Soil detection method and system for soil improvement

Abstract

The invention relates to the technical field of soil environment monitoring, and particularly discloses a soil detection method and system for soil improvement. The method comprises the following steps: acquiring a real-time output signal of a sensor, identifying an initial moment when the signal fluctuates to define a monitoring time period, taking the signal in the time period as a parameter signal, dividing the parameter signal into a parameter group to generate a comprehensive score for representing the soil state, comparing the comprehensive score with a preset deviation value to judge whether a correction triggering condition is met or not, if yes, issuing a correction instruction, and calculating a correction time node based on the response time of the sensor and the reference factor to serve as the execution moment of correction operation. According to the method, decision making is carried out through comprehensive scores, instantaneous signal interference can be effectively smoothed, and timeliness and precision of system response are enhanced through accurate calculation of correction time nodes, so that the overall accuracy and reliability of detection are improved.

Description

Technical Field

[0001] This invention belongs to the field of soil environmental monitoring technology, specifically relating to a soil testing method and system for soil improvement. Background Technology

[0002] Soil quality is directly related to the sustainable development of agriculture. With the development of modern agriculture, the need for scientific management and improvement of soil conditions is becoming increasingly prominent. Real-time and accurate monitoring of key parameters such as soil moisture, salinity, and nutrient content is a prerequisite for realizing data-driven precision agriculture. Monitoring soil environmental data provides a basis for decision-making on dynamically adjusting irrigation, fertilization and other improvement measures to improve crop yield and resource utilization efficiency.

[0003] Existing soil quality testing technologies have many inherent shortcomings in practical applications. Traditional laboratory testing methods require on-site sampling, sample delivery, and laboratory analysis, resulting in long testing cycles. The data obtained cannot reflect real-time dynamic changes in the soil, causing adjustments to remediation measures to miss the intervention window and making it difficult to cope with sudden soil problems. Existing online monitoring solutions using sensors deployed in complex field environments are easily affected by environmental factors such as local airflow, instantaneous changes in surface temperature, and uneven water conduction. This makes it impossible to distinguish between transient data jumps caused by environmental disturbances and true, continuous trends in soil quality deterioration, leading to frequent false alarms. High false alarm rates can directly trigger unnecessary soil remediation actions, easily causing secondary damage to the soil due to excessive intervention.

[0004] Therefore, the industry urgently needs a new soil testing method and system to overcome the problems of inaccurate improvement measures and resource waste caused by unreliable data in existing technologies. Summary of the Invention

[0005] The purpose of this invention is to provide a soil testing method for soil improvement. This method can avoid the uncertainty of periodic tasks, autonomously identify the start time of signal fluctuations, and thus dynamically perform signal analysis.

[0006] Another object of the present invention is to provide a soil testing system for soil improvement that can perform the methods provided by the present invention.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows: A soil detection method based on smart sensors includes the following steps: Acquire the real-time output signals of the sensors within the detection area; A comprehensive score is generated based on real-time output signal processing: the starting moment of signal fluctuation in the real-time output signal is identified, and the time period after the starting moment is defined as the monitoring period; the real-time output signal within the monitoring period is defined as the parameter signal; the parameter signal is divided into at least one parameter group according to a preset classification rule; at least one parameter group is evaluated according to a preset scoring standard to generate a comprehensive score, which is used to characterize the soil condition; the comprehensive score is compared with a preset deviation value, and if the difference between the comprehensive score and the preset deviation value exceeds the preset difference range, the correction trigger condition is determined to be met. In response to the calibration triggering condition being met, a calibration command is issued.

[0008] Preferably, the step of evaluating at least one parameter group includes: dividing at least one parameter group into a first parameter group and a second parameter group; scoring the first parameter group and the second parameter group to obtain a first score and a second score respectively; and calculating a comprehensive score based on a first weight, a second weight, the first score, and the second score.

[0009] Preferably, the method further includes: calculating the correction time node and using the correction time node as the execution time of the correction operation corresponding to the correction instruction.

[0010] Preferably, the step of calculating the correction time node includes: identifying the starting moment when the signal fluctuates in the real-time output signal; defining the signal change value occurring within the sensor's response time as a reference signal; processing the reference signal to generate a reference factor; and calculating the correction time node based on the sensor's response time and the reference factor.

[0011] Preferably, the step of generating reference factors includes: performing weighted calculations on multiple data values ​​contained in the reference signal according to a preset functional relationship to generate an index for calculating the correction time node, wherein the reference factors include the index.

[0012] A soil detection system based on intelligent sensors includes: The signal acquisition module is used to acquire the real-time output signal of the sensor within the detection area; The state assessment and decision-making module is used to generate a comprehensive score to characterize the soil state based on real-time output signals, and compare the comprehensive score with a preset deviation value to determine whether the correction triggering condition is met. The instruction issuing module is used to issue correction instructions when the correction triggering conditions are met.

[0013] Preferably, the state evaluation and decision-making module, in processing and generating the comprehensive score, includes: identifying the starting moment of signal fluctuation in the real-time output signal, defining the time period after the starting moment as the monitoring period; defining the real-time output signal within the monitoring period as the parameter signal; dividing the parameter signal into at least one parameter group according to a preset classification rule; and evaluating at least one parameter group according to a preset scoring standard to generate a comprehensive score.

[0014] Preferably, the state evaluation and decision-making module evaluates at least one parameter group by: dividing at least one parameter group into a first parameter group and a second parameter group; scoring the first parameter group and the second parameter group to obtain a first score and a second score respectively; and calculating a comprehensive score based on a first weight, a second weight, the first score, and the second score.

[0015] Preferably, the status evaluation and decision-making module compares the comprehensive score with a preset deviation value to determine whether the correction trigger condition is met, including: if the difference between the comprehensive score and the preset deviation value exceeds a preset difference range, then the correction trigger condition is determined to be met.

[0016] Preferably, the state evaluation and decision module is further used to: calculate the correction time node and use the correction time node as the execution time of the correction operation corresponding to the correction instruction; The calculation of the correction time point includes: identifying the starting moment of signal fluctuation in the real-time output signal; defining the signal change value occurring within the sensor's response time as the reference signal; processing the reference signal to generate reference factors; and calculating the correction time point based on the sensor's response time and the reference factors.

[0017] Beneficial effects This invention divides the parameter signals within a monitoring period into at least one parameter group according to a preset classification rule, and performs a weighted evaluation of each parameter group according to a preset scoring standard to generate a comprehensive score. The comprehensive score is compared with a preset deviation value to determine whether to issue a correction command. Through the comprehensive scoring and comparison decision mechanism, a judgment is made based on an overall assessment of the multidimensional characteristics of the signal, rather than relying on a single threshold trigger. This can smooth out instantaneous signal fluctuations or reduce interference that may be caused by noise, and avoid misjudgments caused by local abnormal data, thereby improving the accuracy of soil condition assessment and the reliability of correction command issuance.

[0018] This invention acquires the response time of a sensor, defines the signal change value within the response time as a reference signal, processes the reference signal to generate a reference factor, and calculates a correction time node based on the sensor's response time, sensor delay parameters, and the reference factor as the basis for executing the correction operation. By combining the inherent physical characteristics of the sensor with the initial dynamic changes of the real-time signal, the correction operation is triggered at an effective time point, overcoming the problem of untimely intervention due to response lag and improving the response speed and accuracy of soil improvement.

[0019] This invention identifies the starting moment of signal fluctuations and defines the preceding period as the pre-detection period. Priority is assigned by extracting the difference between adjacent signals within this period. During the monitoring period after the starting moment, high-frequency components of the parameter signal are sampled to form amplitude data. The signal trend is determined based on the characteristics of amplitude data changes. This processing mode, which first analyzes the pre-detection period and then performs dynamic trend analysis on the monitoring period, can distinguish between background noise and actual signal changes. By establishing a judgment benchmark through the analysis of the pre-detection period, the accuracy of subsequent trend determination is enhanced, and the robustness and adaptability of the entire detection process to environmental changes are improved. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0021] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0022] Example 1 Please see Figure 1 This embodiment provides a soil testing method for soil improvement, suitable for real-time assessment and dynamic improvement of soil conditions in farmland or industrial parks. It distinguishes between real-world environmental changes and sensor noise or drift by segmenting the dynamic characteristics of the sensor signal, thereby achieving accurate and timely soil condition assessment and correction decisions. The method mainly includes the following steps: The system acquires the real-time output signals of the sensor array deployed in the detection area and uses this continuous signal stream as a detection sample. Depending on the application scenario, the sensors may include, but are not limited to, soil temperature and humidity sensors, EC conductivity sensors, and pH sensors, to monitor key soil physicochemical indicators that affect crop growth.

[0023] Furthermore, the starting moment of signal fluctuation in the real-time output signal is identified, and the period after the starting moment is defined as the monitoring period.

[0024] Specifically, the detected samples are processed to identify the precise time point when the signal transitions from a relatively stable state to a significantly fluctuating state. The processing includes continuous differential processing of the real-time output signal to obtain a signal change rate sequence, and setting a fluctuation trigger threshold. The fluctuation trigger threshold is a preset value used to compare with the absolute values ​​of data points in the signal change rate sequence to determine whether the signal has transitioned from a stable state to a significantly fluctuating state. When the absolute value of one or more consecutive data points in the signal change rate sequence first exceeds the fluctuation trigger threshold, the time point corresponding to that data point is determined as the start time. The start time marks the beginning of a potential soil condition change event that requires attention. The period before the start time is defined as the pre-detection period, which is used to assess the baseline stability and accuracy of the signal. The period after the start time is defined as the monitoring period, and the signal data within the monitoring period is used to analyze the trend and magnitude of the actual change event.

[0025] Furthermore, the analysis of the pre-detection period and the determination of signal accuracy are as follows: To ensure the reliability of subsequent analysis, the signal quality within the pre-detection period needs to be evaluated. The real-time output signals within the pre-detection period are arranged in chronological order, and the difference between two temporally adjacent real-time output signals is extracted. This difference sequence is defined as the judgment criterion. The judgment criterion is used to quantitatively assess whether there is unidirectional drift rather than normal random fluctuation in the signal within the pre-detection period. The sign and amplitude of the judgment criterion are statistically analyzed.

[0026] Specifically, the system statistically determines the number of positive and negative values ​​and calculates the total duration of their consecutive occurrences. A continuous period with a positive difference is defined as a positive period, and a continuous period with a negative difference is defined as a negative period. Then, based on the total duration of the positive and negative periods, a trend coefficient is calculated. This trend coefficient is a quantitative indicator that can determine whether there is unidirectional drift in the signal during the previous detection period. In this embodiment, the trend coefficient can be calculated as the ratio of the difference between the two to their sum. A drift judgment threshold is preset. Comparing the drift judgment threshold with the absolute value of the trend coefficient determines whether the signal accuracy is sufficient. When the absolute value of the trend coefficient is greater than a preset drift judgment threshold, i.e., when the coefficient value of the trend coefficient significantly deviates from zero, it indicates that the signal has obvious unidirectional drift, suggesting sensor aging or environmental interference. In this case, the signal accuracy is insufficient. Conversely, when the coefficient value of the trend coefficient is close to zero, it indicates that the signal mainly fluctuates symmetrically and randomly near the baseline, and the signal accuracy is high. Based on this judgment result, it can be decided whether to continue with subsequent steps or trigger a sensor calibration or maintenance alarm.

[0027] Furthermore, priority grading and reference signal establishment: After confirming the accuracy of the signal, in order to distinguish change events of different magnitudes and arrange the priority of responses, a first preset value and a second preset value are set. The first preset value is greater than the second preset value. The judgment criteria are prioritized according to the first and second preset values. Priority grading is used to determine the order of correction responses. Change events corresponding to judgment criteria where the absolute value of the change trend coefficient is greater than the first preset value are classified as high priority, indicating that a drastic change has occurred and an immediate response is required; change events where the absolute value of the change trend coefficient is between the first and second preset values ​​are classified as medium priority; and change events where the absolute value of the change trend coefficient is lower than the second preset value are classified as low priority.

[0028] Specifically, the signal change value occurring within the sensor's response time is defined as the reference signal. The grading result is used to determine the order of subsequent correction commands. At the same time, the sensor's inherent response time is obtained, which is the time required for the sensor's output signal to reach a stable state after being stimulated by external factors. A segment of signal change data collected after the initial moment within the sensor's inherent response time is defined as the reference signal. This reference signal captures the initial and rapid signal response characteristics triggered by external events, providing a key dynamic input for subsequent accurate calculation of correction timing. In this embodiment, the external event is irrigation or fertilization.

[0029] Furthermore, trend analysis during the monitoring period: To extract the true trend of change from the original signal containing noise, the parameter signal during the monitoring period is first subjected to high-frequency filtering. This can be achieved by calculating the average value within a sliding time window to smooth short-term disturbances and separate the low-frequency trend component and high-frequency component of the signal. The high-frequency component is periodically sampled to form amplitude data that reflects the intensity of short-term fluctuations in the signal. Historical amplitude data representing the normal fluctuation range is obtained from the stored data. The historical amplitude data is combined with the amplitude data generated during the current monitoring period to form the judgment benchmark data for trend judgment. Analyzing the dynamic characteristics of the judgment benchmark data can determine the overall trend of the parameter signal.

[0030] Specifically, the process of analyzing the dynamic characteristics of the benchmark data includes: calculating the moving average of the amplitude data within a time window, comparing the current slope of the moving average with a preset stable slope range, and determining whether the signal change trend is a stable trend. If the slope is significantly positive, the change trend is determined to be an upward trend; if the slope is significantly negative, the change trend is determined to be a downward trend; if the slope fluctuates within the stable slope range, the change trend is determined to be a stable trend. This effectively identifies long-term, unidirectional changes and distinguishes them from trendless, periodic short-term fluctuations.

[0031] Further, the reference signal is processed to generate a reference factor: The reference signal is processed to generate a reference factor. The processing includes: applying a set of preset weight coefficients to the data point sequence within the reference signal for weighted summation. The weight coefficient allocation rule is to assign higher weight values ​​to data points that are closer to the end of the response time in time. An index that can reflect the stabilization trend at the end of the signal change is calculated. The reference factor includes this index. A larger index value may mean that the signal tends to stabilize faster. The reference factor is used to reflect the stabilization trend at the end of the signal change and serves as one of the inputs for calculating the correction time node.

[0032] Furthermore, based on the sensor's response time and reference factors, the correction time node is calculated: taking into account the sensor's response time, a preset sensor delay parameter, and reference factors, the preset sensor delay parameter represents the time buffer between signal stabilization and reliable reading. The final correction time node is determined through a preset calculation logic. In this embodiment, the correction time node can be based on the initial time, plus the response time and sensor delay parameter, and finely adjusted according to the magnitude of the reference factors. The correction time node ensures that the triggering of the correction operation is based on a fully developed and stable signal state, rather than an instantaneous disturbance, and ensures that the correction operation is performed at the most appropriate time.

[0033] Furthermore, the parameter signals are divided into at least one parameter group according to a preset classification rule: In order to comprehensively assess the overall health status of the current soil, multiple parameter signals within the monitoring period are divided into a specific set as a parameter group according to the preset classification rule. In this embodiment, the preset classification rule is based on the agronomic significance of the parameters. The parameters that directly reflect the crop's nutrient absorption status, such as the nitrogen, phosphorus, and potassium concentrations in the soil solution, are classified into the first parameter group, while the environmental basic parameters that affect nutrient availability, such as soil temperature and moisture content, are classified into the second parameter group. To reflect the difference in importance between different parameter groups, a first weight and a second weight are set for the first parameter group and the second parameter group, respectively. Each parameter group is evaluated independently to obtain a score. The first weight and the second weight are used to reflect the importance of the corresponding parameter group.

[0034] Specifically, the process of independently evaluating each parameter group and generating a comprehensive score involves: comparing the current value of each parameter in the first parameter group with its respective ideal value range determined based on crop type and growth stage; finding and obtaining the corresponding score from a pre-defined score mapping table based on the degree of deviation from the ideal range; the score mapping table is a lookup table that records the mapping relationship between the degree of deviation of parameter values ​​from the ideal range and their corresponding scores; then, taking a weighted average of the scores of all parameters in the group to obtain the first score; applying a similar process to the second parameter group to obtain the second score; and finally, calculating the comprehensive score by weighting and summing the first weight, second weight, first score, and second score. The comprehensive score is a quantitative score that can comprehensively and selectively reflect the overall state of the current soil, providing a quantitative basis for subsequent decision-making.

[0035] Furthermore, the comprehensive score is compared with a preset target state reference value. The target state reference value is an ideal comprehensive score value determined based on the current crop growth stage, variety, and external climate conditions, representing the best soil health state. The difference between the comprehensive score and the target state reference value is calculated as a preset deviation value, and it is determined whether the absolute value of the preset deviation value exceeds a preset difference range. The preset difference range defines an acceptable range of score fluctuations that do not require intervention. The preset difference range is used to determine whether the difference between the comprehensive score and the target state reference value is significant, thereby determining whether intervention is needed. If the difference exceeds the preset difference range, it is determined that the correction trigger condition is met, indicating that the soil state has significantly deviated from the ideal state. In response to the correction trigger condition being met, a correction instruction is issued to initiate soil improvement measures. If the difference does not exceed the preset difference range, it is determined that no correction instruction is issued, and monitoring continues.

[0036] Furthermore, in a preferred embodiment of this method, after determining that a correction command has been issued, the real-time output signal value at the correction time node is obtained, and this set of real-time output signal values ​​is recorded as a correction reference quantity. This provides an accurate snapshot of the state before the correction for subsequent evaluation of the improvement effect. By comparing future sensor readings with this correction reference quantity, the actual effect after the correction command is executed can be quantitatively evaluated, thereby achieving closed-loop feedback and optimization of the soil improvement strategy.

[0037] Furthermore, in a preferred embodiment of this method, if significant fluctuations are found synchronously in signals from multiple sensors at different physical locations or measuring different parameters, the start time and duration of the fluctuations of each signal can be compared. If they are highly correlated in time, they can be collectively identified as a special event caused by a unified external factor such as concentrated irrigation or sudden weather changes. In the trend analysis step, the high-frequency component can be averaged within a sliding time window to smooth short-term disturbances and improve the accuracy of trend judgment. In the comprehensive scoring calculation, a correction coefficient based on soil type is introduced. In this embodiment, different correction coefficients are set for sandy soil, loam, and clay to adjust the scoring calculation logic, so that the method can better adapt to the physical and chemical properties of different soil textures, thereby having wider applicability.

[0038] Example 2 Please see Figure 2 This embodiment provides a soil detection system based on intelligent sensors. By evaluating the soil condition in real time and autonomously making decisions and issuing correction instructions when a deviation in condition is detected, it achieves closed-loop and intelligent management of the soil environment.

[0039] In its implementation, this system can be deployed in a central server, edge computing device, or embedded controller, and communicates with multiple smart sensors and related soil improvement execution modules deployed in the detection area via wired or wireless networks. Specifically, it includes the following modules: The signal acquisition module is configured to acquire real-time output signals from one or more sensors within a detection area and use these signal streams as detection samples. In specific applications, the detection area can be a farmland, a greenhouse, or an ecological monitoring plot. The sensors are smart sensors, which are one or more combinations of sensors capable of measuring soil pH, humidity, electrical conductivity (EC), nitrogen, phosphorus, and potassium content, and temperature. The signal acquisition module establishes a continuous data connection with these sensors and receives and aggregates real-time output signals at a preset sampling frequency. In this embodiment, the sampling frequency is once per minute. These signals are typically time-series data, where each data point contains a timestamp and the corresponding sensor measurement value. The signal acquisition module also has a built-in signal change detection logic for processing the detection samples. By comparing the signal change rate with the fluctuation trigger threshold, it identifies the starting time of significant signal fluctuations and divides the signal stream into a pre-detection period and a monitoring period based on the starting time, and then transmits them to subsequent modules.

[0040] The status assessment and decision-making module includes an analysis module and a decision-making module; The analysis module, one of the core processing units of this system, is responsible for analyzing and evaluating the received real-time output signals. It receives data from the monitoring period segmented by the signal acquisition module and executes the following processing flow: It performs a state evaluation process to generate a comprehensive score. The real-time output signals collected during the monitoring period are defined as parameter signals. This module divides the parameter signals into at least one parameter group according to preset classification rules. The preset classification rules classify parameters reflecting the intensity of signal fluctuations, such as variance and peak-to-valley difference, into the first parameter group, and parameters reflecting the overall offset level of the signal, such as mean and median, into the second parameter group. These parameter groups are evaluated according to preset scoring criteria, and scores are given to the first and second parameter groups respectively to obtain a first score and a second score. The scoring criteria can be a score mapping table. Based on preset first and second weights, the first and second scores are weighted and summed to calculate the final comprehensive score. This comprehensive score can comprehensively characterize the soil state during the current monitoring period.

[0041] The analysis module is also used to execute the calibration timing calculation process, calculate an optimal calibration time node, and define the signal change data occurring within the sensor's own response time as the reference signal. The response time is an inherent property of the sensor, representing the time required for it to go from sensing environmental changes to outputting a stable signal. The reference signal is processed to generate reference factors. According to a preset functional relationship, multiple data values ​​contained in the reference signal are weighted and calculated. In this embodiment, data closer to the current moment is given a higher weight to generate an index for calculating the calibration time node. This index is part of the reference factors. Based on the sensor's response time and the calculated reference factors, the final calibration time node is determined through a preset calculation model.

[0042] The decision module receives the comprehensive score from the analysis module and executes a correction decision process. It compares the generated comprehensive score with a target state reference value, which represents the ideal or acceptable soil state score. If the absolute value of the difference between the comprehensive score and the target state reference value exceeds a preset difference range, the module determines that the correction trigger condition is met and generates a decision signal, indicating that the current soil state has significantly deviated from the normal range and intervention is required.

[0043] The instruction issuing module responds to the decision signal from the decision module, receives the decision signal from the decision module and the correction time node calculated by the analysis module, generates and issues a correction instruction. The correction instruction includes the type of correction operation to be performed, such as irrigation or fertilization, and the execution time specified by the correction time node. The instruction is sent to the corresponding soil improvement execution module. In this embodiment, the instruction can control the water pump to start at the specified correction time node to perform the irrigation correction operation.

[0044] Through the coordinated work of the aforementioned signal acquisition module, state evaluation and decision-making module, and command issuance module, soil conditions can be monitored automatically and continuously. Based on a data-driven comprehensive evaluation model, anomalies can be accurately identified, and the optimal intervention time can be calculated, thereby achieving precise control of the soil environment. This technology is applicable to modern precision agriculture, smart landscaping, and other fields, and can improve management efficiency and resource utilization.

[0045] The above are merely preferred embodiments of this application and are not intended to limit this application. For those skilled in the art, this application can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A soil testing method for soil improvement, characterized in that, Includes the following steps: Acquire the real-time output signals of the sensors within the detection area; A comprehensive score is generated based on real-time output signal processing: the starting moment of signal fluctuation in the real-time output signal is identified, and the time period after the starting moment is defined as the monitoring period; the real-time output signal within the monitoring period is defined as the parameter signal; the parameter signal is divided into at least one parameter group according to a preset classification rule; at least one parameter group is evaluated according to a preset scoring standard to generate a comprehensive score, which is used to characterize the soil condition; the comprehensive score is compared with a preset deviation value, and if the difference between the comprehensive score and the preset deviation value exceeds the preset difference range, the correction trigger condition is determined to be met. In response to the calibration triggering condition being met, a calibration command is issued.

2. The soil testing method for soil improvement according to claim 1, characterized in that, The step of evaluating at least one parameter group includes: dividing at least one parameter group into a first parameter group and a second parameter group; scoring the first parameter group and the second parameter group to obtain a first score and a second score respectively; and calculating a comprehensive score based on a first weight, a second weight, the first score, and the second score.

3. The soil testing method for soil improvement according to claim 1, characterized in that, The method further includes: calculating the correction time node and using the correction time node as the execution time of the correction operation corresponding to the correction instruction.

4. A soil testing method for soil improvement according to claim 3, characterized in that, The steps for calculating the correction time node include: identifying the starting moment when the signal fluctuates in the real-time output signal; defining the signal change value occurring within the sensor's response time as a reference signal; processing the reference signal to generate a reference factor; and calculating the correction time node based on the sensor's response time and the reference factor.

5. A soil testing method for soil improvement according to claim 4, characterized in that, The step of generating reference factors includes: performing weighted calculations on multiple data values ​​contained in the reference signal according to a preset functional relationship to generate an index for calculating the correction time node, wherein the reference factors include the index.

6. A soil detection system based on intelligent sensors, characterized in that, include: The signal acquisition module is used to acquire the real-time output signal of the sensor within the detection area; The state assessment and decision-making module is used to generate a comprehensive score to characterize the soil state based on real-time output signals, and compare the comprehensive score with a preset deviation value to determine whether the correction triggering condition is met. The instruction issuing module is used to issue correction instructions when the correction triggering conditions are met.

7. A soil testing system for soil improvement according to claim 6, characterized in that, The state evaluation and decision-making module, in processing and generating the comprehensive score, includes: identifying the starting moment of signal fluctuation in the real-time output signal, defining the time period after the starting moment as the monitoring period; defining the real-time output signal within the monitoring period as the parameter signal; dividing the parameter signal into at least one parameter group according to a preset classification rule; and evaluating at least one parameter group according to a preset scoring standard to generate a comprehensive score.

8. A soil testing system for soil improvement according to claim 7, characterized in that, The state evaluation and decision-making module evaluates at least one parameter group by: dividing at least one parameter group into a first parameter group and a second parameter group; scoring the first parameter group and the second parameter group to obtain a first score and a second score respectively; and calculating a comprehensive score based on a first weight, a second weight, the first score, and the second score.

9. A soil testing system for soil improvement according to claim 6, characterized in that, The status evaluation and decision-making module compares the comprehensive score with the preset deviation value to determine whether the correction trigger condition is met. This includes: if the difference between the comprehensive score and the preset deviation value exceeds the preset difference range, then the correction trigger condition is determined to be met.

10. A soil testing system for soil improvement according to claim 6, characterized in that, The state evaluation and decision module is further used to: calculate the correction time node and use the correction time node as the execution time of the correction operation corresponding to the correction instruction; The calculation of the correction time point includes: identifying the starting moment of signal fluctuation in the real-time output signal; defining the signal change value occurring within the sensor's response time as the reference signal; processing the reference signal to generate reference factors; and calculating the correction time point based on the sensor's response time and the reference factors.

Images