sensors, used to collect environmental data in the target area in real time. To ensure data timeliness, the environmental sensors collect data at a set sampling frequency, monitoring key factors such as soil moisture, weather changes, and crop requirements in real time. All collected spectral and environmental data are wirelessly transmitted via IoT devices and uploaded to a central data platform for storage and synchronization, ensuring data integrity and real-time performance.

[0070] Preferably, the data acquisition module includes a data acquisition unit and a data transmission unit;

[0071] The data acquisition unit is equipped with an intelligent sampling strategy. The intelligent sampling strategy adjusts the sampling frequency and range according to the crop growth cycle and meteorological conditions of the target area. Through UAV flight planning and automatic sensor control, it collects spectral data and environmental data in a timely, fixed-point, and quantitative manner, and records the timestamps and location coordinates of the spectral data and environmental data collection to ensure that the data coverage is broad and representative.

[0072] The data transmission unit is equipped with an optimized transmission strategy. Based on network conditions, the optimized transmission strategy selects the wireless communication method through IoT devices to transmit spectral data and environmental data to the central data platform.

[0073] Please see Figure 2 Preferably, the specific steps of the intelligent sampling strategy include:

[0074] The initial sampling frequency is set according to the growth stage of the crop, including the seedling stage, vegetative growth stage, flowering stage, and fruit enlargement stage, so that different sampling frequencies are configured for different crop growth stages. During the vigorous growth period, including the flowering stage and fruit enlargement stage, a higher sampling frequency is used.

[0075] The system receives meteorological data, including temperature, precipitation, wind speed, and humidity. Based on changes in meteorological data and extreme weather conditions, the sampling frequency is dynamically adjusted. When meteorological data changes significantly or exhibits extreme variations, the sampling frequency is increased. The dynamic adjustment formula for the sampling frequency is as follows:

[0076] ;

[0077] in, It is the dynamically adjusted sampling frequency. The initial sampling frequency is set according to the crop's growth stage. It is the change in temperature. It is the change in precipitation. It is the change in humidity. It is the change in wind speed. , , and These are weighting coefficients for changes in temperature, precipitation, humidity, and wind speed, used to represent the degree of influence of each meteorological factor on the sampling frequency. The value range is [0.05, 0.2]. The value range is [0.01, 0.1]. The value range is [0.02, 0.1]. The value range is [0.01, 0.05]. It is a weighted coefficient for the impact of extreme weather on sampling frequency. The value range is [0.1, 1]. It is the floor function. It represents the variation in extreme weather conditions, derived from the degree of extremeness of meteorological data. The degree of extremeness of meteorological data is obtained by the deviation between the received meteorological data and the meteorological standard data, i.e.:

[0078] ;

[0079] in, , , and These are the extreme temperatures. Extreme precipitation extreme humidity levels and extreme wind speed The weighting coefficients, The value range is [0.05, 0.2]. The value range is [0.01, 0.1]. The value range is [0.02, 0.1]. The values ​​range from [0.01, 0.05]. The extreme values ​​for temperature, precipitation, humidity, and wind speed are derived from the deviations between the received meteorological data and the meteorological standard data. It is a function for finding the maximum value;

[0080] The environmental sensor's acquisition time is adjusted according to the dynamically adjusted acquisition frequency to obtain environmental data, and the UAV's flight frequency is adjusted to meet the requirements of the dynamically adjusted acquisition frequency. The flight path is also recorded to ensure consistency in the path for acquiring spectral data.

[0081] Preferably, the specific steps for optimizing the transmission strategy include:

[0082] Regularly monitor network conditions within the target area, including signal strength, network latency, bandwidth, and communication stability, to detect network anomalies, such as signal loss and network disconnection, and assess the current network quality.

[0083] Based on network quality and the drone's flight area within the target region, select the wireless communication method for environmental sensors and remote sensing sensors. For example: LoRa is suitable for long-distance, low-data-rate communication; Wi-Fi is suitable for data transmission within a smaller range, providing higher bandwidth; cellular networks are suitable for scenarios with a large range and requiring high data transmission speeds and low latency.

[0084] Based on real-time network quality, the system automatically selects the communication method. For example, when the network signal is strong and bandwidth is sufficient, it prioritizes Wi-Fi or cellular networks; when the network is unstable or the distance is far, it switches to LoRa or satellite communication.

[0085] The collected spectral and environmental data are compressed to reduce the amount of data and improve transmission efficiency; and the data transmission rate is dynamically adjusted according to real-time network bandwidth and latency to ensure that transmission is not interrupted when the network is congested and to maintain the stability of data transmission.

[0086] The data processing module is used to preprocess the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the outliers of the key indicators, identify abnormal cluster areas, and fit the operating parameters of the gate of the abnormal cluster areas by establishing a key indicator change model.

[0087] In this embodiment, the collected spectral and environmental data are first preprocessed, including noise reduction, filtering, and standardization. Subsequently, multivariate analysis is used to further analyze the preprocessed data, combined with machine learning algorithms to identify potential anomalies, such as excessively low field slurry levels or abnormal crop growth. The machine learning algorithms used include cluster analysis and anomaly detection algorithms. If the data processing module detects an anomaly, an alarm mechanism is triggered to promptly notify relevant personnel for handling. To improve the accuracy of the early warning, a key indicator change model is established using historical meteorological data and key indicator data for detected anomalies. Furthermore, through comprehensive analysis of meteorological conditions and soil moisture, the opening duration and cumulative water flow of the gates within the anomaly cluster area are obtained.

[0088] Preferably, the data processing module includes a preprocessing unit, an anomaly detection unit, and a predictive modeling unit;

[0089] The preprocessing unit is equipped with a data cleaning strategy, which is used to preprocess the collected multidimensional dataset, including data denoising, data imputation and data standardization, to remove data errors in the multidimensional dataset, improve data quality and ensure the accuracy of subsequent analysis models.

[0090] The anomaly detection unit is equipped with a multi-level identification strategy. The multi-level identification strategy is used to identify anomalies in key indicators in the multidimensional dataset after data preprocessing. Through multi-level identification and judgment, abnormal data points are gradually filtered and marked to identify areas of abnormal clusters.

[0091] The predictive modeling unit is equipped with a multidimensional analysis strategy. This strategy obtains gate usage information, combines it with historical meteorological data to establish a key indicator change model, and fits the model to estimate the impact coefficient. This allows the system to obtain the gate's operating parameters based on the key indicator requirements of the abnormal clustering area.

[0092] Preferably, the specific steps of the data cleaning strategy include:

[0093] Based on the timestamps and coordinates in the multidimensional dataset, the spectral and environmental data are fused. This involves spatially correlating the spectral and environmental data, using a buffer radius centered on the key node where the environmental sensor is located, and dividing the spectral data into regions. A spatial interpolation algorithm is then used to convert the original image data into regular grid data, thereby extracting the spectral data corresponding to each key node's region. For example, at the same timestamp, the target region is divided into multiple sub-regions based on the coordinates of the environmental data. For each sub-region, the spectral data within that sub-region is averaged to synthesize a representative spectral dataset for subsequent analysis, ensuring regional consistency and achieving coordinate alignment between the environmental and spectral data.

[0094] Data cleaning and denoising are performed on the integrated multidimensional dataset. Noise filtering techniques are used to remove random noise from the multidimensional dataset to ensure the smoothness and usability of the data. For spectral data, bandpass filters and smoothing filters are used to reduce the impact of environmental interference.

[0095] Statistical methods are used to examine each dimension of the multidimensional dataset, detect and label missing values. For environmental data in the multidimensional dataset, methods such as forward imputation, backward imputation, and linear interpolation are used for imputation. For spectral data, spatial interpolation methods, including IDW interpolation and Kriging interpolation, are used to estimate missing values ​​based on the spectral data of surrounding adjacent locations to ensure the spatial consistency of the spectral data.

[0096] Spectral and environmental data collected from different sensors are standardized to eliminate unit differences between the sensors and ensure that all data are on the same scale, which facilitates subsequent analysis and model training.

[0097] Please see Figure 3 , Figure 4 The preferred multi-level recognition strategy includes the following specific steps:

[0098] Configure a time threshold to determine the time period to be obtained from the cube. Extract multidimensional data points from the cube within the time threshold range from the current timestamp to form the anomaly detection dataset.

[0099] ;

[0100] in, It is an anomaly detection dataset. It is the first in the anomaly detection dataset. The timestamp of the first Multidimensional data points at each coordinate location, including crop chlorophyll content, crop water content, crop growth index, air temperature, soil temperature, air humidity, soil humidity, light intensity, carbon dioxide concentration, and field liquid level. It is the first The timestamp of each time location. It is the current timestamp. It is a time threshold; by setting a time threshold, a time period is defined, and multi-dimensional data points between the current timestamp and that time period are extracted to form an anomaly identification dataset for anomaly detection. This narrows the scope of the data, ensures that only data within the most recent time window is considered, and improves the timeliness and targeting of anomaly detection.

[0101] Based on the crop type in the target area, key indicators affecting crop growth are selected. Crop types include aquatic crops and non-aquatic crops. For aquatic crops, field liquid level is selected as the key indicator affecting crop growth, and for non-aquatic crops, soil moisture is selected as the key indicator affecting crop growth.

[0102] Based on the growth stage of crops in the target area, indicator thresholds are configured, including upper and lower thresholds. Preliminary anomaly detection is performed on the anomaly identification dataset. If the key indicators of multi-dimensional data points in the anomaly identification dataset are greater than the upper threshold or less than the lower threshold, the data point is marked as a potential anomaly to identify key indicator anomalies in the target area and provide clues for subsequent in-depth analysis.

[0103] Time series analysis is used to fit the long-term trend of key indicators in a multidimensional dataset to describe the long-term variation patterns of these key indicators, and the trend deviation between the key indicators containing potential outliers and the fitted trend values ​​is calculated.

[0104] ;

[0105] in, It is the first Trend deviation of key indicators for potential outliers It is the first Key indicators for potential outliers It is the first The fitted trend values ​​of key indicators for potential outliers are used to quantify the degree of anomaly of key indicators through trend deviation, which helps to further screen out the real outliers.

[0106] Configure a trend deviation threshold to assess the reasons for the anomalies in key indicators of potential outliers. If the trend deviation of the key indicators of a potential outlier is greater than the trend deviation threshold, it is marked as a sudden change outlier; otherwise, it is marked as a trend outlier. By using the trend deviation threshold, the abnormal nature of potential outliers can be clarified, providing guidance for subsequent analysis.

[0107] Configure a neighborhood threshold to divide the neighborhood area range for neighborhood analysis of mutation anomalies and trend anomalies. Calculate the pattern similarity between mutation anomalies and trend anomalies and the multidimensional data points in their neighborhood areas to determine whether there are spatially abnormal clustering patterns and to discover regional or clustered anomalies. The pattern similarity calculation methods include Euclidean distance, Manhattan distance, etc.

[0108] A similarity threshold is configured, and the number of multidimensional data points with pattern similarity greater than the threshold within a neighborhood area is counted. By calculating the ratio of the number of multidimensional data points with pattern similarity greater than the threshold to the total number of multidimensional data points in the neighborhood area, the similarity frequency of the neighborhood areas containing mutation anomalies and trend anomalies is obtained. A frequency threshold is configured. If the similarity frequency in the neighborhood area containing a mutation anomaly is greater than the frequency domain threshold, the neighborhood area is marked as a mutation anomaly cluster area; otherwise, a mutation warning is issued for the mutation anomaly. If the similarity frequency in the neighborhood area containing a trend anomaly is greater than the frequency domain threshold, the neighborhood area is marked as a trend anomaly cluster area; otherwise, a trend warning is issued for the trend anomaly. By using neighborhood analysis to identify the spatial clustering of abnormal patterns, more refined anomaly determination can be made for specific areas.

[0109] Output all multidimensional data points within the abnormal clustering region, which includes abrupt change abnormal clustering region and trend abnormal clustering region.

[0110] Please see Figure 5 The preferred multidimensional analysis strategy includes the following specific steps:

[0111] Obtain usage information of gates within the target area, including location coordinates, opening timestamp, closing timestamp, opening duration, water flow velocity, and cumulative water flow.

[0112] Configure the impact area threshold and impact time threshold. The impact area threshold is used to measure the impact area of ​​the gate's use, and the impact time threshold is used to measure the time that the gate's use has an effect on the impact area. Select multidimensional data points from the multidimensional dataset that are within the gate's impact area threshold and within the impact time threshold from the gate's closing timestamp to construct the impact dataset.

[0113] Based on the impact dataset and historical meteorological data, a key indicator change model is constructed. This model is jointly determined by key indicators within the threshold range of the gate's impact area, the gate's opening duration, the gate's cumulative flow, the distance of multi-dimensional data points in the impact dataset from the gate, and meteorological factors. Specifically:

[0114] ;

[0115] in, It is the first Multidimensional data points at the gate's closing timestamp After the time threshold of influence Key indicators It is the first Multidimensional data points at the gate opening timestamp Key indicators at the time It refers to the duration the gate remains open. It is the cumulative water flow of the sluice gate. It is the first The distance of each multidimensional data point from the gate. , , , These are the influence coefficients of operating time, cumulative water flow, location distance, and meteorological factors on key indicators. It affects the time range [ , The comprehensive impact value of internal meteorological data, namely:

[0116] ;

[0117] in, It affects the time range [ , The average temperature within the area, It affects the time range [ , The average humidity within the area, It affects the time range [ , The average wind speed within the area, It affects the time range [ , The cumulative precipitation within the area, , , , These are the influence coefficients of average temperature, average humidity, average wind speed, and cumulative precipitation on the comprehensive impact value, respectively.

[0118] Based on the fitted key indicator change model of the impact dataset, estimate each impact coefficient, and solve the gate opening time and cumulative flow in the abnormal cluster area based on the multidimensional data points of the abnormal cluster area and its required key indicators.

[0119] The execution feedback module generates execution instructions for the gates based on the operating parameters of the gates in the abnormal cluster area, monitors the execution status of the gates through status sensors installed on the gates, issues abnormal warnings for gates with execution abnormalities, and transmits the execution results to the data processing module through IoT devices to update the execution instructions;

[0120] In this embodiment, the execution feedback module generates gate execution commands based on the operating parameters of the gates within the abnormal cluster area. These commands are sent to the gates as digital signals via IoT devices, instructing them to open, close, or adjust the flow rate. Status sensors installed on the gates monitor and provide real-time feedback on key information such as the gate's opening / closing status and flow rate. These sensors include flow sensors, pressure sensors, and position sensors. This information is transmitted in real-time to the data processing module via IoT devices to ensure timely feedback on the execution of control signals. Continuous monitoring of the status sensor data determines whether there are any execution anomalies in the gates, such as failure to open / close as instructed or insufficient flow rate. If an anomaly is detected, the system automatically triggers an early warning and notifies relevant personnel for handling. When an execution deviation is detected, the control system dynamically adjusts the gate control strategy based on the feedback information to ensure the normal operation of the irrigation system. Furthermore, the execution feedback module synchronizes the execution status of the control strategy with new environmental data to the data processing module, serving as an important basis for adjusting water resource demand. Based on this feedback data and real-time monitoring information, the water resource demand in the target area is adjusted, thereby further optimizing the water-saving strategy.

[0121] Preferably, the execution feedback module includes an instruction generation unit, a status detection unit, and a data update unit;

[0122] The instruction generation unit is configured with an instruction optimization strategy. The instruction optimization strategy generates gate execution instructions based on the gate's operating parameters and transmits instructions to the gate for opening, closing, or adjusting flow through IoT devices.

[0123] The status detection unit is equipped with a real-time feedback strategy. The real-time feedback strategy collects key parameters of the gate, including opening and closing status, flow rate, and pressure, through status sensors installed on the gate. It detects execution deviations and judges anomalies to automatically trigger anomaly warnings.

[0124] The data update unit is configured with a feedback adjustment strategy. This strategy is used to collect spectral and environmental data after the execution of the command and transmit them to the data processing module via IoT devices to update the water resource demand of the target area in order to obtain new gate execution commands.

[0125] Preferably, the specific steps of the instruction optimization strategy include:

[0126] Receive the opening duration and cumulative water flow of the gates within the abnormal cluster area, and obtain the location coordinates of the gates. Based on the location coordinates of the gates, obtain the crop growth stage in the multi-dimensional data points within the threshold of the gate's influence area.

[0127] The water flow rate threshold is set according to the crop growth stage. Crops at different growth stages have different water flow rate requirements. For example, a lower water flow rate is required during the seedling stage, while a higher water flow rate is required during the vegetative growth stage.

[0128] Calculate the average flow rate based on the gate's opening duration and cumulative flow volume, i.e.: ,in, It is the average flow rate of the gate. It is the cumulative water flow of the sluice gate. It refers to the duration the gate remains open;

[0129] Based on the default gate opening and the average flow rate, the water flow velocity is calculated, i.e.: ,in, This is the water flow velocity at the default gate opening. This is the default opening degree of the gate;

[0130] If the water flow velocity at the default gate opening is greater than the water flow velocity threshold, then the gate opening will be adjusted, i.e.: ,in, It refers to the adjusted gate opening. If it is the water flow velocity threshold, output the adjusted gate opening and water flow velocity threshold; otherwise, output the default gate opening and water flow velocity.

[0131] The system generates execution commands based on the gate opening degree and water flow velocity, including the gate opening duration, cumulative flow volume, opening degree, and water flow velocity.

[0132] Preferably, the specific steps of the real-time feedback strategy include:

[0133] Key data during the execution of commands are collected in real time through flow sensors, pressure sensors, and switch status sensors on the gate, including flow data, switch status data, and pressure data.

[0134] Configure a flow velocity deviation threshold and detect the actual water flow velocity passing through the gate in real time. If the deviation from the water flow velocity of the executed command is greater than the flow velocity deviation threshold, a flow velocity warning will be issued.

[0135] Configure a flow rate deviation threshold, count the actual flow rate passing through the gate when the opening time is met, calculate the flow rate deviation between the actual flow rate and the cumulative flow rate of the executed command, and issue a flow rate warning if the flow rate deviation is greater than the flow rate deviation threshold.

[0136] Warning information is sent to managers or farm operators via SMS or email for manual intervention.

[0137] Preferably, the specific steps of the feedback adjustment strategy include:

[0138] Configure feedback threshold At the end of the execution instruction After a certain period of time, spectral and environmental data of the abnormal clustering areas were collected to detect crop growth and soil changes.

[0139] The collected data will be transmitted to the central data platform via IoT devices for cleaning, standardization, and anomaly handling. Based on changes in key indicators and meteorological conditions, the control commands for the gates will be regenerated to optimize irrigation results.

[0140] Example 2

[0141] Please see Figure 6 This embodiment introduces a remote control method for smart gates based on remote sensing and Internet of Things (IoT) technologies, including the following steps:

[0142] Step S1: Using an intelligent sampling strategy, the spectral data of the target area is captured by a drone equipped with a remote sensing sensor, and the environmental data of the target area is captured by environmental sensors deployed at key nodes in the target area. The spectral data and environmental data are transmitted to a central data platform through IoT devices for aggregation and construction of a multidimensional dataset.

[0143] Step S2: After preprocessing the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the outliers of the key indicators, identify the abnormal clustering areas, and fit the operation parameters of the gate of the abnormal clustering areas by establishing a key indicator change model.

[0144] Step S3: Generate gate execution instructions based on the operating parameters of the gates in the abnormal cluster area, monitor the gate execution status through status sensors installed on the gates, issue abnormal warnings for gates with execution abnormalities, and transmit the execution results to the data processing module through IoT devices to update the execution instructions.

[0145] Specifically, the steps for acquiring spectral and environmental data include:

[0146] Set the initial sampling frequency according to the crop's growth stage;

[0147] The system receives meteorological data and dynamically adjusts the sampling frequency based on the changes in meteorological data and extreme weather conditions. The changes in meteorological data include changes in temperature, precipitation, humidity, and wind speed. The changes in extreme weather conditions are obtained based on the extreme nature of the meteorological data, which is determined by the deviation between the received meteorological data and meteorological standard data.

[0148] The environmental sensor's acquisition time is adjusted according to the dynamically adjusted acquisition frequency to obtain environmental data, and the UAV's flight frequency is adjusted to meet the requirements of the dynamically adjusted acquisition frequency. The flight path is also recorded to ensure consistency in the path for acquiring spectral data.

[0149] Specifically, the steps for obtaining abnormal clustering areas include:

[0150] Configure a time threshold to extract multidimensional data points within the time threshold range from the current timestamp from the multidimensional dataset, forming an anomaly identification dataset;

[0151] Based on the crop type in the target area, key indicators affecting crop growth are selected. Crop types include aquatic crops and non-aquatic crops. For aquatic crops, field liquid level is selected as the key indicator affecting crop growth, and for non-aquatic crops, soil moisture is selected as the key indicator affecting crop growth.

[0152] Based on the growth stage of crops in the target area, indicator thresholds are configured, including upper and lower thresholds. Preliminary anomaly detection is performed on the anomaly identification dataset. If the key indicators of multidimensional data points in the anomaly identification dataset are greater than the upper threshold or less than the lower threshold, the data point is marked as a potential anomaly.

[0153] Time series analysis is used to fit the long-term trend of key indicators in a multidimensional dataset, and the trend deviation between the key indicators containing potential outliers and the fitted trend values ​​is calculated.

[0154] Configure a trend deviation threshold to assess the reasons for the anomalies in key indicators of potential outliers. If the trend deviation of the key indicators of a potential outlier is greater than the trend deviation threshold, it is marked as a sudden change outlier; otherwise, it is marked as a trend outlier.

[0155] Configure a neighborhood threshold to divide the neighborhood region range for neighborhood analysis of mutation anomalies and trend anomalies, and calculate the pattern similarity between mutation anomalies and trend anomalies and the multidimensional data points in their neighborhood regions respectively.

[0156] Configure a similarity threshold, count the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold, and obtain the similarity frequency of mutation anomalies and trend anomalies in the neighborhood area by calculating the ratio of the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold to the total number of multidimensional data points in the neighborhood area.

[0157] Configure a frequency threshold. If the similar frequency in the neighborhood of a mutation anomaly is greater than the frequency domain threshold, then mark the neighborhood as a mutation anomaly cluster area; otherwise, issue a mutation warning for the mutation anomaly. If the similar frequency in the neighborhood of a trend anomaly is greater than the frequency domain threshold, then mark the neighborhood as a trend anomaly cluster area; otherwise, issue a trend warning for the trend anomaly.

[0158] Output all multidimensional data points within the abnormal clustering region, which includes abrupt change abnormal clustering region and trend abnormal clustering region.

[0159] Specifically, the steps for constructing a key indicator change model include:

[0160] Obtain usage information of gates within the target area, including location coordinates, opening timestamp, closing timestamp, opening duration, water flow velocity, and cumulative water flow.

[0161] Configure the impact area threshold and impact time threshold, and filter out multidimensional data points from the multidimensional dataset that are within the gate impact area threshold and within the impact time threshold from the gate closing timestamp to construct the impact dataset;

[0162] Based on the impact dataset and historical meteorological data, a key indicator change model is constructed. The key indicator change model is jointly determined by the key indicators within the threshold of the gate's impact area, the gate's opening duration, the gate's cumulative flow, the distance of multi-dimensional data points in the impact dataset from the gate, and meteorological factors.

[0163] The meteorological factors mentioned are the comprehensive impact values ​​of meteorological data within the time range, which are jointly determined by the average temperature, average humidity, average wind speed, and cumulative precipitation within the time range.

[0164] Based on the fitted key indicator change model of the impact dataset, estimate each impact coefficient, and based on the multidimensional data points of the abnormal cluster area and its required key indicators, including the abnormal cluster area of ​​sudden change and the abnormal cluster area of ​​trend, solve the gate opening time and cumulative flow in the abnormal cluster area.

[0165] Working principle and its effects:

[0166] The intelligent gate remote control method and system based on remote sensing and Internet of Things technologies achieves efficient water resource management through three main modules, ensuring reasonable irrigation and thus achieving water conservation.

[0167] The data acquisition module utilizes remote sensing sensors mounted on drones to acquire spectral data and combines this with environmental sensors deployed at key nodes to collect meteorological and environmental data. Based on crop growth status and weather conditions, an intelligent sampling strategy dynamically adjusts the sampling frequency and range to optimize data collection. Data is rapidly transmitted to a central data platform via IoT devices, ensuring real-time performance and accuracy. The data processing module cleans and formats the collected multidimensional data for subsequent analysis. A multi-level identification strategy identifies anomalies in key indicators, marking potential anomalies using set humidity upper and lower thresholds and performing trend analysis to identify areas of anomaly clustering. Combining key indicator change models with historical meteorological data, and based on the needs of areas of anomaly clustering, the module generates gate operation parameters, such as opening duration and flow rate calculations. The execution feedback module formulates corresponding execution instructions based on the operation parameters generated by the data processing module and transmits them via IoT devices. The module monitors the gate's execution status in real time, promptly detecting deviations and triggering warnings. Execution results are fed back, updating the water resource requirements of the target area to continuously optimize execution instructions.

[0168] By accurately monitoring key indicators and environmental changes, it is possible to effectively determine when irrigation is needed, reducing water waste. The combination of drones and sensors automates data collection, and real-time data feedback facilitates rapid response to complex environmental changes. Intelligent irrigation tailored to specific growth stages and conditions optimizes water resource use, thereby improving crop growth and yield. Rational management of irrigation water volume avoids unnecessary water runoff, reduces the risk of soil and water salinization, and protects the ecological environment. By integrating advanced sensing technologies, big data analytics, and intelligent control algorithms, intelligent water resource management is achieved, promoting sustainable agricultural development and contributing to addressing the challenges posed by climate change.

[0169] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A remote control system for intelligent gates based on remote sensing and Internet of Things technologies, characterized in that: It includes a data acquisition module, a data processing module, and an execution feedback module; The data acquisition module utilizes an intelligent sampling strategy to capture spectral data of the target area through a drone equipped with a remote sensing sensor, capture environmental data of the target area through environmental sensors deployed at key nodes within the target area, and transmit the spectral data and environmental data to a central data platform for aggregation via IoT devices to construct a multidimensional dataset. The steps of the intelligent sampling strategy include: Set the initial sampling frequency according to the crop's growth stage; The system receives meteorological data and dynamically adjusts the sampling frequency based on the changes in meteorological data and extreme weather conditions. The changes in meteorological data include changes in temperature, precipitation, humidity, and wind speed. The changes in extreme weather conditions are obtained based on the extreme nature of the meteorological data, which is determined by the deviation between the received meteorological data and meteorological standard data. The environmental sensor's acquisition time is adjusted according to the dynamically adjusted acquisition frequency to obtain environmental data, and the UAV's flight frequency is adjusted to meet the requirements of the dynamically adjusted acquisition frequency. The flight path is also recorded to collect spectral data. The data processing module is used to preprocess the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the key indicator anomalies, identify abnormal cluster areas, and fit the operation parameters of the gate of abnormal cluster areas by establishing a key indicator change model. The abnormal clustering regions include mutation abnormal clustering regions and trend abnormal clustering regions; The data processing module includes an anomaly detection unit, which is configured with a multi-level identification strategy. The steps of the multi-level identification strategy include: Configure a time threshold to extract multidimensional data points within the time threshold range from the current timestamp from the multidimensional dataset, forming an anomaly identification dataset; Based on the crop type in the target area, key indicators affecting crop growth are selected. Crop types include aquatic crops and non-aquatic crops. For aquatic crops, field liquid level is selected as the key indicator affecting crop growth, and for non-aquatic crops, soil moisture is selected as the key indicator affecting crop growth. Based on the growth stage of crops in the target area, indicator thresholds are configured, including upper and lower thresholds. Preliminary anomaly detection is performed on the anomaly identification dataset. If the key indicators of multidimensional data points in the anomaly identification dataset are greater than the upper threshold or less than the lower threshold, the data point is marked as a potential anomaly. Time series analysis is used to fit the long-term trend of key indicators in a multidimensional dataset, and the trend deviation between the key indicators containing potential outliers and the fitted trend values ​​is calculated. Configure a trend deviation threshold to assess the reasons for the anomalies in key indicators of potential outliers. If the trend deviation of the key indicators of a potential outlier is greater than the trend deviation threshold, it is marked as a sudden change outlier; otherwise, it is marked as a trend outlier. Configure a neighborhood threshold to divide the neighborhood region range for neighborhood analysis of mutation anomalies and trend anomalies, and calculate the pattern similarity between mutation anomalies and trend anomalies and the multidimensional data points in their neighborhood regions respectively. Configure a similarity threshold, count the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold, and obtain the similarity frequency of mutation anomalies and trend anomalies in the neighborhood area by calculating the ratio of the number of multidimensional data points in the neighborhood area whose pattern similarity is greater than the similarity threshold to the total number of multidimensional data points in the neighborhood area. Configure a frequency threshold. If the similar frequency in the neighborhood of a mutation anomaly is greater than the frequency domain threshold, then mark the neighborhood as a mutation anomaly cluster area; otherwise, issue a mutation warning for the mutation anomaly. If the similar frequency in the neighborhood of a trend anomaly is greater than the frequency domain threshold, then mark the neighborhood as a trend anomaly cluster area; otherwise, issue a trend warning for the trend anomaly. Output all multidimensional data points within the abnormal clustering region, which includes abrupt change abnormal clustering region and trend abnormal clustering region; The execution feedback module generates execution instructions for the gates based on the operating parameters of the gates in the abnormal cluster area, monitors the execution status of the gates through status sensors installed on the gates, issues abnormal warnings for gates with execution abnormalities, and transmits the execution results to the data processing module through IoT devices to update the execution instructions.

2. The intelligent gate remote control system based on remote sensing and Internet of Things technology as described in claim 1, characterized in that, The data acquisition module includes a data acquisition unit and a data transmission unit; The intelligent sampling strategy is configured in the data acquisition unit. The intelligent sampling strategy adjusts the sampling frequency and range according to the crop growth cycle and meteorological conditions of the target area. Through UAV flight planning and automatic sensor control, spectral data and environmental data are collected, and the timestamps and location coordinates of the spectral data and environmental data collection are recorded. The data transmission unit is configured with an optimized transmission strategy. Based on network conditions, the optimized transmission strategy selects a wireless communication method through IoT devices to transmit spectral data and environmental data to the central data platform.

3. The intelligent gate remote control system based on remote sensing and Internet of Things technology as described in claim 1, characterized in that, The data processing module includes a preprocessing unit, an anomaly detection unit, and a predictive modeling unit. The anomaly detection unit is configured with a multi-level identification strategy, which is used to identify anomalies in key indicators in the multidimensional dataset after data preprocessing. The multi-level identification judgment gradually filters and marks abnormal data points to identify abnormal cluster areas. The predictive modeling unit is equipped with a multidimensional analysis strategy. This strategy obtains gate usage information, combines historical meteorological data to establish a key indicator change model, and fits the model to estimate the influence coefficient, so as to obtain the gate's operating parameters according to the key indicator requirements of the abnormal cluster area.

4. The intelligent gate remote control system based on remote sensing and Internet of Things technology as described in claim 3, characterized in that, The steps of the multidimensional analysis strategy include: Obtain usage information of gates within the target area, including location coordinates, opening timestamp, closing timestamp, opening duration, water flow velocity, and cumulative water flow. Configure the impact area threshold and impact time threshold, and filter out multidimensional data points from the multidimensional dataset that are within the gate impact area threshold and within the impact time threshold from the gate closing timestamp to construct the impact dataset; Based on the impact dataset and historical meteorological data, a key indicator change model is constructed. The key indicator change model is jointly determined by the key indicators within the threshold of the gate's impact area, the gate's opening duration, the gate's cumulative flow, the distance of multi-dimensional data points in the impact dataset from the gate, and meteorological factors. The meteorological factors mentioned are the comprehensive impact values ​​of meteorological data within the time range, which are jointly determined by the average temperature, average humidity, average wind speed, and cumulative precipitation within the time range. Based on the fitted key indicator change model of the impact dataset, estimate each impact coefficient, and based on the multidimensional data points of the abnormal cluster area and its required key indicators, including the abnormal cluster area of ​​sudden change and the abnormal cluster area of ​​trend, solve the gate opening time and cumulative flow in the abnormal cluster area.

5. The intelligent gate remote control system based on remote sensing and Internet of Things technology as described in claim 1, characterized in that, The execution feedback module includes an instruction generation unit, a status detection unit, and a data update unit; The instruction generation unit is configured with an instruction optimization strategy, which generates gate execution instructions based on the gate's operating parameters and transmits instructions for opening, closing, or adjusting flow through IoT devices. The status detection unit is equipped with a real-time feedback strategy. The real-time feedback strategy collects key parameters of the gate through status sensors installed on the gate, detects execution deviations and judges anomalies, and automatically triggers anomaly warnings. The data update unit is configured with a feedback adjustment strategy, which is used to collect spectral data and environmental data after the execution of the command, and transmit them to the data processing module through IoT devices to update the water resource demand of the target area in order to obtain new gate execution commands.

6. The intelligent gate remote control system based on remote sensing and Internet of Things technology as described in claim 5, characterized in that, The specific steps of the instruction optimization strategy include: Receive the opening duration and cumulative water flow of the gates within the abnormal cluster area, and obtain the location coordinates of the gates. Based on the location coordinates of the gates, obtain the crop growth stage in the multi-dimensional data points within the threshold of the gate's influence area. The water flow velocity threshold is set according to the crop growth stage, and the average flow rate is calculated based on the gate opening time and cumulative flow. Calculate the water flow velocity based on the default gate opening and the average flow rate; If the water flow velocity at the default gate opening is greater than the water flow velocity threshold, the gate opening is adjusted by the ratio of the gate's average flow rate to the water flow velocity threshold, and the adjusted gate opening and water flow velocity threshold are output; otherwise, the default gate opening and water flow velocity are output. The system generates execution commands based on the gate opening degree and water flow velocity, including the gate opening duration, cumulative flow volume, opening degree, and water flow velocity.

7. A remote control method for intelligent gates based on remote sensing and Internet of Things (IoT) technologies, implemented based on any one of claims 1-6, characterized in that... Includes the following steps: Step S1: Using an intelligent sampling strategy, the spectral data of the target area is captured by a drone equipped with a remote sensing sensor, and the environmental data of the target area is captured by environmental sensors deployed at key nodes in the target area. The spectral data and environmental data are transmitted to a central data platform through IoT devices for aggregation and construction of a multidimensional dataset. Step S2: After preprocessing the multidimensional dataset, select key indicators based on the crop type of the target area, gradually screen and mark the outliers of the key indicators, identify the abnormal clustering areas, and fit the operation parameters of the gate of the abnormal clustering areas by establishing a key indicator change model. Step S3: Generate gate execution instructions based on the operating parameters of the gates in the abnormal cluster area, monitor the gate execution status through status sensors installed on the gates, issue abnormal warnings for gates with execution abnormalities, and transmit the execution results to the data processing module through IoT devices to update the execution instructions.

Images