An agricultural soil information management system

Abstract

This invention relates to the field of agricultural management technology, specifically to an agricultural soil information management system. The system includes: a data analysis module, a temperature difference determination module, a channel identification module, a correlation analysis module, and a management generation module. In this invention, multi-depth temperature information and inorganic nitrogen voltage signals from farmland soil profiles are collected to form a data set with time and depth identifiers. Based on changes in vertical temperature gradients, continuous depth thermal migration zones are identified. Further, the accumulated temperature difference results are screened to determine active microbial migration channels. Simultaneously, correlation analysis is conducted by combining time-varying temperature sequences with inorganic nitrogen concentration differences. Lag evolution step lengths are extracted, and historical temperature changes are traced back to form thermal migration memory variables. Finally, the nutrient evolution rate is calculated based on the current inorganic nitrogen concentration, and the information at set depth points is updated. This improves the ability to identify soil nutrient changes and enhances the supporting role of farmland information management data in nutrient regulation decisions.

Description

Technical Field

[0001] This invention relates to the field of agricultural management technology, and in particular to an agricultural soil information management system. Background Technology

[0002] The field of agricultural management technology refers to the technology of systematically managing and organizing matters such as land resources, crop growth environment, agricultural production activities and agricultural resource allocation involved in the agricultural production process by using information technology, sensing technology and data recording and analysis methods. It usually includes basic farmland information collection, agricultural resource data recording, agricultural production process management, soil environment monitoring, crop planting management and agricultural production data storage and retrieval. Through continuous recording, sorting and management of agricultural production environment and agricultural resource information, a systematic management system of relevant information in the agricultural production process is formed.

[0003] The traditional agricultural soil information management system refers to a system used to collect, record, and manage agricultural soil-related information. It mainly addresses technical matters such as recording farmland soil attribute information, managing soil nutrient data, collecting soil moisture and temperature information, and storing historical soil data. Typically, it obtains soil state data by deploying soil temperature probes, soil moisture probes, and soil conductivity probes in the farmland, and sends the collected data to a server for storage through data acquisition terminals. At the same time, it establishes farmland plot numbers, soil type record tables, soil nutrient test record tables, and soil monitoring time series tables in the database to register, query, and maintain soil information for different plots, thereby forming a centralized recording and management method for agricultural soil-related information.

[0004] Traditional agricultural soil information management systems focus on collecting and centrally storing soil temperature, humidity, and nutrient data. The data recording process is mostly based on single-point or single-layer monitoring results, mainly forming static data tables and time series records. There is a lack of systematic analysis of temperature differences and their changes at different soil depths. It is difficult to reflect the correlation between the heat migration state and nutrient transformation process within the soil profile. Soil environmental changes are often presented only as independent parameters, lacking the ability to identify and interpret the linkage changes between data at multiple depths. As a result, the judgment of soil nutrient change trends relies on human experience, making it difficult to reveal the dynamic relationship between temperature changes and inorganic nitrogen migration, thus affecting the accuracy and timeliness of farmland nutrient management decisions. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an agricultural soil information management system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an agricultural soil information management system includes: The data analysis module collects soil temperature and inorganic nitrogen voltage signals from multiple depths of farmland soil profiles and performs analog-to-digital conversion. It reads the corresponding time and depth identifiers and splices them with temperature and inorganic nitrogen concentration to generate a soil dataset and transmit it to the temperature difference determination module. The temperature difference determination module calculates the vertical temperature gradient and determines the sign based on the soil dataset, judges the consistency of the sign results for continuous depths and extracts the corresponding depth identifier, generates the thermal migration segment and transmits it to the channel identification module. The channel identification module accumulates the vertical temperature difference of the thermal migration section to obtain the temperature difference accumulation result, filters out the depth markers corresponding to the microbial activity thresholds that exceed the preset microbial activity threshold, generates active migration channels, and transmits them to the relevant analysis module. The correlation analysis module calculates the temperature time-varying sequence and inorganic nitrogen concentration difference based on the active migration channel, analyzes the correlation sequence and extracts the first and second values, generates the lag evolution step size and transmits it to the management generation module. The management generation module obtains the current time and temperature, analyzes the thermal migration memory variables by backtracking the corresponding time and temperature for the lag evolution step, calculates the nutrient evolution rate by combining the current inorganic nitrogen concentration, and overwrites the set depth point to generate information management data.

[0007] As a further embodiment of the present invention, the soil dataset includes depth labels, soil temperature values, and inorganic nitrogen concentration values; the thermal migration segment includes starting depth, ending depth, and gradient direction identifier; the active migration channel includes channel depth range, accumulated temperature difference value, and active threshold identifier; the lag evolution step size includes time lag amount, response position, and step size unit; and the information management data includes memory variable values, nutrient evolution rate, and nutrient status at target depth.

[0008] As a further aspect of the present invention, the data analysis module includes: The sensing signal submodule acquires the analog voltage signals output by soil temperature sensors and inorganic nitrogen sensors deployed at multiple depths in the farmland soil profile, monitors the potential changes at the voltage ports of multiple sensing nodes and records the corresponding time and sensor installation depth, calculates and organizes the potential amplitude, and generates multi-depth voltage sampling results. The numerical conversion submodule, based on the multi-depth voltage sampling results, calls the voltage calibration coefficients of the temperature sensor and the inorganic nitrogen sensor to perform analog-to-digital conversion on the multi-depth voltage sampling results, calculates the corresponding temperature and inorganic nitrogen concentration, and matches them with the original recording time and depth identifier to obtain the layered temperature and nitrogen concentration. The soil data submodule, based on the timestamp identifier of the layered temperature and nitrogen concentration reading and the sensor installation depth identifier, combines and organizes the temperature and inorganic nitrogen concentration in chronological order and depth sequence, establishes a unified field structure for multiple record rows and performs data integrity checks to generate a soil dataset.

[0009] As a further aspect of the present invention, the temperature difference determination module includes: The gradient calculation submodule, based on the soil dataset, detects the temperature corresponding to adjacent depth points and arranges them by depth, extracts the temperature difference between adjacent depths and the depth interval for pairing, performs vertical difference calculation based on the temperature difference and the depth interval, and analyzes the corresponding ratio of multiple adjacent depth points to obtain a vertical temperature gradient value sequence. The sign determination submodule extracts multiple gradient values ​​based on the vertical temperature gradient value sequence, performs sign determination operation on each gradient, detects the relationship between the value and the zero reference value and assigns positive or negative labels, arranges them according to depth labels and performs consistency judgment on continuous sign results to obtain the continuous interval of vertical gradient signs. The migration segment submodule reads the depth identifier set corresponding to multiple continuous intervals based on the vertical gradient symbol continuous intervals, extracts the start and end depth identifiers for each group of continuous symbol intervals and constructs segment identifier pairs, performs segment identifier combination operations, and generates hot migration segments.

[0010] As a further aspect of the present invention, the channel identification module includes: The temperature difference extraction submodule acquires a vertical temperature data sequence based on the thermal migration section, detects temperature records at adjacent depth nodes and subtracts values ​​point by point, constructs a vertical temperature difference sequence at multiple depth points, performs numerical calculations on the vertical temperature difference sequence to verify the consistency of the difference distribution, and generates a vertical temperature difference distribution. The cumulative calculation submodule calls the corresponding depth index sequence to perform point-by-point numerical superposition according to the vertical temperature difference distribution, records the cumulative temperature difference at multiple depth locations, performs cumulative calculation on the cumulative temperature difference value set, obtains the cumulative difference value at multiple depths, and generates the depth difference cumulative amount. The channel determination submodule obtains the standard parameter setting values ​​for carbon and nitrogen metabolism and establishes a table of microbial activity threshold values. Based on the accumulated depth difference, it calls the corresponding depth number for numerical comparison, filters out the depth identifiers corresponding to the microbial activity thresholds with accumulated depth differences, and generates active migration channels.

[0011] As a further aspect of the present invention, the process of calling the corresponding depth number for numerical comparison based on the depth difference accumulation is specifically as follows: the cumulative temperature difference value corresponding to multiple depth numbers in the depth difference accumulation is compared with the threshold value of the corresponding depth number in the microbial activity threshold value table one by one, and the depth number sequence in which the cumulative temperature difference value exceeds the microbial activity threshold value is recorded.

[0012] As a further aspect of the present invention, the relevant analysis module includes: The temperature sequence submodule extracts temperatures at consecutive timestamps from the soil dataset based on the active migration channel and arranges them in timetamp order to form a temperature sampling sequence. It then performs subtraction calculations based on adjacent timetamp temperature sampling sequences to generate a time-varying temperature sequence. The nitrogen concentration difference molecular module extracts the inorganic nitrogen concentration at continuous timestamps from the soil dataset and constructs a nitrogen concentration sampling sequence according to the timestamp order. Based on the timestamp segment corresponding to the temperature time-varying sequence, it locates the inorganic nitrogen concentration at adjacent time nodes, calculates the concentration difference, and generates the inorganic nitrogen concentration difference. The correlation ranking submodule synchronizes the temperature time-varying sequence with the inorganic nitrogen concentration difference, extracts the temperature change and concentration difference at the corresponding positions and calculates the time correlation, sorts the correlation set in descending order and extracts the time interval corresponding to the first value to generate the lag evolution step.

[0013] As a further aspect of the present invention, the process of sorting the correlation set in descending order and extracting the time interval corresponding to the first value specifically involves: pairing the time-varying temperature sequence with the inorganic nitrogen concentration difference in the same timestamp segment to construct time-corresponding data pairs; forming a correlation set based on the time correlation values ​​calculated from the time-corresponding data pairs; sorting the correlation set in descending order of correlation values; and extracting the timetamp segment length corresponding to the first correlation value as the lag evolution step size.

[0014] As a further aspect of the present invention, the management generation module includes: The time backtracking submodule obtains the current temperature and backtracks to the corresponding temperature according to the lag evolution step size. It calculates the time difference of the temperature sequence according to the time order and accumulates them to obtain the heat migration sequence. It then performs segment matching judgment in combination with the lag evolution step size to generate heat migration memory variables. The rate calculation submodule collects the corresponding depth inorganic nitrogen concentration based on the thermal migration memory variable, calculates the concentration gradient of the depth point concentration sampling value and forms a depth concentration gradient group, calls the thermal migration memory variable and the depth concentration gradient group to calculate the time-normalized evolution rate feature, and obtains the nutrient evolution rate. The overwrite generation submodule collects the depth number sequence and the corresponding depth point rate value, judges the numbering segment of the depth number sequence and reads the corresponding depth identifier overwrite rule value, replaces the nutrient evolution rate with the value and combines it with the corresponding time point number to generate information management data.

[0015] As a further aspect of the present invention, the calculation of the time-normalized evolution rate characteristics by calling the thermal migration memory variable and the depth concentration gradient group adopts the following formula: ; in, The time-normalized evolution rate characteristic is represented by n, which represents the total number of data elements in the depth concentration gradient group, and i represents the data element sequence number identifier associated with the summation symbol. Represents the time-normalized characteristic coefficients. Represents the hot migration memory variable associated with the i-th data element. This represents the equivalent concentration gradient mapping value of the i-th element in the depth concentration gradient group. This represents the background perturbation concentration corresponding to the i-th depth data item. represents the background fluctuation compensation concentration of the i-th data element, and W represents the global environmental damping attenuation factor.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, multi-depth temperature information and inorganic nitrogen voltage signals from farmland soil profiles are collected to form a dataset with time and depth markers. Based on the vertical temperature gradient changes, continuous depth thermal migration segments are identified. The accumulated temperature difference results are further screened to determine active microbial migration channels. Simultaneously, correlation analysis is conducted by combining the time-varying temperature sequence and the inorganic nitrogen concentration difference. The lag evolution step length is extracted, and historical temperature changes are traced back to form thermal migration memory variables. Then, the nutrient evolution rate is estimated by comprehensively considering the current inorganic nitrogen concentration, and the information at the set depth points is updated. This makes the relationship between soil profile temperature changes and nutrient migration a continuous correlation analysis process, thereby improving the ability to identify soil nutrient changes and enhancing the supporting role of farmland information management data in nutrient regulation decisions. Attached Figure Description

[0017] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart of the data analysis module of the present invention; Figure 3 This is a flowchart of the temperature difference determination module of the present invention; Figure 4 This is a flowchart of the channel identification module of the present invention; Figure 5 This is a flowchart of the relevant analysis module of the present invention; Figure 6 This is a flowchart of the management and generation module of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Please see Figure 1 An agricultural soil information management system includes: The data analysis module collects soil temperature and inorganic nitrogen voltage signals from multiple depths of farmland soil profiles and performs analog-to-digital conversion. It reads the corresponding time and depth identifiers and splices them with temperature and inorganic nitrogen concentration to generate a soil dataset and transmit it to the temperature difference determination module. The temperature difference determination module, based on the soil dataset, calculates the vertical temperature gradient and determines the sign, judges the consistency of the sign results for continuous depths and extracts the corresponding depth identifier, generates the thermal migration segment and transmits it to the channel recognition module. The channel identification module accumulates the vertical temperature difference of the thermal migration section to obtain the temperature difference accumulation result, filters the depth markers corresponding to the microbial activity thresholds that exceed the preset microbial activity threshold, generates active migration channels, and transmits them to the relevant analysis modules. The correlation analysis module, based on the active migration channel, calculates the temperature time-varying sequence and the inorganic nitrogen concentration difference, analyzes the correlation sequence and extracts the first and second values, generates the lag evolution step size and passes it to the management generation module; The management generation module obtains the current time and temperature, analyzes the thermal migration memory variables by backtracking the corresponding time and temperature for the lag evolution step, calculates the nutrient evolution rate by combining the current inorganic nitrogen concentration, and overwrites the set depth point to generate information management data.

[0020] The soil dataset includes depth labels, soil temperature values, and inorganic nitrogen concentration values. The thermal migration segment includes the starting depth, ending depth, and gradient direction identifier. The active migration channel includes the channel depth range, cumulative temperature difference value, and active threshold identifier. The lag evolution step includes the time lag, response position, and step unit. The information management data includes memory variable values, nutrient evolution rate, and nutrient status at the target depth.

[0021] Please see Figure 2 The data analysis module includes: The sensing signal submodule acquires the analog voltage signals output by soil temperature sensors and inorganic nitrogen sensors deployed at multiple depths in the farmland soil profile, monitors the potential changes at the voltage ports of multiple sensing nodes and records the corresponding time and sensor installation depth, calculates and organizes the potential amplitude, and generates multi-depth voltage sampling results. The system receives continuous analog voltage signals of 0 to 5 volts from PT1000 platinum resistance temperature sensors and polymer solid-film ion-selective electrode inorganic nitrogen sensors deployed at three depth levels (10 cm, 20 cm, and 30 cm) vertically downwards from the farmland soil profile. A built-in 16-bit analog-to-digital converter chip is configured to poll the multi-channel input ports, with a scan period set to 100 milliseconds. The captured raw analog potential string undergoes hardware-level denoising using a moving average filtering algorithm. Specifically, the instantaneous voltage values ​​of the preceding 10 time windows are extracted to form a buffer array. After removing the maximum and minimum values ​​from the array, the potentials of the remaining 8 sampling points are arithmetically averaged, and the average result is used as the effective voltage amplitude at the current timestamp. A reference potential fluctuation threshold of 0.05 volts is set. When the absolute value of the voltage amplitude change over three consecutive sampling periods is lower than this 0.05 volt threshold, the output potential of the corresponding sensor port is considered to be in a stable state. Under stable conditions, the 13-bit UNIX millisecond-level timestamp generated by the built-in high-precision real-time clock chip is extracted as the corresponding time identifier. For example, a timestamp of 1773381600000 represents March 13, 2026, 14:00:00. Simultaneously, the 4-bit hexadecimal address code pre-bound to the corresponding physical terminal is read. For example, port A0 corresponds to 0x00A1, indicating a depth of 10 cm, and this is used as the sensor installation depth identifier. The effective voltage amplitude, timestamp identifier, and depth identifier are combined to construct a 24-byte single-frame hexadecimal sampling data packet. For concurrent data streams from multiple sensor nodes, a 1024-byte circular buffer is allocated in memory. The single-frame sampling data packets are pushed into the buffer queue sequentially according to the timestamp order, generating multi-depth voltage sampling results.

[0022] The numerical conversion submodule calls the voltage calibration coefficients of the temperature sensor and the inorganic nitrogen sensor based on the multi-depth voltage sampling results to perform analog-to-digital conversion on the multi-depth voltage sampling results, calculates the corresponding temperature and inorganic nitrogen concentration, and matches them with the original recording time and depth identifier to obtain the layered temperature and nitrogen concentration. The system extracts the 4-digit hexadecimal depth identifier and channel number attribute from the data packets in the multi-depth voltage sampling results. Using the depth identifier as the index key, it initiates an addressing request to the static calibration parameter hash table in the built-in flash memory chip. Specific linear mapping conversion coefficients are returned for different sensor types. For a 10cm temperature sensor data packet identified as 0x00A1, the temperature sensor voltage calibration coefficient is retrieved, which includes a proportional gain coefficient of 25.5 and an offset constant of -20.0. For a 10cm inorganic nitrogen sensor data packet identified as 0x00B1, the inorganic nitrogen sensor voltage calibration coefficient is retrieved, which includes a logarithmic conversion gain coefficient of 15.2 and a baseline correction value of 2.5. The system receives the effective voltage amplitude from the corresponding data packet and performs a physical quantity analog-to-digital conversion. In the temperature conversion logic, the acquired 2.45 volt voltage amplitude is multiplied by the proportional gain coefficient of 25.5, and the product 62.475 is calculated. This product is then summed with the offset constant -20.0 to obtain the corresponding physical temperature value of 42.47 degrees Celsius. In the inorganic nitrogen concentration conversion logic, the difference between the actual voltage amplitude and the reference voltage of 1.2 volts is calculated. This difference is multiplied by a logarithmic conversion gain factor of 15.2, and then a baseline correction value of 2.5 is added to obtain the corresponding inorganic nitrogen concentration of 21.5 mg / kg. The calculated physical temperature and inorganic nitrogen concentration values ​​are all converted to 32-bit single-precision floating-point data. The 13-bit UNIX millisecond-level timestamp and 4-bit depth identifier from the original record are extracted. Through memory pointer mapping, the physical quantity values ​​are bound to the original record's time and depth identifier into the same structure variable to complete field matching. The parity check code of the data bits is verified. After confirming the conversion is correct, the memory address is locked to obtain the stratified temperature nitrogen concentration.

[0023] The soil data submodule reads the timestamp identifier and sensor installation depth identifier based on the stratified temperature and nitrogen concentration, combines and organizes the temperature and inorganic nitrogen concentration in chronological order and depth sequence, establishes a unified field structure for multiple record rows and performs data integrity checks to generate a soil dataset. The system iterates through the generated structure variables in the physical memory block, parsing the 13-bit UNIX millisecond-level timestamp identifier corresponding to the layered temperature and nitrogen concentration, along with the hexadecimal sensor installation depth identifier (e.g., 0x00A1). A doubly linked list data structure is initialized in memory. A bubble sort algorithm is used to prioritize the extracted timestamp identifiers in ascending order. When two records with identical timestamp identifiers are identified, they are considered to be from the same sampling batch. This triggers a secondary call to the depth sequence comparison logic, performing a second position swap on the records according to the physical depths of 10 cm, 20 cm, and 30 cm in ascending order. Temperature and inorganic nitrogen concentration are then strictly combined with the soil depth spatial sequence in chronological order. A 256-byte unified field structure is declared in memory, enforcing a standardized data storage paradigm for multiple records. The fields are divided into an 8-byte long integer timestamp, a 4-byte string depth number, a 4-byte single-precision floating-point temperature, a 4-byte floating-point nitrogen concentration, and a 2-byte status check code. The sorted combined information is written line by line according to the unified field structure. Subsequently, a data integrity check mechanism is initiated, scanning the legality of the floating-point field boundaries line by line. The system determines whether the temperature value in a single record falls within the reasonable range of -30 to 80 degrees Celsius, and whether the inorganic nitrogen concentration is within the effective detection range of 0 to 500 mg / kg. If an outlier value outside the range is detected in a record, or if a field is null due to packet loss, the status check code for that record is changed from the default 0x00 to 0xFF (error flag), and a physical deletion operation is performed. After filtering out abnormal and redundant items, all remaining checks are serialized and packaged using a structure record to generate a soil dataset containing header descriptions and data entities.

[0024] Please see Figure 3 The temperature difference determination module includes: The gradient calculation submodule, based on the soil dataset, detects the temperature corresponding to adjacent depth points and arranges them by depth. It extracts the temperature difference and depth spacing of adjacent depths and pairs them. It performs vertical difference calculation based on the temperature difference and depth spacing and analyzes the corresponding ratio of multiple adjacent depth points to obtain the vertical temperature gradient value sequence. The generated soil dataset is read, and pointer offsets are used to locate multiple record rows under the same 13-bit UNIX millisecond-level timestamp. A 4-digit string depth number and a 4-byte single-precision floating-point temperature field are extracted from each record row. For example, for the data cluster with timestamp 1773381600000, the corresponding temperature values ​​of 42.47 degrees Celsius at a depth of 10 cm, 38.25 degrees Celsius at a depth of 20 cm, and 35.10 degrees Celsius at a depth of 30 cm are extracted sequentially. The arithmetic logic unit is activated, and the acquired temperature points are grouped in pairs according to the physical spatial order from shallowest to deepest. The temperature difference between adjacent depths is obtained by subtracting the temperature of the lower depth from the temperature of the upper depth. The specific operation involves subtracting the temperature at 10 cm from the temperature at 20 cm (38.25 degrees Celsius) to obtain a first interlayer temperature difference of -4.22 degrees Celsius. Simultaneously, subtracting the temperature at 20 cm from the temperature at 30 cm (35.10 degrees Celsius) to obtain a second interlayer temperature difference of -3.15 degrees Celsius. A preset physical space configuration file is read, and the vertical installation distance between adjacent sensors is used as the depth spacing. In this scenario, the depth spacing between 10 cm and 20 cm, and between 20 cm and 30 cm, is precisely set to 0.1 meters. The calculated -4.22 degree Celsius temperature difference is divided by the 0.1-meter depth spacing to perform a vertical difference calculation, yielding a first interlayer vertical temperature gradient of -42.2 degrees Celsius per meter. Similarly, the second set of data is compared, with -3.15 degrees Celsius divided by 0.1 meters to obtain a second interlayer vertical temperature gradient of -31.5 degrees Celsius per meter. These floating-point ratios, along with their corresponding spatial start and end depths, are bound in memory. A single-precision floating-point array containing the above calculation results is then constructed in dynamic random access memory to obtain the vertical temperature gradient value sequence.

[0025] The sign determination submodule extracts multiple gradient values ​​based on the vertical temperature gradient value sequence, performs sign determination operation on each gradient, detects the relationship between the value and the zero reference value and assigns positive or negative labels, arranges them according to depth labels and performs consistency judgment on continuous sign results to obtain the continuous interval of vertical gradient signs. The vertical temperature gradient value sequence is traversed, and multiple gradient values ​​contained in the floating-point array are extracted one by one. The digital comparator unit inside the microcontroller is activated, a strict zero reference value of 0 degrees Celsius per meter is set, and a hardware-level dead zone determination threshold with a width of 0.5 degrees Celsius per meter is written into the register. This dead zone determination threshold is based on 50 statistical analyses of environmental thermal noise under calm wind conditions at night in farmland, and is specifically used to shield disordered small-amplitude potential jumps caused by white noise from high-sensitivity sensors. For the extracted first interlayer vertical temperature gradient of -42.2 degrees Celsius per meter, the digital comparator performs a logical comparison with the zero reference value and the dead zone threshold. Since -42.2 degrees Celsius per meter is less than the boundary of -0.5 degrees Celsius per meter, the logic control unit assigns it an 8-bit signed integer negative identifier of -1. For the second interlayer vertical temperature gradient of -31.5 degrees Celsius per meter, the same hardware logic comparison is performed, and a negative identifier of -1 is also assigned to it. If a gradient value detected by the scan is between -0.5 degrees Celsius per meter and +0.5 degrees Celsius per meter, a zero identifier (0) is forcibly assigned; if the value is greater than +0.5 degrees Celsius per meter, a positive identifier (1) is assigned. The assigned 8-bit signed integer identifiers are pushed into a first-in-first-out (FIFO) queue according to their corresponding physical spatial depth identifiers (e.g., 10-20 cm, 20-30 cm, etc.). A sliding window detection mechanism is activated, with a fixed window length of 2. The consecutive symbol results in the FIFO queue are bitwise XORed to perform a consistency check. Two adjacent negative identifiers (-1 and -1) are extracted and compared using a Boolean comparison. A zero XOR result indicates a consistent symbol state, and the logic unit merges these two consecutive spatial depth ranges within the address space. Through a global scan of the queue and merging of similar items, a continuous storage block with a spatial span from 10 cm to 30 cm and uniformly labeled with a negative identifier (-1) is partitioned in memory to obtain the continuous interval of vertical gradient symbols.

[0026] The migration segment submodule reads the depth identifier set corresponding to multiple continuous intervals based on the continuous intervals of vertical gradient symbols, extracts the start and end depth identifiers for each group of continuous symbol intervals and constructs segment identifier pairs, performs segment identifier combination operations, and generates hot migration segments. Obtain the physical starting address of the continuous interval of the vertical gradient symbol, and read the multiple sets of depth identifiers bound to this interval. Recognizing that the current continuous interval covers a physical space range from 10 cm to 30 cm, extract the address code corresponding to the top spatial node of this interval from the depth identifier set as the starting depth identifier, obtaining the hexadecimal code 0x00A1 representing 10 cm. Simultaneously, extract the address code corresponding to the bottom spatial node of this continuous interval as the ending depth identifier, obtaining the hexadecimal code 0x00C1 representing 30 cm. Allocate 16 bytes of continuous address space in the built-in static random access memory, and concatenate the extracted 0x00A1 and 0x00C1 in a high-low byte manner to construct a complete 16-bit segment identifier pair. Activate the hardware verification logic gate circuit, read the 8-bit signed integer identifier bound inside the segment identifier pair, and confirm that its state variable is a negative identifier -1. This negative identifier strictly corresponds to the vertical conduction behavior of heat from top to bottom in the physical thermodynamic mapping table. Based on the consistency status detection result, the arithmetic logic unit performs byte-level combination operations on the hexadecimal segment identifiers 0x00A1 and 0x00C1, the timestamp 1773381600000, and the negative status attribute representing downward propagation, reconstructing a new 32-byte data structure variable. A 4-byte cyclic redundancy check (CRC) code is appended to the end of this structure to ensure the tamper-proof integrity of subsequent data during bus transfer. After the combination operation is completed, it is pushed into the high-speed data bus buffer. This structure, containing start depth, end depth, timestamp, and heat conduction direction attributes, serves as an independent functional processing unit, generating hot-migrating segments on the bus.

[0027] Please see Figure 4 The channel identification module includes: The temperature difference extraction submodule obtains the vertical temperature data sequence based on the thermal migration section, detects the temperature records of adjacent depth nodes and subtracts them point by point, constructs a vertical temperature difference sequence at multiple depth points, performs numerical calculations on the vertical temperature difference sequence to verify the consistency of the difference distribution, and generates the vertical temperature difference distribution. The generated thermal migration segment structure variable is invoked, and the internal 0x00A1 start depth identifier and 0x00C1 end depth identifier are parsed. The floating-point temperature field mounted under the same 13-bit UNIX millisecond-level timestamp 1773381600000 is extracted, obtaining temperature records of 42.47 degrees Celsius at a physical location of 10 cm, 38.25 degrees Celsius at a physical location of 20 cm, and 35.10 degrees Celsius at a physical location of 30 cm. The arithmetic logic unit performs a subtraction operation on the temperature data of adjacent depth physical nodes, subtracting 38.25 degrees Celsius from 42.47 degrees Celsius to obtain the first inter-layer difference of 4.22 degrees Celsius, and subtracting 35.10 degrees Celsius from 38.25 degrees Celsius to obtain the second inter-layer difference of 3.15 degrees Celsius. A multi-depth point vertical temperature difference sequence containing the elements 4.22 degrees Celsius and 3.15 degrees Celsius is constructed in the dynamic random access memory. The microprocessor calls the floating-point unit to calculate the dispersion of the sequence to check the distribution consistency. It sums 4.22 degrees Celsius and 3.15 degrees Celsius, divides by the total number of elements (2) to obtain the arithmetic mean of 3.685 degrees Celsius. It then calculates and sums the squares of the deviations of each element from the mean, i.e., the square of 0.535 degrees Celsius plus the square of -0.535 degrees Celsius, obtaining the variance result as 0.57245 degrees Celsius squared. The microcontroller encapsulates the original difference sequence and this variance result into a 16-byte data packet to generate the vertical temperature difference distribution.

[0028] The cumulative calculation submodule calls the corresponding depth index order to perform point-by-point numerical superposition according to the vertical temperature difference distribution, records the cumulative temperature difference at multiple depth locations, performs cumulative calculation on the cumulative temperature difference value set, obtains the cumulative difference value at multiple depths, and generates the depth difference cumulative amount. The system addresses a 16-byte data packet of vertical temperature difference distribution temporarily stored in the microcontroller's internal memory mapping table, parsing it to extract a double-precision floating-point temperature difference sequence containing 4.22 degrees Celsius and 3.15 degrees Celsius. It then calls the pre-programmed sequential depth index table (0x00A1 to 0x00C1) in the built-in static random access memory. The hardware accumulator initializes its internal high-order register to zero according to the increasing physical mapping direction. The pointer points to the first physical layer interval (10 to 20 cm), and a floating-point addition operation is performed between the initial zero value of the register and the distribution value of 4.22 degrees Celsius obtained from the call within this interval. The result, 4.22 degrees Celsius, is directly written to and temporarily stored in the first accumulator register, thus mapping and recording the cumulative temperature difference at a physical depth of 20 cm as 4.22 degrees Celsius in memory. The memory addressing pointer moves down to the second physical level interval of 20 to 30 centimeters, triggering the cascaded addition hardware instruction of the arithmetic logic unit. This instruction sums the 4.22 degrees Celsius in the first accumulator register with the 3.15 degrees Celsius extracted from the second level, obtaining a superimposed result of 7.37 degrees Celsius, which is then written into the second accumulator register, completing the construction of the multi-depth location cumulative temperature difference record set. The direct memory access controller converts the generated 4.22 degrees Celsius and 7.37 degrees Celsius into a standard single-precision floating-point array and pushes it into the contiguous address space. This data is then packaged to construct multi-depth cumulative difference numerical data blocks, generating depth difference accumulation values ​​used to identify the intensity of thermal gradient accumulation.

[0029] The channel determination submodule obtains the standard parameter settings for carbon and nitrogen metabolism and establishes a table of microbial activity threshold values. Based on the cumulative depth difference, it calls the corresponding depth number for numerical comparison, filters out the depth identifiers corresponding to the microbial activity thresholds with cumulative depth differences, and generates active migration channels. The system reads pre-configured carbon and nitrogen metabolism standard parameter settings, which are derived from 50 sets of measured statistical mapping results of oxygen consumption and respiration rates of nitrifying bacteria in farmland soil within a temperature range of 15 to 40 degrees Celsius. The microcontroller's internal firmware establishes a table of microbial activity threshold values ​​based on the respiration heat equivalent conversion coefficient. Within this lookup table, the statically calibrated baseline for cumulative thermal stress is set to a constant 5.0 degrees Celsius. The system reads the depth difference accumulation array written to memory space by the preceding module via the high-speed data bus, activates the internal hardware digital comparator module, and uses the depth numbers 0x00B1 and 0x00C1 in the storage unit to perform bit-by-bit mapping and comparison with the data in the array. The hardware comparator extracts the cumulative temperature difference record of 4.22 degrees Celsius corresponding to the 20 cm depth marker. The input logic gate compares this difference with a 5.0 degree Celsius threshold. Since 4.22 degrees Celsius is less than 5.0 degrees Celsius, the judgment circuit outputs a low-level invalid signal. Subsequently, it extracts the cumulative difference of 7.37 degrees Celsius corresponding to the 30 cm depth marker 0x00C1 and compares it with the 5.0 degree Celsius threshold. The difference is 2.37 degrees Celsius, thus crossing the zero-potential threshold. The hardware comparator then toggles its output to a high level and triggers a kernel interrupt request. The interrupt service routine then extracts the high-level output of the 30 cm depth hexadecimal marker 0x00C1 into the active memory queue. After filtering the corresponding depth markers, it merges and packages them with the attributes of the upper conduction path, generating an active migration channel with strong biological metabolic exothermic characteristics within the buffer area.

[0030] Please see Figure 5 The relevant analysis module includes: The temperature sequence submodule, based on the active migration channel, extracts temperatures at consecutive timestamps from the soil dataset and arranges them in timetamp order to form a temperature sampling sequence. It then performs subtraction calculations based on adjacent timetamp temperature sampling sequences to generate a time-varying temperature sequence. The generated active migration channel identifier 0x00C1, characterized by strong biological metabolic exothermic activity, is read. This hexadecimal identifier strictly corresponds to a physical spatial node at a depth of 30 cm in the physical memory mapping table. Based on this addressing information, the data retrieval logic gate sends concurrent read commands to a high-speed solid-state drive connected to the peripheral component interconnect bus to retrieve temperature records corresponding to the past five consecutive sampling periods from the soil dataset file cluster. The extracted raw data stream is first subjected to hardware-level cleaning through the Kalman filter channel of the digital signal processor. A variance convergence threshold of 0.05 degrees Celsius is set to filter out high-frequency white noise from the sensor. After processing, five consecutive node temperature values ​​at the same depth of 30 cm are obtained, with timestamps from 1773381600000 to 1773396000000, which are 35.10 degrees Celsius, 35.50 degrees Celsius, 36.20 degrees Celsius, 37.10 degrees Celsius, and 38.50 degrees Celsius, respectively. The clock synchronization controller verifies the monotonically increasing attribute of the 13-bit UNIX millisecond-level timestamps embedded in each data packet. It pushes the verified numerical sequence into a first-in-first-out (FIFO) queue memory, constructing a one-dimensional single-precision floating-point temperature sampling sequence in the internal static random access memory. The arithmetic logic unit activates the hardware instruction pipeline, scheduling register resources to perform interleaved readings of adjacent nodes, sequentially inputting the backward and forward time node data into the subtractor hardware circuit. The subtractor performs floating-point subtraction operations with nanosecond-level delays. Specifically, 35.50 - 35.10 = 0.40 degrees Celsius, 36.20 - 35.50 = 0.70 degrees Celsius, 37.10 - 36.20 = 0.90 degrees Celsius, and 38.50 - 37.10 = 1.40 degrees Celsius. These incremental results are transported by the direct memory access controller to a contiguous physical address space and encapsulated into a dynamically reflecting temperature time-varying sequence array.

[0031] The nitrogen concentration difference molecular module extracts inorganic nitrogen concentrations at continuous timestamps from the soil dataset and constructs a nitrogen concentration sampling sequence according to the timestamp order. It locates the inorganic nitrogen concentrations of adjacent time nodes based on the timestamp segments corresponding to the temperature time-varying sequence, calculates the concentration difference, and generates the inorganic nitrogen concentration difference. The direct memory access controller, using the same 0x00C1 depth hexadecimal identifier addressing instruction, accesses the biochemical sensor log storage area in the soil dataset in parallel, reading the inorganic nitrogen concentration digital values ​​of the five consecutive timestamp nodes corresponding to the 30 cm depth. The input data flows through a hardware-level numerical range window comparator, configured with a lower limit of 0.01 mg / kg and an upper limit of 50.00 mg / kg, specifically used to remove out-of-bounds garbled data generated by the ion-selective electrode in dry soil. The valid inorganic nitrogen concentration values ​​are extracted into registers as 5.20 mg / kg, 5.35 mg / kg, 5.60 mg / kg, 5.90 mg / kg, and 6.40 mg / kg. The clock distribution network extracts the start and end timestamps of each interval in the aforementioned temperature time-varying sequence as reference alignment anchors, performs hardware timestamp registration between the inorganic nitrogen concentration values ​​and these anchors, and constructs a nitrogen concentration sampling sequence in static random access memory. The multiplexer channel within the triggering mechanism pumps the inorganic nitrogen concentration values ​​from adjacent time nodes into a dedicated floating-point subtractor core, strictly adhering to the time anchor alignment sequence. The subtractor hardware asynchronously performs the difference operation: 5.35 - 5.20 = 0.15 mg / kg, 5.60 - 5.35 = 0.25 mg / kg, 5.90 - 5.60 = 0.30 mg / kg, and 6.40 - 5.90 = 0.50 mg / kg. After the calculation, the direct memory access controller arranges the four floating-point differences in ascending order of memory address, generating an inorganic nitrogen concentration difference data block. To ensure end-to-end hardware traceability of biochemical data, the data transfer results are formatted as a table and stored in external flash memory.

[0032] Table 1 Time-series difference mapping table of inorganic nitrogen concentration As shown in Table 1, the concentration difference reflects the net biochemical metabolic output within the corresponding period.

[0033] The correlation ranking submodule synchronizes the temperature time-varying sequence with the inorganic nitrogen concentration difference, extracts the temperature change and concentration difference at the corresponding position and calculates the time correlation, sorts the correlation set in descending order and extracts the time interval corresponding to the first value to generate the lag evolution step. Align the time-varying temperature sequence with the inorganic nitrogen concentration difference data block, load the multiply-accumulate operation core to perform time correlation calculation, using the following formula: . Represents the time correlation series value; K represents the lag step size sequence number; K represents the total number of data nodes in the synchronization alignment sequence; j represents the node index sequence number; Represents the confidence level weight; and These represent the characteristic values ​​of temperature change at the corresponding nodes and their overall mean, respectively. and These represent the concentration difference feature values ​​and their overall mean after the superposition of lag steps, respectively. The logic circuit determines that the packet loss rate of the underlying network is below 5% and maps the weights accordingly. The constant is 1.0. Four consecutive valid nodes are extracted to obtain temperature characteristics. to Calculate the mean values ​​for 0.40, 0.70, 0.90, and 1.40 respectively. Set the lag step size to 0.85. The value is 1, and the measured inorganic nitrogen concentration characteristics are extracted again. to Calculate the mean values ​​for 0.25, 0.15, 0.30, and 0.50 respectively. The value is 0.30. First, the instantiation operation of the denominator of the formula is performed to obtain the standard deviation product. Then, the aforementioned parameters are substituted into the denominator formula: Subsequently, the instantiation operation of the numerator is performed to construct the eigenvalues ​​of the covariance matrix. The multiply-accumulator hardware circuit of the digital signal substitutes the same weights, temperature deviation, and concentration deviation into the numerator summation formula to perform parallel calculations: The hardware multiplier sequentially executes the product of the deviations at each node, obtaining the products of the deviations as 0.0225, 0, 0, and 0.11, respectively. These products are then added to the high-order register. After completing the loop summation instruction, the specific value of the numerator covariance term is calculated to be 0.155. Substituting the values ​​of the numerator and denominator into the formula... .

[0034] Table 2 Comparison of Hardware Calculation Results for Multi-Step Long-Term Correlation As shown in Table 2, the comparator hardware pushes the relevance set into a bubble sort logic gate array to perform a descending order operation, extracts the first highest value of 0.838, locks its corresponding memory pointer, and derives the lag time step number 1, generating the lag evolution step. This result indicates that the physical conduction cycle of the thermal mutation to the surge of biochemical metabolites is precisely locked as one standard sampling interval, which means that the actual biochemical response delay has been successfully quantified for generating the lag evolution step. The advantage of the formula lies in the introduction of dynamic confidence weights based on network communication packet loss rate quantification. It participates in the covariance and standard deviation scaling operation, effectively isolating the distortion amplification of biochemical coupling characteristics caused by the jitter of the underlying network.

[0035] Please see Figure 6 The management and generation module includes: The time backtracking submodule obtains the current temperature and backtracks to the corresponding temperature based on the lag evolution step size. It calculates the time difference of the temperature sequence according to the time order and accumulates them to obtain the heat migration sequence. It then performs segment matching judgment in combination with the lag evolution step size to generate heat migration memory variables. The internal integrated circuit bus reads the instantaneous temperature value corresponding to the current timestamp of the environmental monitoring probe, which is 38.50℃. The storage management unit receives the hysteresis evolution step size value of 1 from the output of the preceding arithmetic logic unit, and performs a hardware multiplication operation on this step size with the preset standard sampling period of 3600s to calculate the time offset as 3600s. The addressing logic gate circuit subtracts this offset from the current timestamp, generates a backtracking pointer in the high-speed static random access memory, and directly retrieves the backtracking temperature value of 37.10℃ for the corresponding time node. Based on this backtracking interval, the direct memory access controller extracts the temperature sequence of four consecutive nodes within the complete time window: [35.50℃, 36.20℃, 37.10℃, and 38.50℃]. The differential unit of the arithmetic logic unit performs floating-point subtraction operations on the temperature values ​​of adjacent time nodes according to the nanosecond-level clock beat, and successively calculates the temperature time difference subsets: 0.70℃, 0.90℃, and 1.40℃. Subsequently, the hardware accumulator is activated, storing the initial deviation of 0.70℃ into the high-order register. Subsequent difference values ​​are then added sequentially, and the calculation outputs 0.70℃, 1.60℃, and 3.00℃, encapsulating this set of increasing single-precision floating-point numbers into a heat migration sequence. The digital signal comparator hardware reads the preset segment matching threshold of 1.00℃ from the read-only memory. This threshold is set using a lookup table based on the statistical mean of the normal distribution of soil heat flux at the same depth over the past 30 days. The hardware logic gate performs a voltage-level parallel comparison of each value in the heat migration sequence with 1.00℃. The comparison logic is as follows: when the value is greater than or equal to 1.00℃, a high-level logic 1 is output; when the value is less than 1.00℃, a low-level logic 0 is output. After comparison, the first value of 0.70℃ is mapped to 0, and subsequent values ​​of 1.60℃ and 3.00℃ are mapped to 1. The shift register concatenates the generated binary sequence 011 to generate a heat migration memory variable containing 3 bits of characteristics.

[0036] The rate calculation submodule collects the corresponding depth inorganic nitrogen concentration based on the thermal migration memory variable, calculates the concentration gradient of the depth point concentration sampling value and forms a depth concentration gradient group, calls the thermal migration memory variable and the depth concentration gradient group to calculate the time-normalized evolution rate feature, and obtains the nutrient evolution rate. Based on the generated sequence of hot migration memory variables, in the hot migration memory variables The result values ​​0, 1, and 1, substituted into the preceding parameter sequence, are converted into numerical data that can be recognized by a floating-point arithmetic unit. The quantization process for this non-numerical Boolean variable is as follows: the hardware phase detector maps the abnormal heat conduction alarm level (high level) fed back by the sensor to a constant 1, and the background level (low level) where no abnormality was detected to a constant 0, corresponding to the detection results at depths of 10cm, 20cm, and 30cm respectively, thus obtaining... =0, =1, The value is 1. Subsequently, the multiplexer calls the thermal migration memory variable and the depth concentration gradient group to calculate the time-normalized evolution rate characteristic, using the formula: In this formula, Represents the time-normalized evolution rate characteristic; n represents the total number of data elements in the depth concentration gradient group; i represents the summation subscript; This represents performing a loop accumulation operation on the independent feature terms of all physical depths; Represents the time-normalized characteristic coefficients; Represents the hot-transfer memory variable after mapping for the i-th term; Represents the depth concentration gradient mapping value; Represents the baseline perturbation concentration; represents the background fluctuation compensation concentration; W represents the global environmental damping attenuation factor. In logic circuits, this formula is expressed through multiplication. Logical gating to achieve thermodynamic and kinetic characteristics, retaining only the effective gradients at the levels where thermal migration occurs; through absolute value and square root... Nested operations force the extraction of positive perturbation amplitude to avoid sign cancellation when positive and negative noise are added; division is used to... Background noise is scaled proportionally using an attenuation factor to prevent overcompensation; then, the noise compensation is removed by subtraction to purify the signal; addition integrates the effective gating concentration with the background purification perturbation; finally, after accumulation, the signal is divided by... Perform dimensional normalization across multiple time scales. (In the depth concentration gradient group) The resulting values, 0.15, 0.20, and 0.10, were obtained by substituting the preceding parameters. The background perturbation concentration was read from the idle calibration buffer. The concentrations were measured in dormant soil under windless, sunny conditions, with actual results of -0.04, 0.03, and -0.05, respectively. The low-pass filter was used to smooth the background fluctuations extracted from the past 24 hours of data to compensate for the concentrations. The values ​​are 0.02, 0.04, and 0.02, respectively. Regarding the global environmental damping attenuation factor W, it is calculated by dividing the measured soil volumetric moisture content by the standard reference moisture content of 10%. Since the real-time moisture content measured by the current capacitive moisture meter is 20%, the calculated specific setting value for W is 2.0. This setting reasonably reflects the physical hysteresis effect of moisture on nitrogen diffusion. Regarding the time normalized characteristic coefficient... It is set based on the comparison rate between the current sampling period and the standard time of 1 hour. Since the sampling window for this batch is accurate to 1 hour, The value is assigned to 1.0. The total number of elements, n, corresponds to 3 probe depths, and is assigned a value of 3. These values ​​are then substituted into the mentioned formula to perform the calculation: The internal sub-items of the loop accumulation instruction are calculated sequentially: the first item is... The second operation is The third operation is Finally, the hardware accumulator adds the three results together. The final result is the time-normalized evolution rate characteristic. The value is 0.38. The comparator hardware compares the calculated final value of 0.38 with the preset biochemical activity trigger interval [0.25, 0.50]. The value of 0.38 falls precisely within this interval. This result indicates that deep soil nutrients are currently in a stable and rapidly releasing active phase. For obtaining the nutrient evolution rate, this means that the interference from shallow static background has been successfully removed, and the precise quantification of the deep biochemical surge activity has been completed. This value will be directly latched and used as the core data source for subsequent rule overwriting steps. The advantage of the formula is that by introducing a global environmental damping attenuation factor W to participate in the division dimensionality reduction operation, the background fluctuations are dynamically scaled, effectively filtering out the sensor baseline drift interference caused by soil moisture infiltration.

[0037] The overwrite generation submodule collects the depth number sequence and the corresponding depth point rate value, judges the numbering segment of the depth number sequence and reads the corresponding depth identifier overwrite rule value, replaces the nutrient evolution rate with the numerical value and combines it with the corresponding time point number to generate information management data. The system receives and latches the output nutrient evolution rate value of 0.38, and simultaneously acquires the depth number sequence corresponding to the current probe array from the physical layer topology table. The acquired depth numbers are 0x0A, 0x14, and 0x1E, representing physical layers of 10cm, 20cm, and 30cm, respectively. The dual threshold comparator hardware within the arithmetic logic unit is activated to perform parallel logic judgment on the numbered segments of the input depth number sequence. The comparator is configured with a lower threshold of 0x00 and an upper threshold of 0x0F as the first decision window, and a lower threshold of 0x10 and an upper threshold of 0x1F as the second decision window. The logic gate circuit performs hardware-level judgment on the lowest-level number 0x1E. Since this value is greater than 0x10 and less than 0x1F, the comparator triggers the output pin of the second decision window, generating a high-level activation signal. The read-only memory receives this activation signal and reads the corresponding depth identifier overwrite rule value through address line offset addressing to obtain the rule correction coefficient of 1.20 used to compensate for physical errors in deep soil porosity. The floating-point multiplier hardware then invokes the rule correction coefficient to perform a value replacement operation on the original nutrient evolution rate value of 0.38. The multiplier calculates 0.38 * 1.20 = 0.456 in single-precision floating-point format using hardware multiplication logic to obtain the new rate value of 0.456. The shift register extracts the time point number 1255 from the clock distribution network's built-in timeframe header, converts it into a 16-bit binary code stream, and simultaneously converts the replaced value 0.456 into 32-bit machine code using a hardware encoder. The splicing logic gates strictly synchronize and integrate the two according to the bit-filling rule of storing the time point number in the high-order bits and the rate machine code in the low-order bits, completing the combined encoding. Based on this, the encoder synthesizer generates a continuous pulse data packet containing 48 bits, pushes it into the first-in-first-out buffer queue of the external data bus, and finally generates information management data that can be directly accessed by the host computer.

[0038] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An agricultural soil information management system, characterized in that, The system includes: The data analysis module collects soil temperature and inorganic nitrogen voltage signals from multiple depths of farmland soil profiles and performs analog-to-digital conversion. It reads the corresponding time and depth identifiers and splices them with temperature and inorganic nitrogen concentration to generate a soil dataset and transmit it to the temperature difference determination module. The temperature difference determination module calculates the vertical temperature gradient and determines the sign based on the soil dataset, judges the consistency of the sign results for continuous depths and extracts the corresponding depth identifier, generates the thermal migration segment and transmits it to the channel identification module. The channel identification module accumulates the vertical temperature difference of the thermal migration section to obtain the temperature difference accumulation result, filters out the depth markers corresponding to the microbial activity thresholds that exceed the preset microbial activity threshold, generates active migration channels, and transmits them to the relevant analysis module. The correlation analysis module calculates the temperature time-varying sequence and inorganic nitrogen concentration difference based on the active migration channel, analyzes the correlation sequence and extracts the first and second values, generates the lag evolution step size and transmits it to the management generation module. The management generation module obtains the current time and temperature, analyzes the thermal migration memory variables by backtracking the corresponding time and temperature for the lag evolution step, calculates the nutrient evolution rate by combining the current inorganic nitrogen concentration, and overwrites the set depth point to generate information management data.

2. The agricultural soil information management system according to claim 1, characterized in that, The soil dataset includes depth labels, soil temperature values, and inorganic nitrogen concentration values. The thermal migration segment includes starting depth, ending depth, and gradient direction identifier. The active migration channel includes channel depth range, accumulated temperature difference value, and active threshold identifier. The lag evolution step size includes time lag, response position, and step size unit. The information management data includes memory variable values, nutrient evolution rate, and nutrient status at target depth.

3. The agricultural soil information management system according to claim 1, characterized in that, The data analysis module includes: The sensing signal submodule acquires the analog voltage signals output by soil temperature sensors and inorganic nitrogen sensors deployed at multiple depths in the farmland soil profile, monitors the potential changes at the voltage ports of multiple sensing nodes and records the corresponding time and sensor installation depth, calculates and organizes the potential amplitude, and generates multi-depth voltage sampling results. The numerical conversion submodule, based on the multi-depth voltage sampling results, calls the voltage calibration coefficients of the temperature sensor and the inorganic nitrogen sensor to perform analog-to-digital conversion on the multi-depth voltage sampling results, calculates the corresponding temperature and inorganic nitrogen concentration, and matches them with the original recording time and depth identifier to obtain the layered temperature and nitrogen concentration. The soil data submodule, based on the timestamp identifier of the layered temperature and nitrogen concentration reading and the sensor installation depth identifier, combines and organizes the temperature and inorganic nitrogen concentration in chronological order and depth sequence, establishes a unified field structure for multiple record rows and performs data integrity checks to generate a soil dataset.

4. The agricultural soil information management system according to claim 1, characterized in that, The temperature difference determination module includes: The gradient calculation submodule, based on the soil dataset, detects the temperature corresponding to adjacent depth points and arranges them by depth, extracts the temperature difference between adjacent depths and the depth interval for pairing, performs vertical difference calculation based on the temperature difference and the depth interval, and analyzes the corresponding ratio of multiple adjacent depth points to obtain a vertical temperature gradient value sequence. The sign determination submodule extracts multiple gradient values ​​based on the vertical temperature gradient value sequence, performs sign determination operation on each gradient, detects the relationship between the value and the zero reference value and assigns positive or negative labels, arranges them according to depth labels and performs consistency judgment on continuous sign results to obtain the continuous interval of vertical gradient signs. The migration segment submodule reads the depth identifier set corresponding to multiple continuous intervals based on the vertical gradient symbol continuous intervals, extracts the start and end depth identifiers for each group of continuous symbol intervals and constructs segment identifier pairs, performs segment identifier combination operations, and generates hot migration segments.

5. An agricultural soil information management system according to claim 1, characterized in that, The channel identification module includes: The temperature difference extraction submodule acquires a vertical temperature data sequence based on the thermal migration section, detects temperature records at adjacent depth nodes and subtracts values ​​point by point, constructs a vertical temperature difference sequence at multiple depth points, performs numerical calculations on the vertical temperature difference sequence to verify the consistency of the difference distribution, and generates a vertical temperature difference distribution. The cumulative calculation submodule calls the corresponding depth index sequence to perform point-by-point numerical superposition according to the vertical temperature difference distribution, records the cumulative temperature difference at multiple depth locations, performs cumulative calculation on the cumulative temperature difference value set, obtains the cumulative difference value at multiple depths, and generates the depth difference cumulative amount. The channel determination submodule obtains the standard parameter setting values ​​for carbon and nitrogen metabolism and establishes a table of microbial activity threshold values. Based on the accumulated depth difference, it calls the corresponding depth number for numerical comparison, filters out the depth identifiers corresponding to the microbial activity thresholds with accumulated depth differences, and generates active migration channels.

6. An agricultural soil information management system according to claim 5, characterized in that, The process of calling the corresponding depth number for numerical comparison based on the depth difference accumulation is as follows: the cumulative temperature difference value corresponding to multiple depth numbers in the depth difference accumulation is compared with the threshold value of the corresponding depth number in the microbial activity threshold value table one by one, and the depth number sequence where the cumulative temperature difference value exceeds the microbial activity threshold value is recorded.

7. An agricultural soil information management system according to claim 1, characterized in that, The relevant analysis module includes: The temperature sequence submodule extracts temperatures at consecutive timestamps from the soil dataset based on the active migration channel and arranges them in timetamp order to form a temperature sampling sequence. It then performs subtraction calculations based on adjacent timetamp temperature sampling sequences to generate a time-varying temperature sequence. The nitrogen concentration difference molecular module extracts the inorganic nitrogen concentration at continuous timestamps from the soil dataset and constructs a nitrogen concentration sampling sequence according to the timestamp order. Based on the timestamp segment corresponding to the temperature time-varying sequence, it locates the inorganic nitrogen concentration at adjacent time nodes, calculates the concentration difference, and generates the inorganic nitrogen concentration difference. The correlation ranking submodule synchronizes the temperature time-varying sequence with the inorganic nitrogen concentration difference, extracts the temperature change and concentration difference at the corresponding positions and calculates the time correlation, sorts the correlation set in descending order and extracts the time interval corresponding to the first value to generate the lag evolution step.

8. An agricultural soil information management system according to claim 7, characterized in that, The process of sorting the correlation set in descending order and extracting the time interval corresponding to the first value is as follows: the time-varying temperature sequence and the inorganic nitrogen concentration difference are paired item by item in the same timestamp segment to construct time-corresponding data pairs. The time correlation values ​​calculated based on the time-corresponding data pairs form a correlation set. The correlation set is sorted in descending order of correlation values, and the length of the timestamp segment corresponding to the first correlation value is extracted as the lag evolution step size.

9. An agricultural soil information management system according to claim 1, characterized in that, The management generation module includes: The time backtracking submodule obtains the current temperature and backtracks to the corresponding temperature according to the lag evolution step size. It calculates the time difference of the temperature sequence according to the time order and accumulates them to obtain the heat migration sequence. It then performs segment matching judgment in combination with the lag evolution step size to generate heat migration memory variables. The rate calculation submodule collects the corresponding depth inorganic nitrogen concentration based on the thermal migration memory variable, calculates the concentration gradient of the depth point concentration sampling value and forms a depth concentration gradient group, calls the thermal migration memory variable and the depth concentration gradient group to calculate the time-normalized evolution rate feature, and obtains the nutrient evolution rate. The overwrite generation submodule collects the depth number sequence and the corresponding depth point rate value, judges the numbering segment of the depth number sequence and reads the corresponding depth identifier overwrite rule value, replaces the nutrient evolution rate with the value and combines it with the corresponding time point number to generate information management data.

10. An agricultural soil information management system according to claim 9, characterized in that, The calculation of the time-normalized evolution rate characteristics by invoking the thermal migration memory variable and the depth concentration gradient group uses the following formula: ； in, The time-normalized evolution rate characteristic is represented by n, which represents the total number of data elements in the depth concentration gradient group, and i represents the data element sequence number identifier associated with the summation symbol. Represents the time-normalized characteristic coefficients. Represents the hot migration memory variable associated with the i-th data element. This represents the equivalent concentration gradient mapping value of the i-th element in the depth concentration gradient group. This represents the background perturbation concentration corresponding to the i-th depth data item. represents the background fluctuation compensation concentration of the i-th data element, and W represents the global environmental damping attenuation factor.

Images