Geological intelligent data base station

Abstract

The invention discloses a geographical intelligent data base station, which comprises a data sensing layer, a communication transmission layer, a management and calculation layer and a storage layer, and relates to the technical field of data communication. According to the geographical intelligent data base station, the intelligent data base station is arranged, the function boundary of a traditional base station is broken through, data localization storage, preprocessing and multi-modal analysis are achieved at the hierarchy of the base station, real-time decision feedback and data collaboration between distributed nodes are supported, and through a built-in self-adaptive algorithm, the intelligent data base station is obtained. Different scene requirements can be dynamically matched to complete tasks such as predictive analysis, anomaly detection and intelligent optimization, closed-loop data processing ecology is formed, response efficiency and system robustness are remarkably improved, infrastructure-level support is provided for intelligent internet-of-things application in a complex environment, various environmental factors are associated to establish an evaluation model, and the evaluation efficiency is improved. Data analysis and resource allocation are efficiently realized, and an execution strategy is timely generated for regulation and control.

Description

Technical Field

[0001] The present invention relates to the field of data communication technology, and in particular to a localized intelligent data base station. Background Art

[0002] Most base stations in the current market are limited to the basic data transmission function of uploading and downloading data, and only serve as information transfer nodes to realize simple interaction between remote commands and terminal devices. Such products rely on cloud processing architecture and need to transmit raw data back to the central node for analysis and decision-making. As a result, they lack real-time performance and are heavily dependent on network stability. It is difficult to ensure service continuity in remote areas or scenarios with limited network conditions.

[0003] The reference patent name is: A data processing method based on a base station group (patent publication number: CN113271613A, patent publication date: 2021-08-17), including: determining a task transfer plan for the triggered task, the task transfer plan including: at least one processing task; determining the target base station assigned to each of the processing tasks; sending the processing task to the assigned target base station; receiving the processing results of the processing task fed back by the target base station; and obtaining the processing results of the triggering task based on the processing results of the processing task.

[0004] Based on the description in the above-mentioned documents, it is difficult to fully obtain various data in the field environment after existing data collection. In addition, base stations in the current market are mostly limited to the basic data transmission function of uploading and downloading data, and only serve as information transfer nodes to realize simple interaction between remote commands and terminal devices. After data collection, the collected data usually needs to be directly transmitted to the remote server for processing. There are problems of high data transmission delay and low processing efficiency, resulting in insufficient real-time performance and serious dependence on network stability. It is difficult to ensure service continuity in remote areas or scenarios with limited network conditions. For this reason, the present invention provides a geographical intelligent data base station. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a localized intelligent data base station, which solves the problem of insufficient real-time performance caused by the single function and strong cloud dependence of traditional local data base stations.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A geographically intelligent data base station, comprising:

[0007] The data perception layer collects crop environment data in real time through a multimodal sensor array;

[0008] The communication transmission layer evaluates the network quality within the area and selects different communication methods to implement data transmission operations;

[0009] The management and computing layer extracts historical data to associate nodes of various categories to establish an evaluation model. The evaluation model first combines various environmental factors to derive the evaluation status of different regions, and then determines the corresponding execution strategy based on different evaluation status. Then, the real-time data is extracted and introduced into the evaluation model. After implementing resource allocation, the required execution strategy is derived and the final execution strategy is transmitted.

[0010] The storage layer stores the collected data and uses data redundancy backup algorithms to back up the data.

[0011] Preferably, the multimodal sensor array includes a soil sensor subarray, a meteorological sensor subarray and a crop growth sensor subarray. The soil sensor subarray is used to monitor the moisture, temperature and pH parameters of the soil in real time. The meteorological sensor subarray is used to obtain meteorological information in the field environment. The crop growth sensor subarray is used to monitor the growth status of crops.

[0012] Preferably, the operation of evaluating the network quality within the area in the communication transport layer is:

[0013] The communication modes 5G communication, LoRaWAN low-power wide area network communication, and Sigfox long-distance communication are represented by A1, A2, and A3, and the network quality is evaluated based on signal coverage, transmission distance, and transmission volume:

[0014] Step 1: Determine whether the signal coverage of the current area includes A1, A2 and A3, so as to determine the communication mode that can be transmitted;

[0015] Step 2: Determine the transmission distance from the current area to the receiving end, extract the historical distance thresholds of different communication methods and compare them with the transmission distance, and determine the communication method that can be transmitted based on step 1;

[0016] Step 3: Determine the amount of data that needs to be transmitted in the current area, extract the historical transmission volume thresholds of different communication methods and compare them with the current data volume required to be transmitted, and determine the communication method that can be used for transmission based on step 2;

[0017] If all three communication modes are available for transmission after the operations of steps 1 to 3 are met, the priority order is A1, A3 and A2, and the A1 communication mode is used first.

[0018] Preferably, the operation of extracting historical data in the management and calculation layer to associate various category nodes to establish an evaluation model is:

[0019] The names of each sub-array in the multimodal sensor array are used as the main category nodes at the same level, and the corresponding data category content collected under the sub-array is used as the sub-category node;

[0020] Establish an association relationship between sub-category nodes, extract the parameter values ​​between the two sub-category nodes, and combine the adaptive changes between the two in historical data. When the previous sub-category node changes, the subsequent sub-category nodes also need to be adjusted. Then, the association wire between the two sub-category nodes is directed from the previous sub-category node to the subsequent sub-category node.

[0021] By analogy, the association relationship between each sub-category node under the main category node is established to form an evaluation model for analysis and processing.

[0022] Preferably, the evaluation model first combines various environmental factors to derive the evaluation status of different areas as follows:

[0023] Extract each main category node for separate analysis, extract parameter values ​​and compare them with current crop demand values ​​to determine data anomalies in the current area, and mark sub-category nodes with abnormal parameter values;

[0024] Operations for analyzing soil sensor subarray data: Set the soil weight of the current area to G, and the soil moisture data value based on historical data to B t , the data value of soil temperature change per unit time is C, the data value of soil pH change per unit time is P, and the moisture value of the current area is calculated;

[0025] Operations on meteorological sensor subarray data analysis: Analyze the collected meteorological cloud map data to predict the future weather conditions in the current area;

[0026] Operations for analyzing crop growth sensor subarray data: obtaining crop growth status based on crop cultivation time, crop leaf condition, and crop stem size;

[0027] The evaluation results of different areas are determined by comprehensively analyzing the analysis results of each main category node.

[0028] Preferably, the analysis operation of the future weather conditions is:

[0029] Extract the meteorological cloud map data directly above the current area, taking the center of the current area as the center, and extract the characteristics of each cloud cluster on the meteorological cloud map data;

[0030] The basic meteorological data set is matched with the characteristics of each cloud cluster to determine the characteristics of the rainfall cloud cluster. The movement of the rainfall cloud cluster characteristics is determined based on the wind direction and speed in the collected data.

[0031] Assuming that the distance between the rainfall cloud feature moving toward the center of the current area and the center of the current area is L, and the wind speed is V, the predicted time of future rainfall is J=L / V.

[0032] Preferably, the analysis operation of the crop growth condition is:

[0033] Extract the overhead image data of the crop, convert the image into grayscale, and divide the characteristic parts of the area according to the grayscale values;

[0034] Extract the crop features from the historical data and match them with the divided feature parts to determine the leaf surface features, and judge whether there is a gap problem based on the boundary features of the leaf surface features. Then, the center surrounded by various leaf surface features is the crop stem feature, and a line segment is set based on the circle diameter feature of the crop stem feature to intersect with the boundary of the circle diameter feature. A second line segment is set at a right angle to the midpoint of the line segment and intersects with the boundary of the circle diameter feature. The length of the second line segment is the crop stem size data;

[0035] The notch problem is combined with the stalk size data to determine the growth of the crop.

[0036] Preferably, the moisture value calculation formula of the current area is:

[0037] F=(B t / G)-(CP)×T×n;

[0038] And F is the moisture value calculated over time, B t is the soil moisture data value detected in the tth time period, T represents the unit cycle time, and n represents the number of unit cycle times from the start of detection to the current time.

[0039] Preferably, the management and computing layer determines the corresponding execution strategy operation based on different evaluation situations:

[0040] Set the moisture threshold interval to [D1, D2];

[0041] Result 1: When F < [D1, D2], the corresponding execution strategy requires watering the current area;

[0042] Result 2: When F∈[D1, D2], the corresponding execution strategy does not need to perform water irrigation on the current area;

[0043] Result 3: When F>[D1, D2], the corresponding execution strategy needs to drain the water in the current area;

[0044] Then, based on the abnormal parameter values, the various parameter problems in the current area are determined. When analyzing the crop growth sensor subarray data, if defects are found, insect repellent strategies and fertilization strategies are generated. If the soil pH change data is abnormal, an acid-base balance strategy is generated.

[0045] If the first result is met and there is a rain problem in the future, the watering strategy will be removed. Otherwise, the strategy of the first result will be maintained.

[0046] If result three is met and rain problems are predicted in the future, preventive strategies need to be implemented in advance.

[0047] Preferably, the operation of extracting real-time data in the management and computing layer, introducing it into the evaluation model, and implementing resource allocation to derive the required execution strategy is:

[0048] After real-time data is introduced into the evaluation model, the soil sensor subarray collects a large amount of data after multiple acquisitions, so resource utilization is achieved by establishing an independent container. The meteorological sensor subarray and the crop growth sensor subarray are analyzed by allocating fixed resources.

[0049] Then, the evaluation status of the region is determined and the corresponding execution strategy is matched. The execution strategy is formed into instructions and then compressed and transmitted to each execution unit.

[0050] The present invention provides a localized intelligent data base station. Compared with the existing technology, it has the following advantages:

[0051] 1. This geo-based intelligent data base station, through the data perception layer, communication transmission layer, management and computing layer, and storage layer, breaks through the functional boundaries of traditional base stations, realizes data local storage, preprocessing and multimodal analysis at the base station level, supports real-time decision feedback and data collaboration between distributed nodes, and through built-in adaptive algorithms, can dynamically match the needs of different scenarios to complete tasks such as predictive analysis, anomaly detection and intelligent optimization, forming a closed-loop data processing ecosystem, significantly improving response efficiency and system robustness, and providing infrastructure-level support for intelligent IoT applications in complex environments. It also associates various environmental factors to establish an evaluation model, efficiently realizes data analysis and resource allocation, and timely generates execution strategies for regulation.

[0052] 2. This geo-based intelligent data base station collects multi-source data through soil sensor subarrays, meteorological sensor subarrays, and crop growth sensor subarrays, improving data accuracy and reliability. It then selects the optimal communication method based on different regional environmental conditions through communication adaptive algorithms and data cache compression operations, ensuring stable and timely data transmission and reducing transmission time.

[0053] 3. The geographically-based intelligent data base station establishes an evaluation model by extracting historical data to associate nodes of various categories. The evaluation model first combines various environmental factors to derive the evaluation status of different regions, and then determines the corresponding execution strategy based on different evaluation status. According to the characteristics of different data processing tasks, computing resources are reasonably allocated, and the evaluation status of the current region is combined with the anomalies of other factors to generate a specific and simple strategy for rapid analysis and processing, thereby achieving adaptive analysis and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a principle block diagram of the intelligent data base station of the present invention;

[0055] Figure 2 This is an operational flow chart of the network quality assessment of the present invention;

[0056] Figure 3 The figure is a flowchart of the operation of establishing the evaluation model of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] See also Figure 1-Figure 3 , the present invention provides two technical solutions:

[0059] Embodiment 1: A geographically-based intelligent data base station, comprising:

[0060] The data perception layer collects crop environment data in real time through a multimodal sensor array;

[0061] The communication transmission layer evaluates the network quality within the area and selects different communication methods to implement data transmission operations;

[0062] The management and computing layer extracts historical data to associate nodes of various categories to establish an evaluation model. The evaluation model first combines various environmental factors to derive the evaluation status of different regions, and then determines the corresponding execution strategy based on different evaluation status. Then, the real-time data is extracted and introduced into the evaluation model. After implementing resource allocation, the required execution strategy is derived and the final execution strategy is transmitted.

[0063] The storage layer stores the collected data and uses data redundancy backup algorithms to back up the data.

[0064] Core components include a multimodal sensor array, a heterogeneous computing processing unit, a distributed storage matrix, and an adaptive communication module. The sensor array, comprised of specialized sensors for temperature, humidity, vibration, and electromagnetic environment, connects to the main processing unit via a MIPI interface to enable real-time acquisition of physical environmental parameters. The processing unit utilizes a heterogeneous FPGA+CPU architecture, with the FPGA responsible for data cleaning and feature extraction, and the CPU running a machine learning algorithm library. The two exchange data via a PCIe 3.0 bus. The storage system utilizes a hybrid architecture of NVMe SSDs and 3D XPoint memory, with data redundancy protected by RAID technology. Critical business data utilizes AES-256 hardware encryption. The communication module integrates 5G NR, LoRa, and Sigfox triple-mode chips, dynamically switching network topologies through software-defined radio technology to ensure communication reliability in complex environments.

[0065] Through the data perception layer, communication transmission layer, management and computing layer, and storage layer, it breaks through the functional boundaries of traditional base stations, realizes data local storage, preprocessing and multimodal analysis at the base station level, supports real-time decision feedback and data collaboration between distributed nodes, and through the built-in adaptive algorithm, it can dynamically match the needs of different scenarios to complete tasks such as predictive analysis, anomaly detection and intelligent optimization, forming a closed-loop data processing ecosystem, significantly improving response efficiency and system robustness, and providing infrastructure-level support for smart IoT applications in complex environments. It also associates various environmental factors to establish an evaluation model, efficiently realizes data analysis and resource allocation, and generates execution strategies for timely regulation.

[0066] In an embodiment of the present invention, the multimodal sensor array includes a soil sensor subarray, a meteorological sensor subarray, and a crop growth sensor subarray. The soil sensor subarray is used to monitor soil moisture, temperature, and pH parameters in real time. The meteorological sensor subarray is used to obtain meteorological information in a field environment. The crop growth sensor subarray is used to monitor the growth status of crops.

[0067] In the embodiment of the present invention, the operation of evaluating the network quality within the area in the communication transport layer is:

[0068] The communication modes 5G communication, LoRaWAN low-power wide area network communication, and Sigfox long-distance communication are represented by A1, A2, and A3, and the network quality is evaluated based on signal coverage, transmission distance, and transmission volume:

[0069] Step 1: Determine whether the signal coverage of the current area includes A1, A2 and A3, so as to determine the communication mode that can be transmitted;

[0070] Step 2: Determine the transmission distance from the current area to the receiving end, extract the historical distance thresholds of different communication methods and compare them with the transmission distance, and determine the communication method that can be transmitted based on step 1;

[0071] Step 3: Determine the amount of data that needs to be transmitted in the current area, extract the historical transmission volume thresholds of different communication methods and compare them with the current data volume required to be transmitted, and determine the communication method that can be used for transmission based on step 2;

[0072] If all three communication modes are available for transmission after the operations of steps 1 to 3 are met, the priority order is A1, A3 and A2, and the A1 communication mode is used first.

[0073] Among them, multi-source data collection operations are realized through soil sensor subarrays, meteorological sensor subarrays and crop growth sensor subarrays to improve data accuracy and reliability. Then, through communication adaptive algorithms and data cache compression operations, the optimal communication method is selected according to different regional environmental conditions to ensure stable and timely data transmission and shorten the transmission process.

[0074] In the embodiment of the present invention, the operation of extracting historical data from the management and computing layer to associate various category nodes and establish an evaluation model is as follows:

[0075] The names of each sub-array in the multimodal sensor array are used as the main category nodes at the same level, and the corresponding data category content collected under the sub-array is used as the sub-category node;

[0076] Establish an association relationship between sub-category nodes, extract the parameter values ​​between the two sub-category nodes, and combine the adaptive changes between the two in historical data. When the previous sub-category node changes, the subsequent sub-category nodes also need to be adjusted. Then, the association wire between the two sub-category nodes is directed from the previous sub-category node to the subsequent sub-category node.

[0077] By analogy, the association relationship between each sub-category node under the main category node is established to form an evaluation model for analysis and processing.

[0078] In the embodiment of the present invention, the evaluation model first combines various environmental factors to obtain the evaluation status of different areas. The operation is as follows:

[0079] Extract each main category node for separate analysis, extract parameter values ​​and compare them with current crop demand values ​​to determine data anomalies in the current area, and mark sub-category nodes with abnormal parameter values;

[0080] Operations for analyzing soil sensor subarray data: Set the soil weight of the current area to G, and the soil moisture data value based on historical data to B t, the data value of soil temperature change per unit time is C, the data value of soil pH change per unit time is P, and the moisture value of the current area is calculated;

[0081] Operations on meteorological sensor subarray data analysis: Analyze the collected meteorological cloud map data to predict the future weather conditions in the current area;

[0082] Operations for analyzing crop growth sensor subarray data: obtaining crop growth status based on crop cultivation time, crop leaf condition, and crop stem size;

[0083] The evaluation results of different areas are determined by comprehensively analyzing the analysis results of each main category node.

[0084] In the embodiment of the present invention, the analysis operation of future weather conditions is as follows:

[0085] Extract the meteorological cloud map data directly above the current area, taking the center of the current area as the center, and extract the characteristics of each cloud cluster on the meteorological cloud map data;

[0086] The basic meteorological data set is matched with the characteristics of each cloud cluster to determine the characteristics of the rainfall cloud cluster. The movement of the rainfall cloud cluster characteristics is determined based on the wind direction and speed in the collected data.

[0087] Assuming that the distance between the rainfall cloud feature moving toward the center of the current area and the center of the current area is L, and the wind speed is V, the predicted time of future rainfall is J=L / V.

[0088] In the embodiment of the present invention, the analysis operation of the crop growth condition is as follows:

[0089] Extract the overhead image data of the crop, convert the image into grayscale, and divide the characteristic parts of the area according to the grayscale values;

[0090] Extract the crop features from the historical data and match them with the divided feature parts to determine the leaf surface features, and judge whether there is a gap problem based on the boundary features of the leaf surface features. Then, the center surrounded by various leaf surface features is the crop stem feature, and a line segment is set based on the circle diameter feature of the crop stem feature to intersect with the boundary of the circle diameter feature. A second line segment is set at a right angle to the midpoint of the line segment and intersects with the boundary of the circle diameter feature. The length of the second line segment is the crop stem size data;

[0091] The notch problem is combined with the stalk size data to determine the growth of the crop.

[0092] In the embodiment of the present invention, the formula for calculating the moisture value of the current area is:

[0093] F=(B t / G)-(CP)×T×n;

[0094] And F is the moisture value calculated over time, B t is the soil moisture data value detected in the tth time period, T represents the unit cycle time, and n represents the number of unit cycle times from the start of detection to the current time.

[0095] In the embodiment of the present invention, the management and computing layers determine the corresponding execution strategy operations based on different evaluation conditions:

[0096] Set the moisture threshold interval to [D1, D2];

[0097] Result 1: When F < [D1, D2], the corresponding execution strategy requires watering the current area;

[0098] Result 2: When F∈[D1, D2], the corresponding execution strategy does not need to perform water irrigation on the current area;

[0099] Result 3: When F>[D1, D2], the corresponding execution strategy needs to drain the water in the current area;

[0100] Then, based on the abnormal parameter values, the various parameter problems in the current area are determined. When analyzing the crop growth sensor subarray data, if defects are found, insect repellent strategies and fertilization strategies are generated. If the soil pH change data is abnormal, an acid-base balance strategy is generated.

[0101] If the first result is met and there is a rain problem in the future, the watering strategy will be removed. Otherwise, the strategy of the first result will be maintained.

[0102] If result three is met and rain problems are predicted in the future, preventive strategies need to be implemented in advance.

[0103] Among them, the evaluation model is established by extracting historical data to realize the association of various category nodes. The evaluation model first combines various environmental factors to obtain the evaluation status of different areas, and determines the corresponding execution strategy based on different evaluation status. According to the characteristics of different data processing tasks, computing resources are reasonably allocated, and the evaluation status of the current area is combined with the anomalies of other factors to generate specific and simple strategies to achieve rapid analysis and processing, and realize adaptive analysis and processing.

[0104] In the embodiment of the present invention, the operation of extracting real-time data in the management and computing layer, introducing it into the evaluation model, and implementing resource allocation to derive the required execution strategy is as follows:

[0105] After real-time data is introduced into the evaluation model, the soil sensor subarray collects a large amount of data after multiple acquisitions, so resource utilization is achieved by establishing an independent container. The meteorological sensor subarray and the crop growth sensor subarray are analyzed by allocating fixed resources.

[0106] Then, the evaluation status of the region is determined and the corresponding execution strategy is matched. The execution strategy is formed into instructions and then compressed and transmitted to each execution unit.

[0107] The difference between Example 2 and Example 1 is that a composite heat dissipation system of forced air cooling and heat pipes is used inside the intelligent data base station equipment. The surface of key components is coated with a nano-carbon heat dissipation coating to stabilize the operating temperature within the industrial grade range of -40°C to 70°C. The power module is equipped with a dual-redundant UPS and a maximum power point tracking (MPPT) solar charging controller to ensure that the energy storage battery seamlessly takes over the power supply when the mains power is interrupted. The modules are plug-and-play through modular connectors. The backplane bus adopts a 12-layer PCB design, and the signal layer and power layer are independently wired. The key signal lines use differential pairs and are configured with EMI filters to effectively suppress electromagnetic interference. When data anomalies are detected, the FPGA triggers a hardware-level interrupt, and the CPU immediately starts the preset algorithm for pattern recognition. The results are fed back to the execution terminal within 10ms through the communication module, forming a closed-loop control. This design uses hardware acceleration and software optimization to increase the local inference speed by three orders of magnitude, providing low-latency and highly reliable edge intelligent services for scenarios such as industrial Internet of Things and smart agriculture.

[0108] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0109] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0110] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A geographically-based intelligent data base station, characterized by: include: The data perception layer collects crop environment data in real time through a multimodal sensor array; The communication transmission layer evaluates the network quality within the area and selects different communication methods to implement data transmission operations; The management and computing layer extracts historical data to associate nodes of various categories to establish an evaluation model. The evaluation model first combines various environmental factors to derive the evaluation status of different regions, and then determines the corresponding execution strategy based on different evaluation status. Then, the real-time data is extracted and introduced into the evaluation model. After implementing resource allocation, the required execution strategy is derived and the final execution strategy is transmitted. The storage layer stores the collected data and uses data redundancy backup algorithms to back up the data.

2. The localized intelligent data base station according to claim 1, characterized in that: The multimodal sensor array includes a soil sensor subarray, a meteorological sensor subarray, and a crop growth sensor subarray. The soil sensor subarray is used to monitor soil moisture, temperature, and pH parameters in real time. The meteorological sensor subarray is used to obtain meteorological information in the field environment. The crop growth sensor subarray is used to monitor the growth status of crops.

3. The geographically-based intelligent data base station according to claim 1, characterized in that: The operation of evaluating the network quality within the area in the communication transmission layer is: The communication modes 5G communication, LoRaWAN low-power wide area network communication, and Sigfox long-distance communication are represented by A1, A2, and A3, and the network quality is evaluated based on signal coverage, transmission distance, and transmission volume: Step 1: Determine whether the signal coverage of the current area includes A1, A2 and A3, so as to determine the communication mode that can be transmitted; Step 2: Determine the transmission distance from the current area to the receiving end, extract the historical distance thresholds of different communication methods and compare them with the transmission distance, and determine the communication method that can be transmitted based on step 1; Step 3: Determine the amount of data that needs to be transmitted in the current area, extract the historical transmission volume thresholds of different communication methods and compare them with the current data volume required to be transmitted, and determine the communication method that can be used for transmission based on step 2; If all three communication modes are available for transmission after the operations of steps 1 to 3 are met, the priority order is A1, A3 and A2, and the A1 communication mode is used first.

4. The localized intelligent data base station according to claim 2, characterized in that: The operation of extracting historical data in the management and computing layer to associate various category nodes and establish an evaluation model is as follows: The names of each sub-array in the multimodal sensor array are used as the main category nodes at the same level, and the corresponding data category content collected under the sub-array is used as the sub-category node; Establish an association relationship between sub-category nodes, extract the parameter values ​​between the two sub-category nodes, and combine the adaptive changes between the two in historical data. When the previous sub-category node changes, the subsequent sub-category nodes also need to be adjusted. Then, the association wire between the two sub-category nodes is directed from the previous sub-category node to the subsequent sub-category node. By analogy, the association relationship between each sub-category node under the main category node is established to form an evaluation model for analysis and processing.

5. The geographically-based intelligent data base station according to claim 4, characterized in that: The evaluation model first combines various environmental factors to derive the evaluation results of different regions as follows: Extract each main category node for separate analysis, extract parameter values ​​and compare them with current crop demand values ​​to determine data anomalies in the current area, and mark sub-category nodes with abnormal parameter values; Operations for analyzing soil sensor subarray data: Set the soil weight of the current area to G, and the soil moisture data value based on historical data to B t , the data value of soil temperature change per unit time is C, the data value of soil pH change per unit time is P, and the moisture value of the current area is calculated; Operations on meteorological sensor subarray data analysis: Analyze the collected meteorological cloud map data to predict the future weather conditions in the current area; Operations for analyzing crop growth sensor subarray data: obtaining crop growth status based on crop cultivation time, crop leaf condition, and crop stem size; The evaluation results of different areas are determined by comprehensively analyzing the analysis results of each main category node.

6. The geographically-based intelligent data base station according to claim 5, characterized in that: The analysis operation of the future weather conditions is as follows: Extract the meteorological cloud map data directly above the current area, taking the center of the current area as the center, and extract the characteristics of each cloud cluster on the meteorological cloud map data; The basic meteorological data set is matched with the characteristics of each cloud cluster to determine the characteristics of the rainfall cloud cluster. The movement of the rainfall cloud cluster characteristics is determined based on the wind direction and speed in the collected data. Assuming that the distance between the rainfall cloud feature moving toward the center of the current area and the center of the current area is L, and the wind speed is V, the predicted time of future rainfall is J=L / V.

7. The geographically-based intelligent data base station according to claim 5, characterized in that: The analysis operation of the crop growth condition is as follows: Extract the overhead image data of the crop, convert the image into grayscale, and divide the characteristic parts of the area according to the grayscale values; Extract the crop features from the historical data and match them with the divided feature parts to determine the leaf surface features, and judge whether there is a gap problem based on the boundary features of the leaf surface features. Then, the center surrounded by various leaf surface features is the crop stem feature, and a line segment is set based on the circle diameter feature of the crop stem feature to intersect with the boundary of the circle diameter feature. A second line segment is set at a right angle to the midpoint of the line segment and intersects with the boundary of the circle diameter feature. The length of the second line segment is the crop stem size data; The notch problem is combined with the stalk size data to determine the growth of the crop.

8. The geographically-based intelligent data base station according to claim 5, characterized in that: The calculation formula for the moisture value of the current area is: F=(B t / G)-(C-P)×T×n; And F is the moisture value calculated over time, B t is the soil moisture data value detected in the tth time period, T represents the unit cycle time, and n represents the number of unit cycle times from the start of detection to the current time.

9. The geographically-based intelligent data base station according to claim 8, characterized in that: The management and computing layer determines the corresponding execution strategy operations based on different evaluation conditions: Set the moisture threshold interval to [D1, D2]; Result 1: When F < [D1, D2], the corresponding execution strategy requires watering the current area; Result 2: When F∈[D1, D2], the corresponding execution strategy does not need to perform water irrigation on the current area; Result 3: When F>[D1, D2], the corresponding execution strategy needs to drain the water in the current area; Then, based on the abnormal parameter values, the various parameter problems in the current area are determined. When analyzing the crop growth sensor subarray data, if defects are found, insect repellent strategies and fertilization strategies are generated. If the soil pH change data is abnormal, an acid-base balance strategy is generated. If the first result is met and there is a rain problem in the future, the watering strategy will be removed. Otherwise, the strategy of the first result will be maintained. If result three is met and rain problems are predicted in the future, preventive strategies need to be implemented in advance.

10. The geographically-based intelligent data base station according to claim 1, characterized in that: The operations in the management and computing layer to extract real-time data, introduce it into the evaluation model, and implement resource allocation to derive the required execution strategy are as follows: After real-time data is introduced into the evaluation model, the soil sensor subarray collects a large amount of data after multiple acquisitions, so resource utilization is achieved by establishing an independent container. The meteorological sensor subarray and the crop growth sensor subarray are analyzed by allocating fixed resources. Then, the evaluation status of the region is determined and the corresponding execution strategy is matched. The execution strategy is formed into instructions and then compressed and transmitted to each execution unit.

Images