A closed-loop adaptive control method for straw-returning and biochar application

Abstract

This invention discloses a closed-loop adaptive control method for straw return to the field and biochar application. The method includes: calculating the response buffer amount of each soil region to be fertilized based on various scaling indices; comparing the response buffer amount with a preset range threshold to screen sensitive soil regions from each soil region to be fertilized; calculating the soil acidification rate of each sensitive soil region based on the predicted pH value sequence of each sensitive soil region and the pH value change sequence of each soil region to be fertilized at different times; calculating the multi-indicator synergistic sensitivity of each sensitive soil region based on the correlation between various dimensions of soil indicators; calculating the soil response type score of the sensitive soil region based on the soil acidification rate and the multi-indicator synergistic sensitivity; calculating the optimized application amount of straw return to the field and biochar based on the soil response type score and the preset original application amount; and achieving adaptive control of fertilization to improve utilization efficiency.

Description

Technical Field

[0001] This invention relates to the field of agricultural intelligent control technology, and relates to, but is not limited to, a closed-loop adaptive control method for straw return to the field and biochar application. Background Technology

[0002] Returning crop straw to the field and applying biochar are both ways to "turn waste into treasure" by returning crop straw to the soil. The former involves directly processing the straw before application, while the latter involves first pyrolyzing the straw to produce biochar before application. Combining these two methods and introducing closed-loop adaptive control can be understood as an intelligent system that can automatically and accurately decide how to apply these two materials based on real-time soil conditions to achieve the best soil improvement effect. Returning straw to the field is a long-standing method of soil fertilization, where crop straw such as wheat and corn is treated by direct covering, crushing and composting, or piled up for decomposition before being returned to the field. This provides the soil with organic matter, nitrogen, phosphorus, potassium, and other nutrients, improves soil structure, and increases water and fertilizer retention capacity. Applying biochar is an upgraded version of straw reuse, where straw is pyrolyzed at high temperatures under anaerobic or oxygen-limited conditions to produce a charcoal-like substance—biochar. Biochar has an extremely rich porous structure and a huge specific surface area, making it stable and resistant to decomposition. Therefore, applying biochar can not only directly improve soil structure, but more importantly, it can act like a sponge to fix carbon and reduce emissions in the long term, and adsorb and slowly release nutrients to reduce loss.

[0003] Real-time soil index data (organic matter, pH, etc.) are collected, and adaptive fertilization methods such as biochar and straw return are adopted based on the data results to achieve efficient and site-specific control. During the adaptive implementation process, differences in soil buffering capacity due to different fertilization histories (chemical fertilizer vs. organic fertilizer) can occur under the same soil indices, thus affecting subsequent fertilization responses (only analyzing the cross-sectional effect caused by current data). That is, two plots may currently have identical pH, EC, and organic matter, but one part of the soil has been fertilized with organic fertilizer for a long time, while the other part has been fertilized with chemical fertilizer for a long time. After applying the same amount of biochar (or straw return), the biochar response is gradual, while straw return may cause a sharp increase in pH and micronutrient (Zn / Fe) deficiency due to the poor stability of the soil system response, resulting in errors in fertilizer selection and application rate control, and affecting soil yield. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a closed-loop adaptive control method for straw return to the field and biochar application, which aims to overcome the defects of the prior art and address the problem of crop yield being affected by control errors in fertilizer selection and fertilizer amount.

[0005] The specific technical solutions of this invention are as follows: This invention provides a closed-loop adaptive control method for straw return to the field and biochar application, comprising: Configure a pre-test, dividing the pre-test area into several test plots; at multiple preset time points, collect time series data of multi-dimensional soil indicators for each test plot's soil area to be fertilized under each test, wherein the multi-dimensional soil indicators include at least pH value; The DFA algorithm was used to analyze the time series data of soil indicators of each dimension in each soil area to be fertilized to obtain the scaling index corresponding to each dimension, and the response buffer of each soil area to be fertilized was calculated based on the scaling index. The response buffer was compared with a preset range threshold to screen out sensitive soil areas from each soil area to be fertilized. Using historical multidimensional soil index time series data of each sensitive soil region as input, a time series prediction model is constructed to output the prediction sequence of multidimensional soil index of each sensitive soil region; based on the prediction sequence of pH value of each sensitive soil region and the change sequence of pH value of each soil region to be fertilized at different times, the soil acidification rate of each sensitive soil region is calculated. Based on the correlation between various soil indicators in each sensitive soil region, the synergistic sensitivity of multiple indicators in each sensitive soil region is calculated; based on the soil acidification rate and the synergistic sensitivity of multiple indicators, the soil response type score of the sensitive soil region is calculated; based on the soil response type score and the preset original application rate, the optimized application rates of straw return and biochar are calculated, and adaptive fertilization is achieved.

[0006] In some embodiments, calculating the response buffer size of each soil region to be fertilized based on each of the scaling indices includes: The response buffer capacity of each soil zone to be fertilized is calculated using the following formula:

[0007] in, Indicates the first The response buffer capacity of each soil area to be fertilized; The dimension number representing the soil index; For the first Under the soil area to be fertilized, the first Scaling indices were obtained by analyzing time-series data of soil indicators across multiple dimensions using the DFA algorithm.

[0008] In some embodiments, calculating the soil acidification rate of each sensitive soil area based on the predicted pH value sequence of each sensitive soil area and the pH value change sequence of each soil area to be fertilized at different times includes: The soil acidification rate for each sensitive soil region is calculated using the following formula:

[0009] in, Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first The mean of the predicted pH values ​​for a single sensitive soil region; Indicates the first Historical baseline pH mean values ​​for each sensitive soil region; Indicates the first The trend intensity score of the pH prediction sequence for a sensitive soil region is obtained by decomposing the trend component and the residual component using a trend intensity algorithm for time series decomposition. The trend intensity score is then calculated based on the variance of the trend component and the variance of the residual component.

[0010] In some embodiments, calculating the soil response type score of the sensitive soil region based on the soil acidification rate and the multi-indicator synergistic sensitivity includes: The soil response type score for sensitive soil areas is calculated using the following formula:

[0011] in, Indicates the first Soil response type score for each sensitive soil region; Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first Multi-indicator synergistic sensitivity of sensitive soil regions; This represents the normalization function.

[0012] In some embodiments, implementing adaptive fertilization includes: Based on the optimized application rates of straw return to the field and biochar, the corresponding materials are applied to the corresponding sensitive soil areas. After the application of materials is completed, the multi-dimensional indicators of the sensitive soil areas are continuously monitored. The parameters of the time-series prediction model are updated using the time series data of the monitored soil indicators. The material application rate for the next cycle is iteratively optimized to achieve closed-loop adaptive fertilization of straw return to the field and biochar.

[0013] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, sensitive soil domains are determined based on response buffer capacity; the sensitive soil domains are predicted using response result samples to obtain a prediction set; the acidification rate, a single-dimensional indicator in the prediction set, is analyzed, and combined with the synchronous sensitivity of multi-dimensional indicators to determine the soil response type score. Feedback control is then used to determine the application type and amount, achieving adaptive precision fertilization. During soil fertilization, different soils have different fertilization needs. The cost of straw return to the field is much lower than that of biochar. By analyzing relevant soil indicators, corresponding straw return to the field or biochar is used for adaptive fertilization based on different directional results, improving fertilization efficiency and reducing fertilization costs. Since soil fertilization in the current state only considers the current moment and lacks attention to previous and subsequent moments, different fertilization methods (chemical fertilizers and organic fertilizers) have different effects on the soil. To eliminate the control error in subsequent fertilization amounts caused by different soil fertilization types, small-dose application is designed and predicted. Based on the prediction results at different time stages, the subsequent soil response type is judged, and then feedback control is used to determine the application type and amount, improving fertilization precision. Different historical fertilization results lead to different requirements for straw return to the field and biochar application. Excessive or insufficient demand will cause soil indicators to deviate from the normal baseline value. By analyzing and predicting the results, the soil acidification rate and the synchronous sensitivity between multiple indicators are quantified. Combined with the feedback degree of different soil types to the application amount, the amount and type of fertilizer are adaptively controlled to improve utilization efficiency. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is a schematic flowchart of a closed-loop adaptive control method for straw return to the field and biochar application provided in an embodiment of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0017] It should be noted that the terms "first, second, and third" used in the embodiments of the present invention are only used to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, and third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present invention described herein can be implemented in an order other than that illustrated or described herein.

[0018] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which these embodiments of the invention pertain. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0019] Soil "response stability" refers to the ability of a soil system to maintain dynamic equilibrium or undergo gradual changes in its physicochemical properties under external disturbances (such as fertilization). It is positively correlated with the soil's chemical buffering capacity (determined by CEC, organic matter, etc.), but focuses on describing the system's time-dynamic response characteristics rather than its static chemical properties. Soils that have been fertilized with chemical fertilizers for a long time experience continuous mineralization and depletion of organic matter, resulting in a system response that tends to fluctuate randomly (similar to white noise) and exhibits poor response stability. Soils that have been fertilized with organic fertilizers for a long time have abundant organic matter and humus, resulting in a system response that exhibits continuous characteristics and strong response stability.

[0020] Response effect: refers to the phenomenon that, under the same current soil index value, soil produces different responses to the same fertilization measures due to different historical fertilization paths (long-term chemical fertilizer vs. long-term organic fertilizer).

[0021] Closed-loop adaptive control: refers to a control mechanism in which the system continuously monitors the actual response after making a fertilization decision, evaluates the prediction deviation, and triggers model parameter updates when the deviation is too large.

[0022] Figure 1 This is a flowchart illustrating a closed-loop adaptive control method for straw return to the field and biochar application provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes at least the following steps: Step S110: Configure the pre-test and divide the pre-test area into several test plots; at multiple preset time points, collect time series data of multi-dimensional soil indicators for each test plot of the soil area to be fertilized under each test, including at least pH value.

[0023] It should be noted that the system is configured and small-dose break-in test data is collected to obtain the response results of different soil regions under short-term, multi-stage conditions (the type of dose is classified according to general data, and the amount is about 30%, emphasizing that the time must be short, otherwise it will not be effective).

[0024] Delineate the experimental area: Divide the pre-experimental area into several experimental plots according to soil type, planting history and topographical differences. Each experimental plot shall have an area of ​​not less than 100 square meters. The purpose is to ensure clear boundaries and prevent cross-contamination of water and fertilizer.

[0025] Background data collection: Within each experimental plot, 5-7 sampling points were set up along an S-shaped route, and soil samples from the 0-20cm topsoil layer were collected after mixing. The following indicators were measured: pH, EC, organic matter, total nitrogen, available phosphorus, available potassium, cation exchange capacity (CEC), and soil bulk density (8 indicators). Historical fertilization records for each experimental plot were recorded (type, amount, and frequency of chemical fertilizers / organic fertilizers / straw return over the past 3 years).

[0026] Unique ID: Generate a unique code for each test cell and bind the background data with the ID to the system database.

[0027] First application during the break-in period: Determine the application plan: Based on the conventional recommended fertilization rate for the experimental area, uniformly adjust the dosage of straw return to the field or biochar to 20% of the preset conventional recommended rate, which is based on experience. If the conventional practice in the experimental area is straw return to the field, then apply 20% of the straw dosage; if the conventional practice in the experimental area is biochar, then apply 20% of the biochar dosage.

[0028] Precise application: Using manual or small-scale mechanical methods, evenly spread 20% of the material onto the surface of the corresponding test plot, followed by shallow rotary tillage (10-15cm deep) to fully mix the material with the topsoil.

[0029] Initial response data collection: On the 7th day after application, soil samples were collected from the 0-20cm topsoil layer in each experimental plot using the 5-point sampling method. pH, EC, organic matter, available nitrogen, available phosphorus, available potassium, cation exchange capacity (CEC), and soil bulk density were measured, and soil temperature and moisture content were recorded.

[0030] Application during the break-in period (Nth time): Nth application: 14+N days after the first application Fifteen days later (i.e., the 7th day after the first response data collection), the same type of material was applied again in each test plot at the same dosage (20% of the recommended amount), and the operation method was the same as the first time.

[0031] Nth response data collection: On the 7th day after the Nth application, soil index data were collected again, and the test indicators were the same as those collected in the first response data collection.

[0032] Construct a time series dataset: Collect all detection data of each experimental plot at multiple preset time points (e.g., three preset time points: T0 background data (i.e., soil test data before application), T1 7 days after the first application, and T2 7 days after the second application) by plot ID to form a complete response sequence.

[0033] Specifically, at multiple preset time points, time series data of multi-dimensional soil indicators for each experimental plot's soil area to be fertilized are collected for each test under each detection. The multi-dimensional soil indicators include at least pH value.

[0034] It should be noted that the historical fertilization type (chemical fertilizer-dominated / organic fertilizer-dominated / straw return-to-field-dominated) of each experimental plot is labeled in the dataset as label information for subsequent model training. All sequence data are entered into the system as the basic training samples for the subsequent long-term decision-making model.

[0035] The entire trial, from the first application to the completion of data collection, lasted 21 days, with each application being 20% ​​of the recommended dose. This trial captured the differentiated response characteristics of soils with varying historical backgrounds to the same intervention through short-term, multi-stage response curves, providing data support for subsequent personalized fertilization prediction models.

[0036] It should be noted that sensitive soil domains are determined based on response buffer capacity; the sensitive soil domains are predicted and processed using response result samples to obtain a prediction set; the acidification rate of the single-dimensional index in the prediction set is analyzed, and combined with the synchronous sensitivity of multi-dimensional indexes, the soil response type score is determined, and the application type and amount are controlled by feedback to achieve the purpose of adaptive precision fertilization.

[0037] It should be noted that soils with different fertilization backgrounds will have different effects on the application of "straw return" and "biochar". Long-term application of chemical fertilizers can lead to a sharp rise in parameters such as pH due to weak buffering capacity, which will affect subsequent crop yields. Therefore, in order to achieve precise control of application in a closed loop, it is necessary to identify different soil types (organic fertilizer or chemical fertilizer) and adaptively control the type of fertilizer (straw return or biochar) and the amount of fertilizer applied according to the type.

[0038] For example, taking cation exchange capacity (CEC) as an example, chemical fertilizers mainly provide readily available nutrients but do not directly replenish organic matter. Without external organic matter replenishment, the soil's original organic matter will be continuously consumed through mineralization, leading to a reduction in the core carriers of CEC and a decrease in fertilizer retention capacity. Consequently, the effects of fertilization on soil changes will be more pronounced, potentially causing problems such as ion accumulation, and its buffering capacity against fertilization is weak. Organic fertilizers, on the other hand, are a direct source of organic matter and humus. These substances carry a large number of negative charges on their surfaces and mainly participate in the formation of CEC. Therefore, such soils have stronger water and fertilizer retention capacity, and thus a stronger buffering capacity against fertilization.

[0039] It should be noted that soil index parameters will exhibit the above-mentioned variation characteristics due to their inherent features; only CEC will be analyzed as an example.

[0040] For example, the set of detection results corresponding to each test cell is obtained, and this set is the first... Under the soil area to be fertilized, the first Each test result contains soil index data from multiple dimensions, such as pH, EC, organic matter, available nitrogen, available phosphorus, available potassium, soil temperature, and moisture content. Each test result consists of soil indexes from different dimensions within the same soil region at the same testing time, as specifically represented as follows:

[0041]

[0042]

[0043] The first column of data , This indicates the test results for the first data point in the first soil area to be fertilized; similarly, the second column... , This indicates the result of the second data test in the first soil area to be fertilized, and so on.

[0044] Step S120: Perform DFA algorithm analysis on the time series data of soil indicators of each dimension in each soil area to be fertilized to obtain the scaling index corresponding to each dimension, and calculate the response buffer amount of each soil area to be fertilized based on each scaling index; compare the response buffer amount with the preset range threshold, and screen out sensitive soil areas from each soil area to be fertilized.

[0045] Obtain the same soil index (e.g., pH value) within the same area of ​​soil to be fertilized. Indicates the first A sequence of pH changes in a soil area to be fertilized at different times; The time series data of soil indicators of each dimension in each soil area to be fertilized were analyzed by DFA (detrended fluctuation analysis) algorithm to obtain the scaling index corresponding to each dimension.

[0046] The response buffer capacity of the soil area to be fertilized is calculated using the following formula:

[0047] in, Indicates the first The response buffer capacity of each soil area to be fertilized; The dimension number representing the soil index; For the first Under the soil area to be fertilized, the first The scaling index is obtained by analyzing the time series data of soil indicators in each dimension using the DFA algorithm. 8 represents the total number of dimensions of the soil indicators.

[0048] It should be noted that for summation and subtraction within formulas... In part, the absolute value of this difference characterizes the stability of the original sequence data. Specifically, for this algorithm, DFA is mainly used to detect long-range correlation and self-similarity of sequences. When the algorithm's output... When the sequence represents white noise (random and uncorrelated), the closer the difference is to 0, the higher the response of the soil regional data indicators to small sample fertilization results, exhibiting a sharp rise or fall in the data. Therefore, the lower the similarity between the data, the stronger the response of this type of soil to fertilization results, and the lower its buffer capacity. This soil is more likely to have been subjected to long-term chemical fertilizer application, thus increasing the necessity for further fertilization feedback control and making it more likely to be a sensitive soil region. For the summation part, the lower the difference across multiple dimensions, the lower the summation result, indicating that the soil multi-dimensional index data show a more significant and sensitive response to changes in fertilization results, and the lower its buffer capacity.

[0049] It should be noted that the response buffer size is divided when... Within the preset range threshold At that time, it was assumed that the soil lacked buffering capacity and was highly sensitive to the type and amount of fertilizer applied. Based on the distribution of the response buffering capacity of all soil areas to be fertilized in the experiment, combined with historical soil sample data showing negative effects after fertilization in this area, the lower quartile of the response buffering capacity was defined as the threshold range. The threshold range can be adjusted according to different soil types, climatic conditions, and crop varieties. If fertilization is improper, it may cause negative effects; such soils are considered sensitive soil areas.

[0050] Specifically, when the If the response buffer capacity of a soil area to be fertilized is within a preset threshold range, then the first soil area to be fertilized is determined to be within that range. The soil in the area to be fertilized lacks buffering capacity, which will lead to the first The soil areas to be fertilized are designated as sensitive soil areas.

[0051] Step S130: Using the time series data of historical multidimensional soil indicators of each sensitive soil area as input, construct a time series prediction model and output the prediction sequence of multidimensional soil indicators of each sensitive soil area; calculate the soil acidification rate of each sensitive soil area based on the prediction sequence of pH value of each sensitive soil area and the change sequence of pH value of each soil area to be fertilized at different times.

[0052] It should be noted that, using historical multi-dimensional soil index time-series data from various sensitive soil regions as input, a time-series prediction model based on a Long Short-Term Memory (LSTM) network is constructed to output predicted sequences of multi-dimensional soil indices for each sensitive soil region. The reason for this prediction is that soil regions themselves experience long cycles of fertilizer and nutrient absorption and changes; directly monitoring the entire cycle and applying targeted fertilization is impractical and costly. Therefore, to efficiently determine fertilization targets and adaptively adjust fertilization amounts, it is necessary to perform predictive processing on indices for certain special soil regions to obtain corresponding predicted indices.

[0053] For example, taking pH value as an example for analysis, in sensitive soil areas that have been fertilized with chemical fertilizers for a long time, the pH value is usually low and the acidification rate is fast. This is mainly due to the nitrification reaction, where ammonium nitrogen fertilizer is converted into nitrate nitrogen under the action of microorganisms, releasing a large amount of hydrogen ions (H+). + Meanwhile, to maintain charge balance, alkaline cations (such as Ca) in sensitive soil areas... 2+ Mg 2+ Significant amounts of alkaline substances are lost, depleting the soil's buffer capacity and making it extremely sensitive to changes in external pH levels. Conversely, organic fertilizers contain alkaline substances and can significantly increase the soil's pH buffering capacity. Furthermore, humus and other organic colloids contain numerous acidic functional groups that can directly neutralize H+. + Furthermore, increasing the soil's cation exchange capacity (CEC) can adsorb and store a large number of basic ions, effectively resisting acidification.

[0054] Therefore, the higher the acidification rate (i.e., the rate of increase) of the predicted data in a single dimension (such as pH value), the more sensitive the corresponding soil area response type is to the effect of fertilization, and the more necessary it is to reasonably control the type and amount of fertilizer to prevent the loss of essential soil ions caused by excessive fertilization, which could lead to a decrease in yield.

[0055] It should be noted that soil index parameters will exhibit the above-mentioned variation characteristics due to their inherent features; here, only pH is used as an example for analysis.

[0056] Specifically, taking pH value as an example, to obtain the first A predicted pH sequence for a sensitive soil region; the data trend and its increasing / decreasing effect of the predicted sequence are calculated using a trend strength from decomposition algorithm. The input is a time series containing trend and seasonality, i.e., the predicted sequence mentioned above; the output is a trend strength score between 0 and 1. When the score is close to 1, it indicates that the sequence has a strong trend, the rate of increase is fast, and the sequence changes are dominated; when the score is close to 0, it indicates that the sequence has almost no trend or a very weak trend.

[0057] Based on the predicted sequence and the above-mentioned... The pH value variation sequence of a soil area to be fertilized at different times was used to calculate the soil acidification rate of the sensitive soil area using the following formula:

[0058] in, Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first The mean of the predicted pH values ​​for a single sensitive soil region; Indicates the first Historical baseline pH mean values ​​for each sensitive soil region; Indicates the first The trend intensity score of the pH prediction sequence for a sensitive soil region is obtained by decomposing the trend component and residual component using a trend intensity algorithm based on time series decomposition. The ratio of the variance of the trend component to the sum of the trend component and the residual component is used as the trend intensity score.

[0059] It should be noted that, The larger the value, the closer it is to 1, indicating a strong trend in the predicted pH value sequence, a rapid increase, and that the sequence is dominated by changes in pH. This corresponds to the following logic: for sensitive soil areas with long-term chemical fertilizer application, the pH value is usually low, and the acidification rate is fast; therefore, The larger the value, the higher the acidification rate, and the higher the corresponding soil acidification rate. The formula uses the difference between the two values. Taking a portion as the main body, the significance of subtracting and taking the absolute value is: when the predicted sequence data values ​​show rapid acidification, then the pH value itself shows a decreasing effect, and its corresponding mean is relatively lower than the mean. The lower the (monitored actual pH value), the greater the difference, which better aligns with the logic of soil acidification, and consequently, the higher the soil acidification rate. Therefore, The higher the value, the higher the corresponding soil acidification rate, and the more necessary it is to adapt key fertilization parameters such as fertilization amount.

[0060] Step S140: Based on the correlation between soil indicators of various dimensions in each sensitive soil region, calculate the multi-indicator synergistic sensitivity of each sensitive soil region; calculate the soil response type score of the sensitive soil region according to the soil acidification rate and the multi-indicator synergistic sensitivity; calculate the optimized application amount of straw return and biochar according to the soil response type score and the preset original application amount, and realize adaptive fertilization.

[0061] It should be noted that sensitive soil areas are determined based on the response buffer amount; the sensitive soil areas are predicted and processed using response result samples to obtain a prediction set; the acidification rate of the single-dimensional index in the prediction set is analyzed, and combined with the synchronous sensitivity of the multi-dimensional index, the soil response type score is determined, and the application type and amount are controlled by feedback to achieve the purpose of adaptive precision fertilization.

[0062] It should be noted that soil index parameters are quite abundant, and different indicators exhibit corresponding changes after fertilization in the same type of soil, meaning there is a high degree of synchronous sensitivity among the index values. For example, in "sensitive" chemical fertilizer soils: due to weak buffering capacity, even a small dose of alkaline biochar can cause drastic pH fluctuations. Simultaneously, due to low CEC, imported nutrients may be rapidly lost due to poor adsorption capacity, leading to a rapid increase in EC. In contrast, in "robust" organic fertilizer soils: the strong buffering system absorbs and buffers such disturbances, resulting in a smoother pH change. High CEC effectively adsorbs and retains imported nutrients, and EC changes are also more moderate.

[0063] It should be noted that in the prediction sequence, the higher the synchronicity of the changes in the indicators among the multi-dimensional indicators, the higher the corresponding sensitivity. Combined with the acidification rate, the resulting soil response type is closer to the sensitive soil after fertilizer application. The specific application amount needs to be determined based on the soil response type score (the application type can be determined based on the soil type in the existing technology).

[0064] Specifically, the predicted sequences of multi-dimensional indicators for each sensitive soil region are obtained, and the synergistic sensitivity of the multi-indicator indicators for the sensitive soil region is calculated using the following formula:

[0065] in, Indicates the first Multi-indicator synergistic sensitivity of sensitive soil regions; , They represent the first The first sensitive soil area The and the first Time series data of soil indicators in multiple dimensions; Indicates the first The first sensitive soil area The and the first The Pearson correlation coefficients between soil indicators of various dimensions can be used to calculate the correlation values.

[0066] It should be noted that by summing the relevant values, the closer the value is to 1, the higher the correlation coefficient, and the higher the sensitivity of multi-indicator synergy, because the similarity between data of different indicator dimensions is high, which is consistent with the logic of synchronous change mentioned in the above steps.

[0067] Furthermore, based on the aforementioned soil acidification rate and the synergistic sensitivity of multiple indicators, the soil response type score was determined. The soil response type score for sensitive soil areas is calculated using the following formula:

[0068] in, Indicates the first Soil response type score for each sensitive soil region; Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first Multi-indicator synergistic sensitivity of sensitive soil regions; This represents the normalization function. It acts as a product coefficient, that is, the acidification rate of a single-dimensional indicator acts as a coefficient of a multi-dimensional indicator, guiding the change of the comprehensive score. As the product entity, the corresponding score value is determined by combining the changing characteristics of multi-dimensional indicators.

[0069] It should be noted that the above-mentioned... The closer the soil response type score of a sensitive soil area is to 1, the higher the sensitivity of the soil to the application of (straw return or biochar) after the application of chemical fertilizers. In order to avoid problems such as soil ion loss caused by excessive fertilization, optimization suggestions are made for its application amount.

[0070] Determine the type of material to be applied: When the pH of each sensitive soil area is less than the preset acidification threshold of 6.5 and the organic matter content is less than the preset infertility threshold of 2%, biochar should be applied first to utilize its alkaline properties and structural improvement function to simultaneously alleviate acidification and infertility problems; otherwise, straw should be returned to the field first to provide organic matter and nutrients.

[0071] Specifically, when the applied material is straw returned to the field, the optimized application rate of straw returned to the field is calculated using the following formula:

[0072] in, Indicates the first Optimized application rate of straw return to the field in sensitive soil areas; This indicates the original application rate of straw returned to the field; Indicates the first The soil response type score for each sensitive soil area; the higher the score, the lower the recommended application rate, ensuring that soil ions remain within the normal range after application and preventing yield reduction due to over-application. When the value is 1, If the soil response to the material is deemed to have reached a critical sensitivity state, the application of the material should be stopped.

[0073] When the applied material is biochar, the optimized application rate of biochar is calculated using the following formula:

[0074] in, Indicates the first Optimized biochar application rates for sensitive soil areas; This indicates the original application rate of biochar. Indicates the first Soil response type score for each sensitive soil region; The biochar response reduction factor, ranging from 0 to 1, is used to characterize the difference in soil response sensitivity between biochar and straw return.

[0075] It should be noted that the optimized application rate achieves the goal of adaptive, precise, closed-loop fertilization type and application rate selection.

[0076] Based on the optimized application rates of straw return to the field and biochar, the corresponding materials are applied to the corresponding sensitive soil areas. After the application of materials is completed, the multi-dimensional indicators of the sensitive soil areas are continuously monitored. The parameters of the time series prediction model are updated using the time series data of the monitored soil indicators, and the material application rate of the next cycle is iteratively optimized to achieve closed-loop adaptive fertilization of straw return to the field and biochar.

[0077] For example, the complete data of sensitive test plot 2: Test plot 2 is the soil area to be fertilized. Time series data of 8-dimensional soil indicators were collected, and the scaling index of each dimension was obtained through DFA algorithm analysis. The response buffer was calculated to be 0.4025 according to the formula. 0.4025 is greater than the preset range threshold, and it is determined to be a sensitive soil area. The predicted pH value sequence of the sensitive soil area was obtained by prediction, and the soil acidification rate of the sensitive soil area was calculated. Combined with the synergistic sensitivity of multiple indicators, the soil response type score was calculated to be 0.477. The pH of the sensitive soil area is 5.70 and the organic matter is 1.62%. It meets the requirements of pH < 6.5 and organic matter < 2%, so biochar is preferred. The baseline application rate of biochar was set at 3000 kg / hm. 2 Biochar response reduction factor =0.6, the optimized application rate is calculated to be 2141.4 kg / hm. 2 Compared with the preliminary test, a fixed 20% baseline dosage (600 kg / hm) was used. 2 The control group was compared with the control group. After two growth cycles of monitoring, soil acidification continued to develop in the fixed dose group, while the optimized application group of this method effectively curbed the trend of soil acidification and significantly improved the soil buffering capacity, achieving the effect of closed-loop adaptive regulation.

[0078] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of the invention, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the invention. The sequence numbers of the above-described embodiments of the invention are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0079] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0080] In the several embodiments provided by this invention, it should be understood that the disclosed methods can be implemented in other ways. The methods disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments. The features disclosed in the several method embodiments provided by this invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0081] The above description is merely an embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A closed-loop adaptive control method for straw return to the field and biochar application, characterized in that, include: Configure a pre-test, dividing the pre-test area into several test cells; At multiple preset time points, time series data of multi-dimensional soil indicators for each soil area to be fertilized in each experimental plot are collected for each test. The multi-dimensional soil indicators include at least pH value. The DFA algorithm was used to analyze the time series data of soil indicators of each dimension in each soil area to be fertilized to obtain the scaling index corresponding to each dimension, and the response buffer of each soil area to be fertilized was calculated based on the scaling index. The response buffer was compared with a preset range threshold to screen out sensitive soil areas from each soil area to be fertilized. Using historical multidimensional soil index time series data of each sensitive soil region as input, a time series prediction model is constructed to output the prediction sequence of multidimensional soil index of each sensitive soil region; based on the prediction sequence of pH value of each sensitive soil region and the change sequence of pH value of each soil region to be fertilized at different times, the soil acidification rate of each sensitive soil region is calculated. Based on the correlation between various soil indicators in each sensitive soil region, the synergistic sensitivity of multiple indicators in each sensitive soil region is calculated; based on the soil acidification rate and the synergistic sensitivity of multiple indicators, the soil response type score of the sensitive soil region is calculated; based on the soil response type score and the preset original application rate, the optimized application rates of straw return and biochar are calculated, and adaptive fertilization is achieved.

2. The method according to claim 1, characterized in that, The calculation of the response buffer for each soil region to be fertilized, based on each of the scaling indices, includes: The response buffer capacity of each soil zone to be fertilized is calculated using the following formula: in, Indicates the first The response buffer capacity of each soil area to be fertilized; The dimension number representing the soil index; For the first Under the soil area to be fertilized, the first Scaling indices were obtained by analyzing time-series data of soil indicators across multiple dimensions using the DFA algorithm.

3. The method according to claim 2, characterized in that, The step of comparing the response buffer amount with a preset range threshold to screen out sensitive soil areas from each soil area to be fertilized includes: When the If the response buffer amount of a soil area to be fertilized is within a preset threshold range, then the first... The soil in the area to be fertilized does not have buffering capacity, so the first... The soil areas to be fertilized are designated as sensitive soil areas.

4. The method according to claim 1, characterized in that, The calculation of soil acidification rate for each sensitive soil region, based on the predicted pH value sequence for each sensitive soil region and the pH value change sequence for each soil region to be fertilized at different times, includes: The soil acidification rate for each sensitive soil region is calculated using the following formula: in, Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first The mean of the predicted pH values ​​for a single sensitive soil region; Indicates the first Historical baseline pH mean values ​​for each sensitive soil region; Indicates the first The trend intensity score of the pH prediction sequence for a sensitive soil region is obtained by decomposing the trend component and the residual component using a trend intensity algorithm for time series decomposition. The trend intensity score is then calculated based on the variance of the trend component and the variance of the residual component.

5. The method according to claim 1, characterized in that, The calculation of multi-indicator synergistic sensitivity of each sensitive soil region based on the correlation between various dimensions of soil indicators in each sensitive soil region includes: The synergistic sensitivity of multiple indicators in sensitive soil regions is calculated using the following formula: in, Indicates the first Multi-indicator synergistic sensitivity of sensitive soil regions; , They represent the first The first sensitive soil area The and the first Time series data of soil indicators in multiple dimensions; Indicates the first The first sensitive soil area The and the first Pearson correlation coefficients among soil indicators of various dimensions.

6. The method according to claim 1, characterized in that, The step of calculating the soil response type score of the sensitive soil area based on the soil acidification rate and the synergistic sensitivity of multiple indicators includes: The soil response type score for sensitive soil areas is calculated using the following formula: in, Indicates the first Soil response type score for each sensitive soil region; Indicates the first Soil acidification rate in sensitive soil areas; Indicates the first Multi-indicator synergistic sensitivity of sensitive soil regions; This represents the normalization function.

7. The method according to claim 1, characterized in that, The step of calculating the optimized application rates of straw return and biochar based on the soil response type score and the preset original application rate includes: When the applied material is straw returned to the field, the optimized application rate of straw returned to the field is calculated using the following formula: in, Indicates the first Optimized application rate of straw return to the field in sensitive soil areas; This indicates the original application rate of straw returned to the field; Indicates the first Soil response type score for each sensitive soil region; When the applied material is biochar, the optimized application rate of biochar is calculated using the following formula: in, Indicates the first Optimized biochar application rates for sensitive soil areas; This indicates the initial application rate of biochar. Indicates the first Soil response type score for each sensitive soil region; The biochar response reduction factor is used to characterize the difference in soil response sensitivity between biochar and straw return.

8. The method according to claim 7, characterized in that, The implementation of adaptive fertilization includes: Based on the optimized application rates of straw return to the field and biochar, the corresponding materials are applied to the corresponding sensitive soil areas. After the application of materials is completed, the multi-dimensional indicators of the sensitive soil areas are continuously monitored. The parameters of the time-series prediction model are updated using the time series data of the monitored soil indicators. The material application rate for the next cycle is iteratively optimized to achieve closed-loop adaptive fertilization of straw return to the field and biochar.

Images