Soil nutrient detection and comparison system based on big data analysis and management

Abstract

The invention relates to the technical field of soil nutrient detection, and discloses a soil nutrient detection and comparison system based on big data analysis and management. The integrated calibration control module is introduced to be linked with the cloud big data platform, so that the measurement basis of the probe type detection equipment is conveniently and accurately calibrated; an intelligent data conversion unit arranged in the system converts original parameters into a complete data set containing multiple nutrient indexes, real-time comparison and analysis are carried out in combination with a cloud crop standard parameter library, the data is presented in a visual comparison report form, and the dilemma that traditional detection data is obscure, difficult to understand and difficult to directly apply is changed; finally, a complete closed loop from accurate measurement, intelligent analysis to scientific decision is constructed, a user can quickly obtain a customized fertilization guidance scheme without professional knowledge, the technical threshold and decision cost of accurate agriculture are reduced, and the practical value of soil detection data and the intelligent level of agricultural management are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of soil nutrient detection, in particular to a soil nutrient detection comparison system based on big data analysis management. BACKGROUND

[0002] Soil nutrient detection is a key link for guiding scientific fertilization and improving agricultural production efficiency. At present, there are various soil nutrient detection devices and methods in the market, but there are still obvious limitations in real-time, accuracy and intelligence.

[0003] The existing probe type (or probe type) soil rapid tester can realize rapid in-situ detection, but its technical maturity is insufficient, and has the following defects: limited by production process, component consistency and cost, the existing device has large measurement error. More importantly, it generally lacks reliable calibration mechanism, or the calibration process is cumbersome, and cannot effectively correct the measurement drift caused by long-term use and environmental changes of the device, so that the reliability of the measurement data decreases over time. The device can usually only output basic physicochemical parameters such as electrical conductivity (EC value), pH value, or nutrient estimation values converted based on a simple fixed model. These original data or single estimation results are too professional for ordinary farmers and are difficult to understand intuitively, and cannot be directly converted into specific fertilization guidance, so the data application value is low. The data is isolated and lacks intelligent analysis: the existing device is usually operated by a single machine, and the detection data is isolated, which cannot be associated and compared with historical data, regional soil information and nutrient demand standards of different crops. The lack of intelligent comparison and decision support function based on big data makes the detection behavior and the final fertilization management decision disjointed.

[0004] Soil is the basis of agricultural production, and its nutrient status is directly related to the yield and quality of crops. Rapid and accurate detection of soil nutrients and accurate fertilization based on the detection are important means to realize agricultural modernization. At present, the detection methods of soil nutrients mainly include laboratory detection and on-site rapid detection.

[0005] Although the laboratory detection method has high precision, it has the disadvantages of long detection period, high cost and complex process, which is difficult to meet the real-time and rapid monitoring demand of soil nutrients in agricultural production. Therefore, various portable and probe type soil nutrient rapid testers have appeared in the market, which can realize in-situ rapid detection of soil. However, these on-site rapid detection devices still have many obvious defects: Firstly, the measurement accuracy of the existing probe type soil rapid tester is generally not high, and it lacks effective calibration function. Due to the limitation of production process, component consistency and cost, there is a large error between different devices or the same device at different periods.

[0006] Therefore, the field urgently needs a soil nutrient detection and analysis system integrating high-precision calibratable detection, intelligent data model conversion, cloud big data comparison and personalized decision support to overcome the above-mentioned deficiencies of the prior art. SUMMARY

[0007] In view of the deficiencies of the prior art, the present application provides a soil nutrient detection and comparison system based on big data analysis and management to solve the problems in the background art.

[0008] To achieve the above-mentioned purpose, the present application provides the following technical solution: a soil nutrient detection and comparison system based on big data analysis and management, comprising: a probe type detection device for inserting into the soil to detect the original physicochemical parameters of the soil; a cloud big data platform, which stores a standard soil nutrient reference database divided according to different planting varieties and soil types; a data processing and communication terminal, which is in communication connection with the probe type detection device and performs data interaction with the cloud big data platform through a wireless network; the data processing and communication terminal is integrated with: a) a data conversion unit, which is pre-provisioned with an algorithm model for converting the original physicochemical parameters into a plurality of soil nutrient indicators, the plurality of soil nutrient indicators at least including water content, electrical conductivity, soil temperature, available nitrogen, available phosphorus and available potassium content; b) a calibration control module, configured to, when a user inserts the probe type detection device into a standard test liquid for calibration instruction, obtain the standard parameter value of the standard test liquid from the cloud big data platform after the instruction is triggered, receive the real-time measurement value of the probe type detection device and convert it into a plurality of soil nutrient indicators through the data conversion unit, compare the plurality of soil nutrient indicators with the obtained standard parameter value, calculate the calibration parameter for the data conversion unit based on the comparison difference, and use the calibration parameter to mathematically correct the subsequent conversion results of the data conversion unit to make the output value approach the standard value; c) a data comparison and feedback module, configured to receive the planting variety information input by the user, call the corresponding standard reference data from the cloud big data platform, and compare the converted plurality of soil nutrient indicators with the standard reference data side by side to generate a comparison report containing difference analysis; d) an output module for displaying the comparison report, the plurality of soil nutrient indicators and / or the original soil physicochemical parameters on the screen of the data processing and communication terminal, and / or printing them through a printing device connected thereto.

[0009] Preferably, the calibration control module is calibrated by inserting the probe into a standard test solution, and the calibration accuracy meets at least one of the following: Soil temperature: control adjustment accuracy ±0.1℃; Soil moisture content: control adjustment accuracy ±0.1%; Soil conductivity: control adjustment accuracy ±50us / cm; Soil pH: control adjustment accuracy ±0.1; Available nitrogen content in soil: control adjustment accuracy ±1.0mg / kg; Available phosphorus content in soil: control adjustment accuracy ±1.0mg / kg; Available potassium content in soil: control adjustment accuracy ±1.0mg / kg; Water-soluble organic matter content in soil: control adjustment accuracy ±1.0g / kg.

[0010] Preferably, the calibration control module is calibrated by the following method: Calibration of soil temperature, moisture content, conductivity and pH is completed by comparing the soil nutrient indicators calculated by the data conversion unit according to the real-time measurement values with the standard parameter values obtained from the cloud big data platform, in response to the user inserting the probe 70%-80% into water and starting the calibration operation. Calibration of available nitrogen, available phosphorus, available potassium and water-soluble organic matter in soil is completed by comparing the soil nutrient indicators calculated by the data conversion unit according to the real-time measurement values with the standard parameter values obtained from the cloud big data platform, in response to the user inserting the probe 70%-80% into the corresponding standard test solution and starting the calibration operation.

[0011] Preferably, the data conversion unit calculates the content indicators of available nitrogen, available phosphorus, available potassium and water-soluble organic matter in soil based on the preset regression analysis model, taking the soil conductivity value, soil temperature and soil moisture content as input variables.

[0012] Preferably, the output module includes a Bluetooth connection module for connecting a thermal printer to print the measurement data and reference data obtained from the big data platform.

[0013] Preferably, the data processing and communication terminal uploads the original soil physicochemical parameters, the multiple soil nutrient indicators and / or the comparison report to the cloud big data platform through a wireless network to realize data sharing; users remotely obtain soil nutrition parameters and adjust the soil according to the parameters.

[0014] Preferably, the data comparison and feedback module obtains the standard reference values of the planting variety from the cloud big data platform after the user inputs the planting variety, and compares the soil nutrient indicators output by the data conversion unit, and the comparison results are presented in a side-by-side display manner.

[0015] Preferably, the data comparison and feedback module is further configured to: based on the difference analysis in the comparison report, automatically generate a planting guidance scheme containing specific fertilization type and amount suggestions, and present the scheme to the user through the output module.

[0016] Preferably, the data processing and communication terminal further integrates a data storage module for locally storing the soil nutrient indicators calculated by the data conversion unit, the comparison report generated by the data comparison and feedback module, and the soil physical and chemical parameters uploaded by the probe detection device.

[0017] Compared with the prior art, the present application has the following beneficial effects: The present application realizes convenient and accurate calibration of the probe detection device measurement benchmark by introducing an integrated calibration control module and cloud big data platform linkage, solves the problems of large measurement error and poor long-term data stability of traditional devices due to the inability to calibrate; the intelligent data conversion unit built-in the system converts the original parameters into a complete data set containing multiple nutrient indicators, and combines the cloud crop standard parameter library for real-time comparison and analysis, and presents in the form of intuitive comparison report, changes the traditional detection data obscure and difficult to understand, and solves the difficulty of direct application; finally, a complete closed loop from accurate measurement, intelligent analysis to scientific decision is built, the user can quickly obtain the customized fertilization guidance scheme without professional knowledge, reduces the technical threshold and decision cost of precision agriculture, and improves the practical value of soil detection data and the intelligent level of agricultural management.

[0018] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and achieved by the structures indicated in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The present application is a system overall structure diagram. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work belong to the scope of the present application.

[0021] Please refer to Figure 1 The soil nutrient detection comparison system based on big data analysis management in the present application comprises a probe type detection device, a data processing and communication terminal and a cloud big data platform. Precise calibration of the device is realized through a calibration control module, original parameters are converted into multiple nutrient indicators by using a data conversion unit, intelligent comparison and analysis are performed in combination with a cloud standard library, and finally an intuitive comparison report and a customized fertilization scheme are generated, thereby constructing a technical closed loop from measurement to decision.

[0022] 1. System initialization and high-precision calibration process 1.1 Hardware power-on and self-checking The user turns on the "data processing and communication terminal" (such as a smart phone, a tablet computer, a computer, etc.). After the terminal loads an embedded operating system, the first hardware self-checking sequence is executed: Communication link check: By sending a handshake signal, it is confirmed that the LoRa wireless connection with the probe type detection device and the 4G / 5G network connection with the cloud platform are both normal.

[0023] Module state diagnosis: Check the algorithm model loading state of the internal data conversion unit, the calibration parameter storage memory access state of the calibration control module, and the remaining space of the data storage module.

[0024] Sensor preheating: Send instructions to the probe type detection device to make the PT1000 temperature sensor, pH glass electrode, etc. inside it enter a constant temperature preheating state for about 60 seconds to eliminate cold start drift.

[0025] After the self-checking is passed, the terminal interface displays "system ready" and prompts the user to calibrate to obtain the best measurement accuracy.

[0026] 1.2 Deep working mechanism of the calibration control module The user selects the "full parameter calibration" mode. At this time, the calibration control module becomes the command center of the system.

[0027] Step A: Standard value acquisition When the user inserts the probe type detection device into the "rapid potassium standard test solution" and presses the "start" key, the module first initiates an encrypted HTTPS request to the cloud big data platform through the unique device identification code (such as the MAC address) of the terminal.

[0028] The request message includes the "batch number" information of the standard test solution. The cloud platform searches its certified "standard substance database" and returns the authoritative standard value of the standard solution for that batch number at 25°C, for example: the standard concentration of readily available potassium. Standard value of conductivity , Standard value .

[0029] Step B: Measurement Values ​​and Preliminary Conversion The probe device continuously samples the standard solution for 30 seconds (sampling frequency 10Hz), removes obvious outliers (such as those exceeding 3 times the standard deviation), and calculates the average of the original measurements: for example, to obtain the original conductivity signal. Original pH voltage signal .

[0030] These raw signals are fed into the data conversion unit in real time. At this point, the unit performs calculations using the previously used calibration parameters. It will... according to equation (Where R is the gas constant, T is the measured temperature, and F is the Faraday constant) Convert to pH measurement value Similarly, the measured value of available potassium was calculated using the model. .

[0031] Step C: Setting manual calibration parameters based on intuitive comparison This step is part of the entire calibration process, designed to be visual, interactive, and verifiable, ensuring that users can directly control and trust the calibration results.

[0032] Calibration interface presentation and difference identification: The calibration control module generates a two-column comparison interface specifically designed for calibration on the screen of the data processing and communication terminal (such as a smart terminal).

[0033] Left sidebar: Displays authoritative standard parameter values ​​corresponding to the current standard test solution, obtained from the cloud big data platform.

[0034] Right sidebar: Real-time display of the initial nutrient index values ​​calculated by the data conversion unit based on the signal measured by the probe in the current standard solution.

[0035] Core design: Next to each indicator, the system dynamically calculates and displays the absolute deviation between the measured value and the standard value using a highlighted color (e.g., red indicates large deviation, yellow indicates acceptable, and green indicates good).

[0036] Interface example: [Calibration mode: Standard solution for available potassium - Batch: B2024001] | Index | Standard value | Current measurement | Deviation | |------------- |-------------|------------|-------------| | Conductivity (μS / cm) | 2080 | 2150 | +70 (too high) | | Available potassium (mg / kg) | 150.0 | 142.5 | -7.5 (too low) | | pH | 6.80 | 6.72 | -0.08 (too low) | Adjust calibration parameters manually: The system provides a set of clear manual adjustment controls for each index that needs calibration (e.g. available potassium), including a pair of '+' and '-' buttons or a numerical input box, allowing the user to directly adjust the calibration parameters. Calibration parameters include offset and calibration factor, and the system calculates and displays the corrected value in real time. The user adjusts the current measurement value to match the standard value by operating these controls, and the final calibration parameters are saved for subsequent mathematical correction.

[0037] Adjustment logic and real-time feedback: The user's goal is to directly modify the value displayed in the "Current measurement" column to match or closely approach the "Standard value" on the left side by operating these controls.

[0038] Example of operation: The user sees that the current measurement value of "available potassium" is 142.5 mg / kg, while the standard value is 150.0 mg / kg, with a deviation of -7.5.

[0039] The user clicks the "+" button corresponding to "available potassium". Each click increases the raw calculation value output by the data conversion unit by a fixed step size (e.g. 0.1 mg / kg) in the background, and immediately refreshes the "Current measurement" value on the screen.

[0040] The user continues to click the "+" button and observes that the "Current measurement" value on the screen gradually increases from 142.5 to 142.6, 142.7,... until it displays 150.0. At the same time, the "Deviation" column is also updated to 0.0 in real time.

[0041] The essence of this process is that every click the user makes through the interface is directly setting a final offset. The calculation performed internally by the system is: the measurement value displayed to the user = the data conversion unit's raw output value + the offset. The user, through his actions, is effectively adjusting this offset from 0 to +7.5.

[0042] Parameter confirmation and solidification: When the user has adjusted the "current measurement value" of all parameters to match the "standard value" through the above method, click the "Confirm and save calibration" button.

[0043] At this point, the calibration control module writes the calibration correction value (CAV) that the user has finally set for each parameter, i.e. Offset = +7.5 (for available potassium) in the above example, as a permanent calibration parameter into the non-volatile memory of the data storage module.

[0044] In any field measurement, the raw value calculated by the data conversion unit for available potassium will automatically have this saved offset (+7.5) added before being presented to the user (display, print, comparison). This mathematical correction process is automatic and transparent, ensuring that the final output value of the system tends to infinity towards the standard value.

[0045] The user does not need to understand the underlying model, only to focus on the screen numbers and match the two values through simple operations. By directly comparing with certified standard liquids, physical traceability of the measurement reference is achieved, ensuring the accuracy and stability of long-term measurement from the root.

[0046] 1.3 Calibration precision guarantee This process is performed for the eight parameters one by one, and the calibration precision of each parameter is guaranteed through the above closed-loop feedback control: Soil temperature: through comparison with the calibration liquid temperature probe (precision ±0.05℃), a control and adjustment precision of ±0.1℃ is achieved Soil conductivity: through calibration in standard liquids of different conductivity ranges (such as low, medium, high), a control and adjustment precision of ±50μS / cm is achieved within the full range.

[0047] Available nitrogen, phosphorus, potassium: through comparison with internationally certified standard substances, a control and adjustment precision of ±1.0 mg / kg is achieved.

[0048] 2. Field detection and core data conversion engine 2.1 Multi-parameter synchronous acquisition The user inserts the calibrated probe into the field soil. The microcontroller (MCU) inside the probe starts the synchronous acquisition sequence: TDR moisture sensor emits a 1GHz electromagnetic wave and measures the time difference from emission to reflection from the probe tip . According to (Where is the dielectric constant, which is proportional to the square of , the volumetric water content is calculated in real time .

[0049] Four-electrode EC sensor measures the resistance of the soil solution at a frequency of 1kHz, and calculates according to the probe constant .

[0050] pH electrode measures the potential difference relative to the reference electrode.

[0051] PT1000 measures the soil temperature and performs real-time compensation.

[0052] All these raw physical signals are packaged into a data packet and transmitted wirelessly to the data processing and communication terminal through LoRa.

[0053] 2.2 Intelligent model operation of data conversion unit After the raw data packet enters, the data conversion unit starts its preset algorithm model.

[0054] 2.21. Preprocessing and compensation: First, the original signal is preprocessed, including: Temperature compensation: For example, standardize the EC value and convert it to the standard value at 25°C: Where is the temperature coefficient.

[0055] Calibration parameter application: Call the offset and scale factor determined in the calibration phase to make preliminary corrections to the original signal.

[0056] 2.22. Model calculation: This unit adopts a multivariate nonlinear regression model. Through the collection of thousands of sets of laboratory standard measurement values corresponding to the probe raw signals covering different soil types and different nutrient levels, it is trained by using machine learning algorithms (such as gradient descent method).

[0057] The standardized signal after compensation and calibration is sent to the preset multivariate statistical prediction model. This model is trained by a large amount of historical laboratory data and can establish a nonlinear mapping relationship between the basic physical parameters and the complex nutrient content.

[0058] Taking the calculation of available nitrogen (N) as an example, its model may present as: where, , ,..., are the model coefficients trained, pH is calculated by

[0059] Parallel computing: Model parallel operation, one-time output of a complete soil nutrient index data set, at least including: water content (%), conductivity (μS / cm), soil temperature (℃), available nitrogen (mg / kg), available phosphorus (mg / kg), available potassium (mg / kg) and the like.

[0060] 2.23. Output: The data set is marked with time stamp and location information, and sent to the data storage module for local storage. The data storage module is used to save the soil nutrient indicators calculated by the data conversion unit, the comparison report generated by the data comparison and feedback module, and the soil physical and chemical parameters uploaded by the probe detection equipment, to ensure data traceability and offline access.

[0061] 3. Cloud intelligent comparison and personalized report generation This stage is the decision of the system, responsible for converting detection data into executable agricultural decisions.

[0062] 3.1 Dynamic query of cloud big data platform The user inputs or selects the planting variety (such as "summer corn") and growth period (such as "jointing stage") corresponding to this detection in the data processing and communication terminal. The data comparison and feedback module immediately constructs a structured query request and sends it to the cloud through the API gateway.

[0063] Cloud processing: After receiving the request, the cloud platform first determines the geographical area (such as the North China Plain) according to the GPS information uploaded by the terminal. In its knowledge base, it performs multi-dimensional matching (including variety, growth stage, region, soil type, target yield, etc.), dynamically generates and returns a set of optimized standard nutrient reference ranges for multi-dimensional matching, for example: Crop variety: "spring corn" Growth stage: "jointing stage" Soil type: (inferred from location information or user historical data) "brown soil" Result return: The platform returns a structured JSON object containing the most suitable nutrient index range under the condition. For example, the JSON format data is: ​{"nitrogen": {"min": 100, "max": 130, "unit": "mg / kg"}, "phosphorus": {"min": 35, "max": 55,...},...} 3.2 Generation logic of comparison report and fertilization scheme Upon receiving the cloud standard range, the data comparison and feedback module immediately compares each item in the measured data set output by the data conversion unit in depth.

[0064] Difference analysis: Determine "lack", "suitable" or "excess", and calculate the "degree of lack" For example, the measured value of available nitrogen is 80 mg / kg , the lower limit of the standard range is 100 mg / kg . The degree of lack is: Report generation: The data comparison and feedback module calls a report template engine to fill in the comparison results, measured data, user information, time and place, etc. into a pre-set HTML5 template to generate a comparison report containing difference analysis. The report highlights the status of each indicator with color coding (red / yellow / green).

[0065] Intelligent decision-making fertilization scheme generation: The module embeds a fertilization decision tree model.

[0066] Decision trigger: Recognize that "lack of available nitrogen" is the current main limiting factor.

[0067] Fertilizer selection: According to the soil pH and crop characteristics, select "urea" as the recommended nitrogen fertilizer from the built-in fertilizer library.

[0068] Dosage calculation: Use the principle of nutrient balance method for detailed calculation: Where: is the target nutrient value, usually taking the median value in the standard range, such as 105 mg / kg; is the soil bulk density (through user input or platform default, such as 1.35 g / cm 3 ); is the depth of the root activity layer (default 20 cm = 0.2 m); is the unit area, here referring to 1 mu = 667 m 2 ; is the nitrogen content of urea (46%); is the utilization rate of urea in the season (according to the region and experience, take 35%).

[0069] Calculation process: First, unify the units, and convert the target and measured concentration difference (105-85)=20 mg / kg into the amount of nitrogen needed to supplement per mu of cultivated soil.

[0070] Then calculate the amount of urea used:

[0071] Output of the scheme: Finally, generate the specific planting guidance scheme "Suggest: apply about 22 kg of urea per mu".

[0072] 4. Multi-modal output and data closed loop 4.1 Execution of the output module Screen display: The visualization report interface generated by the GPU rendering of the data processing and communication terminal, including data tables, column contrast charts, radar charts, and suggestion texts.

[0073] Bluetooth printing: The Bluetooth driver of the output module communicates with the paired thermal printer to convert the core content of the contrast report, multiple soil nutrient indicators, original physicochemical parameters, and fertilization guidance scheme into printer instruction sets (such as ESC / POS commands), and controls the printer to output a paper-based certificate.

[0074] 4.2 Data upload and sharing At the same time, the terminal automatically uploads all the data generated in this task, including the original soil physicochemical parameters, converted multiple soil nutrient indicators, generated contrast report, and fertilization scheme, to the cloud big data platform through asynchronous threads, achieving data sharing. Users can access these data through remote devices and adjust the soil according to these parameters. Cloud backup and multi-device sharing of user data are achieved, and these real field data (after desensitization) can supplement the cloud model to optimize future standard parameter library and conversion algorithm, forming a perfect data and intelligent closed loop.

[0075] The embodiment of the present application constructs a soil nutrient detection and decision support system integrating hardware sensing, edge computing, and cloud intelligence. The system realizes in-situ rapid detection of soil through a probe-type detection device, ensures measurement accuracy with the calibration control module of the data processing and communication terminal, intelligently converts basic parameters such as conductivity and temperature into multiple nutrient indicators such as available nitrogen, phosphorus, and potassium with the data conversion unit, and realizes dynamic comparison and analysis of detection data and crop standard requirements with the help of the cloud big data platform, finally generating a personalized planting scheme containing specific fertilization types and amounts, forming a technical closed loop from measurement, intelligent analysis to decision-making. The system converts professional soil detection technology into intuitive and easy-to-understand planting guidance, reduces the technical threshold of agriculture, and provides reliable technical support for modern agricultural production.

Claims

1. A soil nutrient detection comparison system based on big data analysis management, characterized in that, The utility model relates to a kind of soil nutrient detection system, comprising: Probe detection equipment is used to detect the original physicochemical parameters of soil by inserting into soil in situ; A cloud big data platform stores standard soil nutrient reference database classified according to different planting varieties and soil types; A data processing and communication terminal is connected with the probe detection equipment and exchanges data with the cloud big data platform through wireless network;The data processing and communication terminal is integrated with: a) Data conversion unit, preset with algorithm model for converting the original physicochemical parameters into multiple soil nutrient indicators, including at least water content, conductivity, soil temperature, available nitrogen, available phosphorus and available potassium content; b) Calibration control module, configured to obtain standard parameter values of standard test liquid from the cloud big data platform after receiving calibration instruction of inserting the probe detection equipment into standard test liquid by user, receive real-time measurement values of the probe detection equipment and convert them into multiple soil nutrient indicators through the data conversion unit, compare the multiple soil nutrient indicators with the obtained standard parameter values, and calculate calibration parameters for the data conversion unit based on comparison difference, which are used to mathematically correct subsequent conversion results of the data conversion unit to make output values close to standard values; c) Data comparison and feedback module, configured to receive planting variety information input by user, call corresponding standard reference data from the cloud big data platform, and compare the converted multiple soil nutrient indicators with the standard reference data to generate comparison report containing difference analysis; d) Output module, configured to display the comparison report, the multiple soil nutrient indicators and / or original soil physicochemical parameters on the screen of the data processing and communication terminal and / or print them through connected printing equipment.

2. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The calibration control module calibrates by inserting probe into standard test liquid, and its calibration accuracy meets at least one of the following: Soil temperature: control adjustment accuracy ±0.1℃; Soil water content: control adjustment accuracy ±0.1%; Soil conductivity: control adjustment accuracy ±50us / cm; Soil pH: control adjustment accuracy ±0.1; Available nitrogen content in soil: control adjustment accuracy ±1.0mg / kg; Available phosphorus content in soil: control adjustment accuracy ±1.0mg / kg; Available potassium content in soil: control adjustment accuracy ±1.0mg / kg; Water-soluble organic matter content in soil: control adjustment accuracy ±1.0g / kg.

3. The soil nutrient detection comparison system based on big data analysis management according to claim 1, characterized in that, The calibration control module is calibrated by the following method: Calibration of soil temperature, water content, conductivity and pH is completed by comparing multiple soil nutrient indicators calculated by data conversion unit according to real-time measurement values with standard parameter values obtained from cloud big data platform in response to calibration operation started by user inserting probe 70%-80% into water. The calibration of the soil available nitrogen, available phosphorus, available potassium and water-soluble organic matter is completed by comparing the soil nutrient indexes calculated by the data conversion unit according to the real-time measurement values with the standard parameter values obtained from the cloud big data platform in response to the user inserting the probe 70%-80% into the corresponding standard test solution and starting the calibration operation.

4. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The data conversion unit calculates the content indexes of the soil available nitrogen, available phosphorus, available potassium and water-soluble organic matter based on the preset regression analysis model, taking the soil conductivity value, soil temperature and soil moisture content as input variables.

5. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The output module includes a Bluetooth connection module for connecting a thermal printer to print the measurement data and the reference data obtained from the big data platform.

6. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The data processing and communication terminal uploads the original soil physicochemical parameters, the soil nutrient indexes and / or the comparison report to the cloud big data platform through a wireless network to realize data sharing; the user remotely obtains the soil nutrition parameters and adjusts the soil according to the parameters.

7. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The data comparison and feedback module obtains the standard reference values of the planting variety from the cloud big data platform after the user inputs the planting variety, and compares the values with the soil nutrient indexes output by the data conversion unit; the comparison results are presented in a side-by-side display manner.

8. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The data comparison and feedback module is further configured to automatically generate a planting guidance scheme including specific fertilization types and amount suggestions based on the difference analysis in the comparison report, and present the scheme to the user through the output module.

9. The soil nutrient detection and comparison system based on big data analysis management according to claim 1, characterized in that, The data processing and communication terminal further integrates a data storage module for locally storing the soil nutrient indexes calculated by the data conversion unit, the comparison report generated by the data comparison and feedback module, and the soil physicochemical parameters uploaded by the probe-type detection device.

Images