A soil environment intelligent monitoring system for fritillaria thunbergii miq planting
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIVERSITY OF TRADITIONAL CHINESE MEDICINE (CHUNAN QIANDAO LAKE) RESEARCH INSTITUTE CO LTD
- Filing Date
- 2026-04-09
- Publication Date
- 2026-07-03
Smart Images
Figure CN121995037B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of soil environmental monitoring technology, specifically to an intelligent soil environmental monitoring system for the cultivation of Fritillaria thunbergii. Background Technology
[0002] Fritillaria thunbergii, an important traditional Chinese medicine, is highly dependent on soil environmental conditions for its growth and development. During continuous cultivation, the roots of Fritillaria thunbergii secrete autotoxic substances such as phenolic acids. The accumulation of these substances in the soil inhibits the growth of subsequent plants, leading to continuous cropping obstacles and seriously affecting the yield and quality of the medicinal material.
[0003] In related technologies, sampling points are often set up, and samples are taken and analyzed regularly to monitor the soil environment based on the content of autotoxic substances in the samples.
[0004] However, tillage operations such as turning and leveling during continuous cropping can disrupt the spatial distribution of the original autotoxic substances in the soil. The analysis results obtained from sampling at the sampling points are difficult to accurately reflect the accumulation of autotoxic substances in the soil caused by continuous cropping, resulting in low accuracy in monitoring and assessing the soil environment. Summary of the Invention
[0005] To address the technical problem that analytical results obtained from sampling at designated sampling points cannot accurately reflect the accumulation of autotoxic substances in the soil caused by continuous cropping, thus resulting in low accuracy in soil environmental monitoring and assessment, this application aims to provide an intelligent soil environmental monitoring system for Fritillaria thunbergii cultivation. The specific technical solution adopted is as follows:
[0006] This application provides an intelligent soil environment monitoring system for Fritillaria thunbergii cultivation. The system includes a data acquisition unit, a data processing unit, a data prediction unit, and a data monitoring unit. The data acquisition unit is used to collect autotoxic substance content from multiple sampling points at a preset sampling period, obtaining the autotoxic substance content of each sampling point at multiple sampling times. The data processing unit is used to determine historical continuous cropping rows based on the changes in autotoxic substance content at each sampling point at multiple sampling times and the location distribution of multiple sampling points, and to identify the sampling points included in the historical continuous cropping rows as continuous cropping points. The data processing unit is also used to determine the cumulative baseline value of continuous cropping and the autotoxic substance content generated by each sampling point within the current sampling period based on the autotoxic substance content of each continuous cropping point at the current sampling time. The data prediction unit is used to input the autotoxic substance content generated by each sampling point within the current sampling period and the cumulative baseline value of continuous cropping into a content prediction model to obtain a predicted value of the autotoxic substance content of each sampling point at the next sampling time. This content prediction model is used to predict the autotoxic substance content. The data monitoring unit is used to monitor the soil environment based on the predicted value of the autotoxic substance content of each sampling point at the next sampling time.
[0007] Optionally, the data processing unit is specifically used for: determining the rate of change of autotoxic substance content at each sampling point based on the autotoxic substance content at multiple sampling times; determining the degree of influence at each sampling point based on the rate of change of autotoxic substance content at each sampling point, wherein the degree of influence is used to characterize the extent to which a sampling point is affected by the current planting; determining the distribution regularity of autotoxic substances in each candidate row based on the degree of influence of the sampling points included in each of the N candidate rows and the autotoxic substance content at each sampling point, where N is an integer greater than or equal to 1; and determining the candidate row with the largest distribution regularity of autotoxic substances as the historical continuous cropping row.
[0008] Optionally, the data processing unit is specifically configured to: determine the maximum and minimum rates of change of autotoxic substance content at sampling points within a first local region, wherein the first local region is the local region where the first sampling point is located, and the first sampling point is one of the plurality of sampling points; and determine the ratio between the first difference and the second difference as the degree of influence of the first sampling point, wherein the first difference is the difference between the rate of change of autotoxic substance content at the first sampling point and the minimum rate of change of autotoxic substance content, and the second difference is the difference between the maximum rate of change of autotoxic substance content and the rate of change of autotoxic substance content at the first sampling point.
[0009] Optionally, the data processing unit is specifically used to: determine M current planting rows based on the positions of multiple sampling points, where M is an integer greater than or equal to 1; rotate the M current planting rows by at least one angle and / or translate them by at least one preset distance to obtain N candidate rows, where N≥M.
[0010] Optionally, the data processing unit is specifically used to: determine the historical continuous cropping reference value of each sampling point based on the degree of influence of the sampling points included in the first candidate row and the content of autotoxic substances at each sampling point, wherein the first candidate row is any one of the N candidate rows; determine the difference in historical continuous cropping reference value between every two adjacent sampling points in the first candidate row; and determine the distribution regularity of autotoxic substances in the first candidate row based on the difference in historical continuous cropping reference value between every two adjacent sampling points in the first candidate row.
[0011] Optionally, the data processing unit is specifically used to: determine the distance between the first sampling point and the root sampling point in the first local area, wherein the root sampling point in the first local area is the sampling point corresponding to the maximum rate of change of autotoxic substance content in the first local area; determine the continuous cropping representativeness of the first sampling point based on the distance and the degree of influence of the first sampling point; and determine the historical continuous cropping reference of the first sampling point based on the continuous cropping representativeness of the first sampling point and the autotoxic substance content of the first sampling point.
[0012] Optionally, the data processing unit is specifically used to: determine the cumulative benchmark value of continuous cropping based on the distribution regularity of autotoxic substances in historical continuous cropping rows, the content of autotoxic substances at all continuous cropping points in historical continuous cropping rows at the current sampling time, and the continuous cropping representativeness of each continuous cropping point; and determine the content of autotoxic substances generated by each sampling point in the current sampling period by the difference between the content of autotoxic substances at each sampling point at the current sampling time and the cumulative benchmark value of continuous cropping.
[0013] Optionally, the system also includes a model building unit, which is used to: generate multiple autotoxic substance content sequences based on the autotoxic substance content of each sampling point at multiple collection times, with one sampling point corresponding to one autotoxic substance content sequence; perform least squares fitting on the multiple autotoxic substance content sequences to obtain initial parameters, which are used to characterize the overall trend of autotoxic substance content changes; and build a content prediction model based on the initial parameters.
[0014] Optionally, the data acquisition unit is further configured to: obtain the actual value of the autotoxic substance content at each sampling point at the next sampling time; the model building unit is further configured to: adjust the initial parameters in the content prediction model based on the actual value of the autotoxic substance content at each sampling point at the next sampling time and the predicted value of the autotoxic substance content at the next sampling time.
[0015] Optionally, the data monitoring unit is specifically used to issue an alarm message when the predicted value of the self-toxic substance content at the first sampling point exceeds a preset risk threshold.
[0016] This application has the following beneficial effects:
[0017] In this embodiment, the intelligent soil environment monitoring system for Fritillaria thunbergii cultivation ensures the timeliness and coverage of data by collecting autotoxic substance content from multiple sampling points and at multiple sampling times. Then, based on the changes in autotoxic substance content at multiple times, it identifies historically continuously cropped rows where autotoxic substance accumulation is severe but less affected by the current planting cycle. Based on the autotoxic substance content of sampling points in these historically continuously cropped rows, it determines the historical cumulative baseline value and separates the autotoxic substance content of the current sampling cycle from the historical cumulative baseline value, improving the accuracy of detecting autotoxic substance accumulation caused by continuous cropping. Thus, based on the autotoxic substance content of the current sampling cycle and the historical cumulative baseline value, it inputs the content prediction model to predict the autotoxic substance content at future times, resulting in more accurate predictions and improving the accuracy and foresight of soil monitoring. Attached Figure Description
[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a structural diagram of an intelligent soil environment monitoring system for Fritillaria thunbergii cultivation, provided in one embodiment of this application.
[0020] Figure 2 This is a flowchart of a method for intelligent monitoring of soil environment for Fritillaria thunbergii cultivation, provided in one embodiment of this application.
[0021] Figure 3 This is a structural diagram of another intelligent soil environment monitoring system for Fritillaria thunbergii cultivation provided in one embodiment of this application. Detailed Implementation
[0022] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent soil environment monitoring system for Fritillaria thunbergii cultivation proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0024] Fritillaria thunbergii is a traditional Chinese medicine with medicinal effects such as clearing heat and resolving phlegm, relieving cough and asthma. Its growth and development are extremely sensitive to soil environmental conditions. Changes in factors such as temperature, humidity, pH, nutrient content, and aeration directly affect bulb formation and the accumulation of effective components. Currently, the empirical management method for cultivating Fritillaria thunbergii relies on manual, periodic sampling and testing of soil indicators. This data is lagging and has limited accuracy, making it difficult to achieve dynamic monitoring and scientific control of the soil environment, resulting in significant fluctuations in yield and quality. Currently, an intelligent soil environment monitoring system has been developed by introducing Internet of Things (IoT) and wireless sensor network technologies. This system consists of multiple sensors, data acquisition modules, communication networks, edge computing units, and a cloud analysis platform. It collects and analyzes parameters such as soil temperature, humidity, pH, conductivity, nutrient content, and organic matter content in real time to assess whether the planting environment is suitable for the growth of Fritillaria thunbergii.
[0025] In existing technologies, during continuous cultivation of Fritillaria thunbergii, its roots secrete large amounts of phenolic acid autotoxic substances, which accumulate in the soil with increasing cropping frequency. This significantly inhibits root growth and seed germination in subsequent plants, exacerbates disease occurrence, and leads to sharp yield reduction and quality decline. When monitoring autotoxic substances under continuous cropping conditions, the accumulation of these phenolic acids varies between planting rows. Furthermore, tillage operations such as turning and leveling during continuous cropping disrupt these spatial differences and blur accumulation boundaries, masking the uneven distribution of autotoxic substances in the soil and making accurate identification and quantitative assessment of the soil environment for cultivation difficult. Therefore, this invention establishes an estimation model for autotoxic substances produced during current Fritillaria thunbergii cultivation and accumulated through continuous cropping, enabling accurate monitoring of the soil environment for Fritillaria thunbergii cultivation.
[0026] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent soil environment monitoring system for Fritillaria thunbergii cultivation provided in this application.
[0027] Please see Figure 1 The diagram shows a structural diagram of an intelligent soil environment monitoring system for Fritillaria thunbergii cultivation, provided in one embodiment of this application.
[0028] like Figure 1 As shown, the intelligent soil environment monitoring system 10 for Fritillaria thunbergii cultivation includes a data acquisition unit 101, a data processing unit 102, a data prediction unit 103, and a data monitoring unit 104.
[0029] The data acquisition unit 101 is used to collect the content of autotoxic substances at multiple sampling points with a preset sampling period length, so as to obtain the content of autotoxic substances at each sampling point at multiple sampling times.
[0030] Optionally, the preset sampling period can be 3 days, meaning that sampling is performed once every 3 days.
[0031] Optionally, key sampling times can be determined based on the growth cycle of Fritillaria thunbergii, such as before and after seedling emergence, during bulb enlargement (high release of autotoxin), before and after harvest, and after tilling and preparing the land.
[0032] Optionally, the sampling frequency can be increased during the growth cycle of autotoxic surge (i.e., the bulb enlargement period).
[0033] In one alternative implementation, sampling points can be set up at different locations in each experimental field, such as rows, between rows, and edge areas. The sampling points must cover at least the midpoint of each row, the midpoint of the space between rows, the end of each row, and corners. During sampling, a stainless steel sampler is used to collect soil samples to avoid metal contamination. An equal amount of soil is taken from each sampling point and mixed to create a composite sample to reduce local noise. After sampling, the samples are immediately placed in a clean, sealed bag, labeled with the number and sampling time, refrigerated or stored at low temperature, and sent for testing as soon as possible.
[0034] In this embodiment of the application, the content of the autotoxic substance is the concentration of phenolic acids or alkaloids in the soil.
[0035] Optionally, liquid chromatography-mass spectrometry (LC-MS / MS) can be used to quantify phenolic acids and alkaloids depending on the type of autotoxic substance.
[0036] Optionally, after obtaining the content of autotoxic substances at each sampling point at multiple sampling times, outliers can be identified and removed.
[0037] The data processing unit 102 is used to determine historical continuous cropping rows based on the changes in the content of autotoxic substances at each sampling point at multiple sampling times and the location distribution of multiple sampling points, and to determine the sampling points included in the historical continuous cropping rows as continuous cropping points.
[0038] It should be understood that historical continuous cropping behavior refers to the distribution of Fritillaria thunbergii planted in the current sampling planting area during historical periods.
[0039] It should be understood that during historical continuous cropping, autotoxic substances are concentrated in the historically cropped rows, and the content of autotoxic substances caused by this historical continuous cropping does not change. During the current planting process, autotoxic substances will be concentrated in the current planting rows, and the content of autotoxic substances will gradually increase as the Fritillaria thunbergii grows. Therefore, the historically cropped row can be determined based on the changes in the content of autotoxic substances at each sampling point at multiple sampling times.
[0040] Specifically, the physiological metabolic intensity of Fritillaria thunbergii varies during its emergence, bulb enlargement, and maturity stages. Consequently, the types and amounts of phenolic acids, steroidal alkaloids, and other secondary metabolites secreted by the roots also change. Some of these metabolites exhibit significant autotoxic effects, inhibiting root vigor, affecting soil microbial community structure, and even hindering the germination and growth of subsequent plants. Relying solely on single-time-point content measurements makes it difficult to determine whether the autotoxic substances are newly released during the current sampling period or accumulated from previous seasons. Therefore, analyzing the changes in autotoxic substance content at different sampling times allows us to capture the dynamic trends of autotoxic substance release and migration.
[0041] It should be understood that the current sampling period is the sampling period between the previous sampling time and the current sampling time.
[0042] It is understandable that when the content of autotoxic substances at a certain sampling point changes significantly, it indicates that the autotoxic substances secreted by the current root system at that sampling point are more abundant; when the content of autotoxic substances at a certain sampling point changes little, it indicates that the autotoxic substances at that sampling point are mainly composed of autotoxic substances accumulated from continuous cropping.
[0043] Optionally, multiple rows can be divided based on the locations of the multiple sampling points, and then historical continuous cropping rows can be determined based on the changes in the content of autotoxic substances at the sampling points in each row.
[0044] The data processing unit 102 is also used to determine the cumulative baseline value of continuous cropping and the content of autotoxic substances generated by each sampling point in the current sampling period based on the content of autotoxic substances at each continuous cropping point at the current sampling time.
[0045] Understandably, the continuous cropping baseline is used to quantify the basic level of autotoxic substances in the soil accumulated through historical continuous cropping processes (i.e., all historical accumulation prior to the last sampling time). This continuous cropping baseline reflects the residual autotoxic substances in the soil caused by long-term continuous cropping.
[0046] It should be understood that a planting cycle refers to the complete growth stage of Fritillaria thunbergii from planting (e.g., seedling stage) to harvest. The sampling period is a preset duration, such as three days, and a planting cycle includes multiple sampling periods. This cumulative baseline value for continuous cropping is not directly affected by the autotoxic substances secreted by the roots in the current planting cycle. Based on this cumulative baseline value for continuous cropping, it is possible to distinguish between the autotoxic substance content accumulated from historical continuous cropping and the newly added autotoxic substance content in the current sampling cycle at a sampling point.
[0047] It is understandable that, since historical continuous cropping rows are greatly affected by historical continuous cropping, the content of autotoxic substances at the continuous cropping points in these historical continuous cropping rows is mostly caused by historical continuous cropping, and a small part is autotoxic substances generated before the previous sampling time in the current planting cycle. The cumulative baseline value of continuous cropping can be determined based on the content of autotoxic substances at these continuous cropping points.
[0048] In one alternative implementation, the mean value of autotoxic substance content at all points in the historical continuous cropping row can be determined as the cumulative baseline value for continuous cropping. Then, the difference between the autotoxic substance content at each sampling point and the cumulative baseline value for continuous cropping can be determined as the autotoxic substance content generated by each sampling point in the current sampling period.
[0049] The data prediction unit 103 is used to input the content of autotoxic substances generated by each sampling point in the current sampling period and the cumulative baseline value of continuous cropping into the content prediction model to obtain the predicted value of autotoxic substance content of each sampling point at the next sampling time.
[0050] The content prediction model is used to predict the content of autotoxic substances.
[0051] It should be understood that the next sampling time is the sampling time after one sampling cycle from the current sampling time. The predicted value of autotoxic substance content at the next sampling time is actually the predicted value of autotoxic substance content in the soil at one sampling point after one sampling cycle.
[0052] Specifically, the content prediction model can simulate the growth trend and growth rate of autotoxic substance content, predict the growth amount of a sampling point in the next sampling period, and based on the growth amount, predict the autotoxic substance content at the next sampling time.
[0053] The data monitoring unit 104 is used to monitor the soil environment based on the predicted content of autotoxic substances at each sampling point at the next sampling time.
[0054] Optionally, a soil autotoxicity risk distribution map can be constructed based on the predicted values of autotoxic substance content at all sampling points. This distribution map presents the degree of autotoxicity at different locations within the planting area in a visual manner.
[0055] In one alternative implementation, the data monitoring unit 104 is specifically used to issue an alarm message when the predicted value of the self-toxic substance content at the first sampling point exceeds a preset risk threshold.
[0056] Optionally, the preset risk threshold can be set according to the requirements of the soil, or different preset risk thresholds can be set according to different growth stages of Fritillaria thunbergii.
[0057] For example, under normal circumstances, the preset risk threshold can be set to 50 mg / kg, and during the bulb enlargement period, the preset risk threshold can be reduced to 30 mg / kg.
[0058] It should be understood that this alarm message is used to prompt managers to carry out agronomic interventions such as soil remediation, crop rotation adjustments, or precision fertilization.
[0059] Optionally, the intelligent soil environment monitoring system 10 for Fritillaria thunbergii cultivation may also include a human-computer interaction interface, on which the alarm information is displayed.
[0060] In this embodiment, the intelligent soil environment monitoring system for Fritillaria thunbergii cultivation ensures the timeliness and coverage of data by collecting autotoxic substance content from multiple sampling points and at multiple sampling times. Then, based on the changes in autotoxic substance content at multiple times, it identifies historically continuously cropped rows where autotoxic substance accumulation is severe but less affected by the current planting cycle. Based on the autotoxic substance content of sampling points in these historically continuously cropped rows, it determines the historical cumulative baseline value and separates the autotoxic substance content of the current sampling cycle from the historical cumulative baseline value, improving the accuracy of detecting autotoxic substance accumulation caused by continuous cropping. Thus, based on the autotoxic substance content of the current sampling cycle and the historical cumulative baseline value, it inputs the content prediction model to predict the autotoxic substance content at future times, resulting in more accurate predictions and improving the accuracy and foresight of soil monitoring.
[0061] Combination Figure 1 ,like Figure 2 As shown, in one implementation of this application embodiment, when the data processing unit 102 is used to identify historical continuous cropping rows from the spatial distribution of multiple sampling points based on the changes in the content of autotoxic substances at each sampling point at multiple sampling times, it is specifically used to execute the intelligent monitoring method for soil environment for Fritillaria thunbergii cultivation provided in S201-S204 below.
[0062] S201. Based on the content of autotoxic substances at each sampling point at multiple sampling times, determine the rate of change of autotoxic substance content at each sampling point.
[0063] Optionally, the rate of change of the autotoxic substance content can be calculated based on the autotoxic substance content at all sampling times. If the number of sampling times is too large, the rate of change of the autotoxic substance content can also be calculated based on the autotoxic substance content at the current sampling time and the previous 5 sampling times, i.e., 6 sampling times.
[0064] Optionally, the rate of change of the autotoxic substance content at a sampling point satisfies the following formula:
[0065]
[0066] in, Indicates sampling point The rate of change in the content of autotoxic substances, Indicates the number of sampling times. Indicates the number of adjacent sampling times. Indicates sampling point In the The content of autotoxic substances at each sampling time. Indicates sampling point In the The content of autotoxic substances at each sampling time. express The length of time represented by each sampling moment express The absolute value of.
[0067] It should be understood that The time length represented by each sampling moment is The duration of each sampling period, for example, assuming The value is 6, and a sampling period is 3 days. It lasts for 18 days.
[0068] In this formula, It represents the absolute value of the difference in the content of autotoxic substances between adjacent sampling times. It is used to quantify the instantaneous fluctuation range of the content of autotoxic substances. Taking the absolute value can avoid the cancellation of positive and negative values, which would lead to the underestimation of the fluctuation range. This represents the total fluctuation amplitude over multiple sampling periods; It represents the mean of the absolute values of the differences in the content of toxic substances between adjacent sampling times, and characterizes the average fluctuation range; This method is used to standardize the average fluctuation range into a rate of change per unit time, ensuring that the rate of change of autotoxic substance content is comparable over time, and allowing the rate of change of autotoxic substance content at different sampling points to be compared under the same time benchmark.
[0069] S202. Based on the rate of change of the content of autotoxic substances at each sampling point, determine the degree of impact on each sampling point.
[0070] The degree of impact is used to characterize the extent to which a sampling point is affected by the current planting, specifically the impact of the autotoxic substance content of a sampling point on the autotoxic substance content generated during the current planting cycle.
[0071] It should be understood that the greater the rate of change in autotoxic substances at a sampling point, the greater the influence of autotoxic substances secreted by Fritillaria thunbergii during the current planting cycle on that sampling point. Conversely, the smaller the rate of change in autotoxic substances at a sampling point, the greater the cumulative impact of historical continuous cropping. The greater the influence on a sampling point, the closer its autotoxic substance content variation characteristics are to the current active root secretion area; the smaller the influence on a sampling point, the more stable its autotoxic substance content variation is, and the more likely it is to reflect the cumulative characteristics of historical continuous cropping.
[0072] In one alternative implementation, the rate of change of autotoxic substance content at multiple sampling points can be subjected to maximum and minimum normalization, and then the normalized value can be used to determine the degree of influence of each sampling point.
[0073] In another alternative implementation, the maximum and minimum rates of change of autotoxic substance content at sampling points within the first local region can be determined; the ratio between the first difference and the second difference is determined as the degree of influence of the first sampling point.
[0074] Wherein, the first local area is the local area where the first sampling point is located, the first sampling point is one of the multiple sampling points, the first difference is the difference between the change rate of the autotoxic substance content of the first sampling point and the minimum change rate of the autotoxic substance content, and the second difference is the difference between the maximum change rate of the autotoxic substance content and the change rate of the autotoxic substance content of the first sampling point.
[0075] Optionally, the distance between five adjacent sampling points on a straight line can be defined as a preset radius, and a circular region centered on the first sampling point with a preset radius can be defined as the first local region.
[0076] Optionally, the rate of change of autotoxic substance content at each sampling point in the first local region can be determined, and then the extreme values (including the maximum and minimum rates of change of autotoxic substance content) can be determined.
[0077] It should be understood that the first difference represents the amount by which the rate of change of autotoxic substance content at the first sampling point exceeds the minimum value in the neighborhood (i.e., the minimum rate of change of autotoxic substance content in the first local region). The larger the first difference, the more significantly the rate of change of autotoxic substance content at the first sampling point is higher than that at the least active point in the neighborhood, and the more obvious the influence of the current planting. The second difference represents the difference between the maximum value in the neighborhood (i.e., the maximum rate of change of autotoxic substance content in the first local region) and the rate of change of autotoxic substance content at the first sampling point. The larger the second difference, the more the rate of change of autotoxic substance content at the first sampling point deviates from the maximum value in the neighborhood, and the less it is affected by the current planting.
[0078] Optionally, the degree of influence of a sampling point satisfies the following formula:
[0079]
[0080] in, Indicates sampling point The extent of the impact, Indicates sampling point The rate of change in the content of autotoxic substances, Indicates sampling point The minimum rate of change in the content of autotoxic substances within the local area. Indicates sampling point The maximum rate of change in the content of autotoxic substances within the local area.
[0081] In this formula, Indicates the first difference. This represents the second difference.
[0082] Optionally, when calculating the degree of influence of a sampling point based on this formula, when When the value is 0, The value is set to 1.
[0083] Understandably, using the extreme value of the rate of change within the spatial neighborhood as a reference benchmark can effectively eliminate environmental noise interference and accurately reflect the differences in the impact of current root secretion activity on sampling points at different locations.
[0084] S203. Based on the degree of influence of the sampling points included in each of the N candidate rows and the content of autotoxic substances in each sampling point, determine the distribution regularity of autotoxic substances in each candidate row.
[0085] Where N is an integer greater than or equal to 1.
[0086] It should be understood that each candidate row represents a possible historical row direction.
[0087] In one alternative implementation, M current planting rows are determined based on the positions of multiple sampling points, and the M current planting rows are rotated by at least one angle and / or translated by at least one preset distance to obtain N candidate rows.
[0088] Where M is an integer greater than or equal to 1, and N≥M.
[0089] It should be understood that the M planting behaviors are the rows defined in the current planting distribution.
[0090] Optionally, the spatial coordinate data of the multiple sampling points can be obtained, and then the sampling points can be grouped based on a spatial clustering algorithm (such as DBSCAN or K-means clustering algorithm) to identify the linear regions with dense spatial distribution. Each linear region corresponds to a current planting row, and the row direction and row spacing of the current planting row are recorded.
[0091] Optionally, based on each current planting behavior, the N candidate rows can be obtained by rotating clockwise multiple times at preset angle intervals, with the midpoint of the current planting row as the center.
[0092] For example, the preset angle interval can be 5°, and the number of rotations can be 5.
[0093] Optionally, the N candidate rows can be obtained by translating multiple times at a preset distance on both sides perpendicular to the current planting row direction.
[0094] For example, the preset distance can be 10cm, and the number of translations can be 3.
[0095] Alternatively, rotation and translation operations can be combined, for example, rotation followed by translation, to generate the N candidate rows.
[0096] Alternatively, the M planting rows can be directly designated as the N candidate rows.
[0097] In one alternative implementation, based on the degree of impact of the sampling points included in the first candidate row and the content of autotoxic substances at each sampling point, the historical continuous cropping reference value of each sampling point is determined, and the difference in historical continuous cropping reference value between every two adjacent sampling points in the first candidate row is determined; based on the difference in historical continuous cropping reference value between every two adjacent sampling points in the first candidate row, the distribution regularity of autotoxic substances in the first candidate row is determined.
[0098] The first candidate row is any one of the N candidate rows.
[0099] It should be understood that the historical continuous cropping reference is used to characterize the degree to which the sampling point can be referenced as a continuous cropping point. The greater the historical continuous cropping reference, the greater the probability that the point is a continuous cropping point.
[0100] Optionally, the distance between the first sampling point and the root sampling point within the first local area is determined; based on the distance and the degree of influence of the first sampling point, the continuous cropping representativeness of the first sampling point is determined; based on the continuous cropping representativeness of the first sampling point and the content of autotoxic substances at the first sampling point, the historical continuous cropping reference value of the first sampling point is determined.
[0101] Specifically, the root sampling points within this first local area are those corresponding to the maximum rate of change in autotoxic substance content within that area. These sampling points, exhibiting the highest rate of change in autotoxic substance content, are highly likely to be the root sampling points currently being planted, and can be understood as the center of the current planting activity, i.e., the hotspot area.
[0102] It should be understood that the representativeness of continuous cropping is used to identify sampling points that can represent the historical continuous cropping accumulation (rather than the current planting). Since historical continuous cropping rows are often parallel or close to the current planting rows, the farther the distance between the first sampling point and the root sampling points in the first local area, the more likely the first sampling point is to deviate from the historical continuous cropping row, the more marginal its position is, and the less representative it is. The closer the distance between the first sampling point and the root sampling points in the first local area, the closer the first sampling point is to the hot spot area, and the stronger the representativeness of the first sampling point.
[0103] Understandably, the lower the degree of influence of a sampling point, the weaker the contribution of the current planting cycle, and the more likely the content of autotoxic substances at that sampling point is to represent historical accumulation, thus the higher the representativeness of continuous cropping at that sampling point.
[0104] Optionally, the representativeness of a sampling point satisfies the following formula:
[0105]
[0106] in, Indicates sampling point Representative of continuous work, Indicates sampling point The extent of the impact, Indicates sampling point The distance between the root sampling points in its local area and the sampling points therein.
[0107] Optionally, when calculating the continuous representativeness of a sampling point based on this formula, when or When the value is 0, it means that the sampling point is not affected by the current planting cycle or that the sampling point is the root sampling point. This indicates that the sampling point is very likely to overlap with historical continuous cropping points. In this case, it can be... The value is set to 1.
[0108] Optionally, the degree of influence and distance can be normalized separately, for example, by max-min normalization, mapping the values to [0, 1], and then the formula can be calculated.
[0109] Based on this formula, it should be understood that the less affected a sampling point is and the closer it is to the root sampling point, the closer the sampling point is to the historical accumulation zone, and the higher the representativeness of continuous cropping.
[0110] In this embodiment, the representativeness of continuous cropping determined by combining distance and degree of impact integrates spatial location and dynamic change information, ensuring the comprehensiveness of the sampling point assessment. Since distance reflects proximity to hotspot areas, and degree of impact quantifies the current contribution, this two-factor assessment reduces the risk of misselection and improves the accuracy of measuring the representativeness of continuous cropping points.
[0111] It is understandable that Fritillaria thunbergii is usually planted in alignment in one direction. The stronger the representativeness of the sampling point as a breakdown of the cumulative amount of autotoxic substances in continuous cropping, the more it can be used as the basis for analyzing the distribution pattern. That is, the representativeness of continuous cropping at a sampling point can be used as the weight of the autotoxic substance content, so that the sampling point can be used as a reference for analyzing historical continuous cropping (i.e., historical continuous cropping reference).
[0112] Optionally, the product of the representativeness of continuous cropping at the first sampling point and the content of autotoxic substances at the first sampling point can be used as the historical continuous cropping reference for the first sampling point.
[0113] It should be understood that this historical continuous cropping reference is used to characterize the weighted reference value of the sampling point for the analysis of historical continuous cropping distribution patterns, and to quantify the importance of the point in spatial distribution pattern recognition.
[0114] Optionally, the absolute value of the historical contiguous reference difference between any two adjacent sampling points in the first candidate row can be determined as the historical contiguous reference difference between the two sampling points.
[0115] Within a candidate row, if all sampling points are more similar in terms of their historical continuous cropping reference for analyzing the distribution pattern of autotoxic substances in Fritillaria thunbergii, then the candidate row is more likely to be a historical continuous cropping row for previous Fritillaria thunbergii plantings. Therefore, the distribution pattern of autotoxic substances in a candidate row can be analyzed based on the historical continuous cropping reference between adjacent sampling points.
[0116] Optionally, the distribution regularity of autotoxic substances in a candidate row satisfies the following formula:
[0117]
[0118] in, Indicates candidate rows The distribution regularity of autotoxic substances, where N represents candidate rows. The number of inline sampling points, Indicates candidate rows The number of adjacent sampling point pairs within the same area. Indicates candidate rows Sampling points within Historical references Indicates candidate rows Sampling points within Historical references This represents an exponential function with the natural constant e as its base.
[0119] In this formula, This exponential function represents the absolute value of the historical reference difference between adjacent sampling points. In the middle, Taking a negative value can achieve the following: The smaller, The larger the number of candidate rows Between adjacent sampling points When both are relatively small, it indicates that the candidate row... The spatial consistency of the distribution of endotoxins is high, and this candidate line The higher the regularity of the distribution of autotoxic substances.
[0120] In this embodiment, historical continuous cropping reference is calculated based on the degree of impact and the content of autotoxic substances, and the distribution regularity is determined by evaluating the reference differences between adjacent points within a row. This effectively captures the spatial continuity and uniformity of autotoxic substances and can effectively distinguish between real historical continuous cropping rows and noisy rows.
[0121] S204. The candidate row with the greatest regularity in the distribution of autotoxic substances is identified as the historical continuous cropping row.
[0122] It is understandable that the more regular the distribution of autotoxic substances, the more the content of autotoxic substances at the sampling points in the candidate row is affected by historical continuous cropping, and the candidate row can be identified as a historical continuous cropping row.
[0123] The methods provided in S201-S204 above achieve precise location of historical continuous cropping rows by introducing a candidate row screening mechanism and distribution regularity assessment. This method fully considers the actual distribution characteristics of autotoxic substances in the field, and by quantitatively analyzing the degree of influence and autotoxic substance content at sampling points within each candidate row, it can effectively identify the spatial distribution patterns that best represent historical continuous cropping patterns.
[0124] In one implementation of this application, after determining the representativeness of continuous cropping at each sampling point, the data processing unit 102, when determining the cumulative benchmark value of continuous cropping and the content of autotoxic substances generated by each sampling point in the current sampling period based on the autotoxic substance content of each continuous cropping point at the current sampling time, may specifically: determine the cumulative benchmark value of continuous cropping based on the distribution regularity of autotoxic substances in historical continuous cropping rows, the content of autotoxic substances of all continuous cropping points in historical continuous cropping rows at the current sampling time, and the representativeness of continuous cropping at each continuous cropping point; and determine the content of autotoxic substances generated by each sampling point in the current sampling period by the difference between the content of autotoxic substances at each sampling point at the current sampling time and the cumulative benchmark value of continuous cropping.
[0125] Optionally, the cumulative baseline value for consecutive operations satisfies the following formula:
[0126]
[0127] in, This represents the cumulative baseline value over consecutive operations. Indicates historical continuity The distribution regularity of autotoxic substances, Indicates historical continuity Zhonglian work point Representative of continuous work, Indicates historical continuity The middle connecting point The content of autotoxic substances, Indicates historical continuity The number of points connected in the middle.
[0128] In this formula, Indicates historical continuity All connecting points The content of autotoxic substances is calculated as a weighted sum based on the representativeness of each compound. For all connected points The sum of representative values of consecutive operations is used as a normalization factor to ensure... For a reasonable weighted average, Indicates historical continuity The cumulative amount of continuous cropping in the middle, It can be understood as The confidence level is higher because the cumulative amount of continuous cropping in historical continuous cropping rows is slightly affected by the current planting. Therefore, the baseline value of continuous cropping cumulative amount obtained by correcting for high continuous cropping cumulative amount is more accurate.
[0129] Since the cumulative baseline value for continuous cropping represents the content of autotoxic substances accumulated before the current sampling time, the content of autotoxic substances at each sampling point collected at the current sampling time can be subtracted from the cumulative baseline value for continuous cropping to obtain the content of autotoxic substances generated at each sampling point in the current sampling period.
[0130] Combination Figure 1 ,like Figure 3 As shown, the intelligent soil environment monitoring system 10 for Fritillaria thunbergii cultivation also includes a model building unit 105.
[0131] Model building unit 105 is used to execute the following S301-S303.
[0132] S301. Based on the content of autotoxic substances at each sampling point at multiple sampling times, generate multiple autotoxic substance content sequences.
[0133] Each sampling point corresponds to a sequence of autotoxic substance content.
[0134] Understandably, the sequence lengths of these multiple autotoxic substance content sequences should be consistent.
[0135] S302. The least squares method is used to fit the content sequences of multiple autotoxic substances to obtain the initial parameters.
[0136] The initial parameter is used to characterize the overall trend of changes in the content of autotoxic substances.
[0137] It should be understood that when the initial parameter is positive, it indicates that the content of autotoxic substances is increasing, and when the initial parameter is negative, it indicates that the content of autotoxic substances is decreasing. The initial parameter obtained by fitting based on the least squares method can accurately reflect the change law of autotoxic substances in the time dimension.
[0138] S303. Construct a content prediction model based on initial parameters.
[0139] Optionally, the content prediction model includes the following formula:
[0140]
[0141] in, Indicates sampling point exist Predicted values of autotoxic substance content at any given time. This represents the cumulative baseline value over consecutive operations. This indicates the content of autotoxic substances generated during the current sampling period. This represents the initial parameters.
[0142] In this formula, This indicates the content of autotoxic substances that may be generated at the next sampling time. By observing the change pattern of autotoxic substance content in each sampling period, the content of autotoxic substances that may be generated in the next sampling period can be predicted. Then, by adding the cumulative baseline value of continuous cropping to the autotoxic substance content in the current sampling period, the cumulative autotoxic substance content in the soil in the next sampling period can be obtained.
[0143] It is understandable that by fitting the autotoxic substance content sequence at multiple sampling times using the least squares method, the overall change pattern of the autotoxic substance content can be accurately captured. Based on the change pattern, the content prediction model can be constructed, which can accurately predict the change of autotoxic substance content at the next sampling time, thereby obtaining the predicted value of autotoxic substance content at the next sampling time.
[0144] It should be understood that after constructing the content prediction model based on the initial parameters, the content prediction model can be optimized and updated based on the actual value obtained each time the content prediction model is used and the predicted value is obtained. Specifically, the initial parameters can be optimized and updated to improve the accuracy of the content prediction model.
[0145] In one optional implementation, the data acquisition unit 101 is further configured to: acquire the actual value of the autotoxic substance content of each sampling point at the next sampling time; the model building unit 105 is further configured to: adjust the initial parameters in the content prediction model based on the actual value of the autotoxic substance content of each sampling point at the next sampling time and the predicted value of the autotoxic substance content at the next sampling time.
[0146] Specifically, the actual value of the autotoxic substance content at each sampling point at the next sampling time can be compared with the predicted value of the autotoxic substance content, the residual can be calculated, and the initial parameters can be adjusted based on the residual and the adaptive weighted correction algorithm so that the error of the adjusted initial parameters is smaller.
[0147] Understandably, through multi-cycle iterative updates during use, the content prediction model can learn the changing patterns under different environmental conditions, improve the stability and generalization ability of the prediction, and achieve real-time, high-precision dynamic prediction of the content of autotoxic substances in Fritillaria thunbergii during cultivation, thus realizing intelligent monitoring of the soil environment with high precision and high frequency.
[0148] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0149] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting, characterized in that, The system includes a data acquisition unit, a data processing unit, a data prediction unit, and a data monitoring unit; The data acquisition unit is used to collect the content of autotoxic substances at multiple sampling points with a preset sampling period length, so as to obtain the content of autotoxic substances at each sampling point at multiple sampling times. The data processing unit is configured to: determine the rate of change of autotoxic substance content at each sampling point based on the autotoxic substance content at multiple sampling times; determine the maximum and minimum rates of change of autotoxic substance content at sampling points within a first local region, wherein the first local region is the local region where the first sampling point is located, and the first sampling point is one of the plurality of sampling points; and determine the ratio between a first difference and a second difference as the degree of influence of the first sampling point, wherein the first difference is the difference between the rate of change of autotoxic substance content at the first sampling point and the minimum rate of change of autotoxic substance content, and the second difference is the difference between the maximum rate of change of autotoxic substance content and the rate of change of autotoxic substance content at the first sampling point, wherein the degree of influence is used to characterize the extent to which a sampling point is affected by the current planting. N candidate rows are determined based on the positions of multiple sampling points, where N is an integer greater than or equal to 1; the distribution regularity of autotoxic substances in each candidate row is determined based on the degree of influence of the sampling points included in each candidate row and the content of autotoxic substances in each sampling point; the candidate row with the largest distribution regularity of autotoxic substances is determined as the historical continuous cropping row, and the sampling points included in the historical continuous cropping row are determined as continuous cropping points; The data processing unit is also used to determine the cumulative benchmark value of continuous cropping based on the distribution regularity of autotoxic substances in historical continuous cropping rows, the content of autotoxic substances at all continuous cropping points in historical continuous cropping rows at the current sampling time, and the continuous cropping representativeness of each continuous cropping point; and to determine the content of autotoxic substances generated by each sampling point in the current sampling period by the difference between the content of autotoxic substances at each sampling point at the current sampling time and the cumulative benchmark value of continuous cropping. The data prediction unit is used to input the content of autotoxic substances generated by each sampling point in the current sampling period and the cumulative baseline value of continuous cropping into the content prediction model to obtain the predicted value of autotoxic substance content of each sampling point at the next sampling time. The content prediction model is used to predict the content of autotoxic substances. The data monitoring unit is used to monitor the soil environment based on the predicted content of autotoxic substances at each sampling point at the next sampling time.
2. The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 1, characterized in that, The data processing unit is specifically used for: M current planting rows are determined based on the positions of multiple sampling points, where M is an integer greater than or equal to 1; Rotate the M current planting rows by at least one angle and / or translate them by at least one preset distance to obtain N candidate rows, where N≥M. 3.The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 1, characterized in that, The data processing unit is specifically used for: Based on the degree of impact of the sampling points included in the first candidate row and the content of autotoxic substances at each sampling point, the historical continuous reference of each sampling point is determined, and the first candidate row is any one of the N candidate rows; Determine the historical contiguous reference differences between every two adjacent sampling points within the first candidate row; Based on the historical continuous reference differences between every two adjacent sampling points in the first candidate row, the distribution regularity of autotoxic substances in the first candidate row is determined.
4. The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 3, characterized in that, The data processing unit is specifically used for: Determine the distance between the first sampling point and the root sampling points in the first local area, wherein the root sampling points in the first local area are the sampling points corresponding to the maximum rate of change of autotoxic substance content in the first local area. Based on the distance and the degree of influence of the first sampling point, the representativeness of the continuous operation of the first sampling point is determined; Based on the representativeness of continuous cropping at the first sampling point and the content of autotoxic substances at the first sampling point, the historical continuous cropping reference value of the first sampling point is determined.
5. The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 1, characterized in that, The system further includes a model building unit, which is used for: Based on the content of autotoxic substances at each sampling point at multiple sampling times, multiple autotoxic substance content sequences are generated, with one sampling point corresponding to one autotoxic substance content sequence. The least squares method was used to fit the content sequences of the multiple autotoxic substances to obtain initial parameters, which were used to characterize the overall trend of the content of autotoxic substances. A content prediction model is constructed based on the initial parameters. 6.The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 5, characterized in that, The data acquisition unit is also used for: Obtain the actual value of the autotoxic substance content at each sampling point at the next sampling time; The model building unit is also used to adjust the initial parameters in the content prediction model based on the actual value of the autotoxic substance content at the next sampling time and the predicted value of the autotoxic substance content at the next sampling time for each sampling point.
7. The soil environment intelligent monitoring system for Fritillaria thunbergii Miq. planting according to claim 1, characterized in that, The data monitoring unit is specifically used for: An alarm will be issued if the predicted content of the self-toxic substance at the first sampling point exceeds the preset risk threshold.