Method and system for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy

Abstract

The application provides a kind of agricultural soil heavy metal detection method and system based on laser-induced breakdown spectroscopy, related to heavy metal detection technical field, by collecting the temperature and humidity data, moisture content data of the agricultural soil to be detected, and corresponding detection batch information;According to the above two types of data, the process of generating plasma when simulating the excitation of the agricultural soil to be detected by using digital twin simulation model and laser-induced breakdown spectroscopy is generated, target laser energy parameters and target acquisition parameters are generated to collect the optical signal of the plasma radiation generated by the agricultural soil to be detected;After separating the target waveband signal from the optical signal, correction, conversion, baseline correction and outlier rejection processing are carried out, to obtain the corrected digital signal, and a multi-waveband feature set is extracted from the corrected digital signal, the set is cross-validated, combined with the detection batch information, to generate the heavy metal detection result of the agricultural soil to be detected, to improve the precision, stability and adaptability of heavy metal detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heavy metal detection, and particularly relates to a heavy metal detection method and system for agricultural soil based on laser-induced breakdown spectroscopy. BACKGROUND

[0002] In modern agricultural production and soil environment monitoring, rapid and accurate detection of heavy metal content in agricultural soil has become a key technical requirement for ensuring the safety of agricultural products and ecological health. Due to the complexity of soil composition and significant influence of environmental factors, the detection method needs to have high sensitivity, strong adaptability and on-site operability, and can effectively cope with the interference of different temperature and humidity conditions and water content on the detection results, so as to realize real-time evaluation and accurate control of heavy metal pollution.

[0003] The detection method of the prior art mainly uses laser-induced breakdown spectroscopy combined with a fixed parameter excitation strategy to analyze soil heavy metals. This method directly obtains plasma radiation spectrum on various soil samples by presetting a unified laser energy and signal acquisition time, and analyzes the characteristics of heavy metal elements by using a standardized model.

[0004] However, the existing method has some defects, for example, it is easily affected by soil temperature, humidity and water content fluctuations in actual application, which reduces the accuracy and reliability of heavy metal quantitative analysis. Especially in the complex environment of the field, the characteristic signal is often distorted or weakened, which is difficult to meet the requirements of high-precision on-site detection. SUMMARY

[0005] The purpose of the present application is to provide a heavy metal detection method and system for agricultural soil based on laser-induced breakdown spectroscopy to solve the problem of unstable spectral signal and low heavy metal detection precision caused by fixed parameters not adapting to changes in soil environment in the prior art.

[0006] To solve the above technical problems, in a first aspect, the present application provides a heavy metal detection method for agricultural soil based on laser-induced breakdown spectroscopy, comprising:

[0007] Collecting temperature and humidity data, water content data of the agricultural soil to be detected, and corresponding detection batch information;

[0008] According to the temperature and humidity data and the water content data, a digital twin simulation model and a laser-induced breakdown spectroscopy are used to simulate the process of generating plasma in the agricultural soil to be detected, to generate target laser energy parameters and target acquisition parameters;

[0009] Collecting the light signal of the plasma radiation generated when the target laser beam acts on the agricultural soil to be detected, and the emission parameters of the target laser beam are determined based on the target laser energy parameters and target acquisition parameters;

[0010] Separating a target waveband signal from the light signal, correcting the target waveband signal to obtain a corrected waveband signal;

[0011] Converting the corrected waveband signal into a digital signal, performing baseline correction and abnormal signal rejection processing on the digital signal to obtain a corrected digital signal;

[0012] Extracting a multi-waveband feature set from the corrected digital signal by using a quantitative analysis model, performing cross-validation on the multi-waveband feature set to obtain correlation feature information, and combining the detection batch information to generate a heavy metal detection result of the agricultural soil to be detected.

[0013] Optionally, according to the temperature and humidity data and the water content data, a digital twin simulation model and laser-induced breakdown spectroscopy are used to simulate the process of generating plasma from the agricultural soil to be detected, to generate target laser energy parameters and target acquisition parameters, including:

[0014] According to the water content data, the particle pore and water affinity parameters of the agricultural soil to be detected are determined;

[0015] According to the particle pore, water affinity parameters and soil particle material parameters of the agricultural soil to be detected, the initial state data when the plasma is formed is generated by the laser simulation module of the digital twin simulation model and the energy propagation process of the laser beam in different soil region units of the laser-induced breakdown spectroscopy.

[0016] According to the temperature and humidity data, the existence duration and energy distribution parameters of the plasma in the initial state data are adjusted to generate final state data;

[0017] According to the final state data, the preset initial laser energy parameters and the preset initial acquisition parameters are adjusted until the target laser energy parameters and the target acquisition parameters when the radiation intensity simulation data of the plasma meets the stable state condition are obtained.

[0018] Optionally, according to the particle pore, water affinity parameters and soil particle material parameters of the agricultural soil to be detected, the initial state data when the plasma is formed is generated by the laser simulation module of the digital twin simulation model and the energy propagation process of the laser beam in different soil region units of the laser-induced breakdown spectroscopy, including:

[0019] According to the particle pore, water affinity parameters and soil particle material parameters of the agricultural soil to be detected, the influence parameter set is determined, which includes soil regions of different void levels, connected path directions, water affinity region distribution and absorption coefficients of different soil particle types to laser energy;

[0020] According to the laser pulse characteristics of laser-induced breakdown spectroscopy, initial laser parameters of the laser beam are set;

[0021] According to the influence parameter set and the initial laser parameters, a laser simulation module of a digital twin simulation model simulates the energy propagation process of the laser beam in different soil area units according to the spatial structure of the agricultural soil to be detected, generates an energy data table, and the energy data table includes the residual energy density of each soil area unit;

[0022] According to the energy data table, the preset ionization energy of the target soil substance in the agricultural soil to be detected is corrected to obtain a corrected ionization energy, and the soil area unit with a residual energy density greater than or equal to the corrected ionization energy is selected as an initial formation area of plasma;

[0023] According to the area characteristics of each initial formation area, the initial ionization degree of the soil substance in each initial formation area is calculated to form initial state data when the plasma is formed.

[0024] Optionally, according to the influence parameter set and the initial laser parameters, a laser simulation module of a digital twin simulation model simulates the energy propagation process of the laser beam in different soil area units according to the spatial structure of the agricultural soil to be detected, generates an energy data table, and the energy data table includes the residual energy density of each soil area unit, including:

[0025] According to the spatial structure of the agricultural soil to be detected, the agricultural soil to be detected is divided into a plurality of soil area units;

[0026] The propagation step length of the laser beam is set, and the laser simulation module of the digital twin simulation model simulates the energy propagation process of the laser beam in each soil area unit according to the propagation step length of the laser beam until the residual energy of the laser beam is lower than a preset energy or the laser beam propagates to a preset depth, and an energy data table is generated.

[0027] Optionally, the corrected wave band signal is converted into a digital signal, and the digital signal is subjected to baseline correction and abnormal signal rejection processing to obtain a corrected digital signal, including:

[0028] A preset sampling frequency corresponding to the characteristic wave band range of the soil heavy metal suitable for the corrected wave band signal is determined, and an analog-to-digital conversion operation is performed on the corrected wave band signal according to the preset sampling frequency to obtain a digital signal;

[0029] The digital signal is subjected to baseline analysis, target data segments in the digital signal that meet signal fluctuation conditions are identified, and a reference value is calculated based on the target data segments;

[0030] According to the reference value, the digital signal is adjusted to obtain an adjusted signal;

[0031] According to the characteristic waveband range, a signal amplitude range is set, and data points with amplitude values exceeding the signal amplitude range in the adjusted signal are marked as abnormal data points;

[0032] A numerical average value of normal data points adjacent to the abnormal data points in the adjusted signal is calculated, the signal numerical value of the abnormal data points in the adjusted signal is replaced by the numerical average value, and a corrected digital signal is obtained.

[0033] Optionally, a multi-waveband feature set is extracted from the corrected digital signal by using a quantitative analysis model, and cross-validation is performed on the multi-waveband feature set to obtain correlation feature information, including:

[0034] Standard spectrum data of soil heavy metal standard samples with different concentration gradients are obtained, and a corresponding relationship between each standard spectrum data and a concentration value of a corresponding soil heavy metal standard sample is established;

[0035] According to the standard spectrum data, a characteristic peak position corresponding to a target heavy metal in the agricultural soil to be detected is identified, and a characteristic waveband corresponding to the target heavy metal and a wavelength range of each characteristic waveband are determined in combination with a detection scene requirement of the agricultural soil to be detected;

[0036] According to the characteristic waveband and the wavelength range of each characteristic waveband, a multi-waveband feature set is extracted from the corrected digital signal by using the quantitative analysis model;

[0037] The multi-waveband feature sets corresponding to different detection batches are cross-validated, and feature weights of waveband features in the multi-waveband feature set that passes the cross-validation are determined according to a detection accuracy requirement of the target heavy metal, a corresponding association rule between the waveband features in the multi-waveband feature set that passes the cross-validation and soil heavy metal concentrations is determined in combination with the multi-waveband feature set that passes the cross-validation and the corresponding relationship, and all corresponding association rules are integrated to obtain correlation feature information.

[0038] Optionally, a target waveband signal is separated from the optical signal, and the target waveband signal is corrected to obtain a corrected waveband signal, including:

[0039] Based on a known characteristic radiation wavelength of the target heavy metal in the agricultural soil to be detected, a target waveband range corresponding to the target heavy metal is determined;

[0040] The target waveband signal within the target waveband range is extracted from the optical signal;

[0041] According to a preset interference relationship table, a target influence ratio corresponding to the temperature and humidity data is queried, and the target influence ratio includes a signal gain ratio and a signal attenuation ratio;

[0042] Based on the target influence ratio, the target waveband signal is corrected to obtain a preliminary correction signal;

[0043] According to the target waveband range, the preliminary correction signal is wavelength calibrated to correct the spectral line drift caused by the temperature and humidity data, to obtain a corrected waveband signal.

[0044] In a second aspect, the present application provides a method and system for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy, comprising:

[0045] A first acquisition module is configured to acquire temperature and humidity data, water content data, and corresponding detection batch information of the agricultural soil to be detected;

[0046] A simulation module is configured to simulate the process of generating plasma in the agricultural soil to be detected based on the temperature and humidity data and the water content data, using a digital twin simulation model and laser-induced breakdown spectroscopy, to generate target laser energy parameters and target acquisition parameters;

[0047] A second acquisition module is configured to acquire a light signal of plasma radiation generated when a target laser beam acts on the agricultural soil to be detected, the emission parameters of the target laser beam being determined based on the target laser energy parameters and the target acquisition parameters;

[0048] A correction module is configured to separate a target waveband signal from the light signal, correct the target waveband signal, and obtain a corrected waveband signal;

[0049] A processing module is configured to convert the corrected waveband signal into a digital signal, perform baseline correction and abnormal signal rejection processing on the digital signal, and obtain a corrected digital signal;

[0050] A generation module is configured to extract a multi-waveband feature set from the corrected digital signal using a quantitative analysis model, perform cross-validation on the multi-waveband feature set, obtain associated feature information, combine the detection batch information, and generate a heavy metal detection result of the agricultural soil to be detected.

[0051] In a third aspect, the present application provides an electronic device, comprising:

[0052] A memory is configured to store a computer program;

[0053] A processor is configured to execute the computer program to implement the steps of the method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to the first aspect described above.

[0054] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program, when executed by a processor, can implement the steps of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method according to the first aspect.

[0055] The laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method provided by the present application collects the temperature, humidity, water content and detection batch information of the soil; then, the digital twin simulation and the laser-induced breakdown spectroscopy technology are combined to dynamically generate laser energy and signal acquisition parameters that adapt to the current soil conditions, thereby ensuring the stability of the plasma excitation process.

[0056] On this basis, the plasma radiation light signal generated by the optimized parameters is accurately collected, and the influence of environmental interference on the original spectrum is suppressed through target waveband separation and correction; then, the corrected light signal is digitized, and baseline correction and abnormal rejection are implemented to improve the signal quality and consistency; finally, the quantitative analysis model is used to extract and cross-verify the multi-waveband features, and the batch information is fused to output the heavy metal detection result, and the overall process realizes high-precision and high-robustness recognition of the soil heavy metal content under complex field conditions, overcoming the shortcomings of the traditional fixed parameter strategy in dynamic environment adaptability.

[0057] Further, based on the soil water content, the particle pore and water affinity characteristics are derived, combined with the material information, the laser energy propagation in different soil microzones is simulated through the laser simulation module in the digital twin model to construct the initial state of the plasma; then, the plasma duration and energy distribution under this state are dynamically adjusted according to the temperature and humidity data to form the final state data; accordingly, the initial laser and acquisition parameters are iteratively optimized until the simulated radiation intensity reaches the stable condition, thereby determining the optimal target parameters; the excitation and acquisition conditions can finely match the real-time changes of the soil microstructure and macro environment, thereby enhancing the repeatability and feature expression ability of the plasma signal and improving the reliability and applicability of the heavy metal detection in the heterogeneous and variable soil environment. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0059] Figure 1 The flowchart of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method provided by the present application is shown in the figure;

[0060] Figure 2 A schematic diagram of generating target laser energy parameters and target acquisition parameters using a digital twin simulation model is provided for an embodiment of the present application;

[0061] Figure 3 A structural schematic diagram of an agricultural soil heavy metal detection system based on laser-induced breakdown spectroscopy is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0062] In view of the problem that the existing laser-induced breakdown spectroscopy method is difficult to adapt to the dynamic changes of soil temperature, humidity and water content in the detection of agricultural soil heavy metals due to the use of fixed excitation and acquisition parameters, the present application acquires the temperature and humidity state and water content characteristics of the soil in real time, and introduces digital twin simulation to pre-act the plasma excitation process, dynamically generates laser energy and signal acquisition parameters adapted to the current sample conditions, and improves the stability of the plasma radiation signal; On this basis, combined with multi-level optimization means such as band selection, signal correction, baseline processing and abnormal rejection, the spectral data quality is strengthened, and through the fusion of batch information and multi-band feature cross-validation, a more environmentally robust quantitative analysis process is constructed, and finally the detection of soil heavy metal content under complex field conditions is realized with high consistency and high adaptability.

[0063] In order for those skilled in the art to better understand the present application scheme, the present application will be further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0064] The core of the present application is to provide an agricultural soil heavy metal detection method based on laser-induced breakdown spectroscopy, and a flowchart of one specific embodiment of the method is shown in Figure 1 As shown in the figure, the method comprises:

[0065] Step 101: Collect the temperature and humidity data, water content data of the agricultural soil to be detected, and the corresponding detection batch information.

[0066] In this step, the detection batch information refers to the identification information recorded when the agricultural soil to be detected is collected. The detection batch information includes sample number, collection time, collection location and sample grouping identification.

[0067] In the present application, the temperature and humidity data of the agricultural soil to be detected are collected by a temperature and humidity sensor, the water content data of the water content in the soil are collected by a soil water content determination device, and the detection batch information corresponding to the soil sample is recorded.

[0068] Step 102: According to the temperature and humidity data and the moisture content data, the process of simulating the generation of plasma in the agricultural soil to be detected when laser-induced breakdown spectroscopy is stimulated is simulated by using a digital twin simulation model and a target laser energy parameter and a target acquisition parameter are generated.

[0069] In this step, the target laser energy parameter refers to the laser emission energy related parameter determined by the digital twin simulation model based on the soil temperature and humidity data and the moisture content data. The target laser energy parameter includes the laser energy size, the pulse duration, etc.

[0070] The target acquisition parameter refers to the signal acquisition related parameter matched with the target laser energy parameter. The target acquisition parameter includes the acquisition time window, the acquisition sensitivity, the sampling frequency, etc.

[0071] Step 103: The light signal of the plasma radiation generated when the target laser beam acts on the agricultural soil to be detected is collected. The emission parameters of the target laser beam are determined based on the target laser energy parameter and the target acquisition parameter.

[0072] In this step, the target laser beam refers to the specific laser beam emitted by the laser emission device to the agricultural soil to be detected based on the emission parameters determined based on the target laser energy parameter and the target acquisition parameter.

[0073] In the embodiments of the present application, first, the emission parameters of the target laser beam are determined based on the target laser energy parameter and the target acquisition parameter. The working state of the laser emission device is adjusted according to the emission parameters, and the target laser beam is emitted to the agricultural soil to be detected in combination with the timing requirements of the target acquisition parameter.

[0074] When the target laser beam acts on the agricultural soil to be detected, the substances in the soil are excited to generate plasma. The plasma radiates light signals including the characteristics of the soil substances in the energy release process. At this time, the light signals radiated by the plasma are captured in real time by the signal collection device according to the acquisition sensitivity, the acquisition time window, etc. set by the target acquisition parameter.

[0075] Step 104: The target waveband signal is separated from the light signal, and the target waveband signal is corrected to obtain a corrected waveband signal.

[0076] In this step, the target waveband signal refers to the specific waveband signal corresponding to the target heavy metal characteristic radiation wavelength extracted from the light signal radiated by the plasma. The target waveband signal includes the material characteristic information of the target heavy metal.

[0077] The corrected waveband signal refers to the target waveband signal after temperature and humidity interference correction and wavelength calibration.

[0078] Step 105: converting the corrected waveband signal into a digital signal, performing baseline correction and abnormal signal elimination on the digital signal, and obtaining a corrected digital signal.

[0079] The digital signal refers to a discretized electrical signal data obtained by analog-to-digital conversion of the corrected waveband signal.

[0080] Step 106: extracting a multi-waveband feature set from the corrected digital signal by using a quantitative analysis model, performing cross-validation on the multi-waveband feature set, obtaining associated feature information, and generating a heavy metal detection result of the agricultural soil to be detected in combination with the detection batch information.

[0081] In this step, the multi-waveband feature set refers to a set of features including a plurality of characteristic waveband signal peak positions, signal peak widths, signal peak intensities, etc. extracted from the corrected digital signal, and each characteristic waveband corresponds to a specific spectral feature of the target heavy metal.

[0082] The associated feature information refers to a set of corresponding association rules between the multi-waveband features and the soil heavy metal concentration determined through cross-validation.

[0083] In the embodiments of the present application, a quantitative analysis model is used to extract a multi-waveband feature set from the corrected digital signal, and cross-validation is performed on the multi-waveband feature set to obtain associated feature information. The quantitative analysis model is based on a convolutional neural network and soil heavy metal standard sample calibration. Then, taking the sample number in the detection batch information as the core matching basis, the associated feature information corresponding to each sample is accurately bound with the detection batch information to form an information association table, ensuring that the associated rules, collection time, collection location, and sample grouping identifier corresponding to each sample number are corresponding.

[0084] Subsequently, based on the information association table, the associated rules of all agricultural soil samples to be detected in the same sample grouping are summarized according to the sample grouping identifier, the occurrence frequency of different soil heavy metal concentration values in each sample grouping is counted, and the concentration value with the highest occurrence frequency is taken as the representative concentration value of the group. According to the collection location field in the information association table, all agricultural soil samples to be detected in the table are spatially classified, and the agricultural soil samples to be detected under the same collection location are integrated into a group. Then, the sample number and collection location information of each group of agricultural soil samples to be detected are extracted, and the associated rules corresponding to each agricultural soil sample to be detected, the representative concentration value of the grouping to which the agricultural soil sample belongs, and the target heavy metal type are combined to finally integrate and form a heavy metal detection result of the agricultural soil to be detected.

[0085] The embodiments of the present application can cope with the interference of temperature, humidity, and water content fluctuations, and balance rapid detection and high accuracy, and are suitable for complex field scenarios.

[0086] A specific embodiment is provided in the present application, for example,Figure 2 As shown, in step 102, according to the temperature and humidity data and the water content data, a digital twin simulation model and laser-induced breakdown spectroscopy are used to simulate the process of generating plasma in the to-be-detected agricultural soil when the target laser energy parameter and the target acquisition parameter are generated, and the specific steps include the following steps:

[0087] Step 201: According to the water content data, the particle pore and water affinity parameters of the to-be-detected agricultural soil are determined.

[0088] In this step, the water affinity parameter refers to a quantitative index obtained based on the water content data to characterize the interaction ability of the to-be-detected agricultural soil with water.

[0089] In the embodiment of the present application, according to the correlation between the soil water content in the water content data and the particle arrangement, the water binding capacity, the influence of the water content on the looseness of the soil particles is analyzed to determine the size of the voids between the soil particles, and the particle pore of the to-be-detected agricultural soil is obtained. Then, according to the contact state of the water and the soil particles, the water adsorption capacity, the diffusion speed of the water in the soil pore, the infiltration range and the uniformity of the soil being wetted by the water are determined, and combined with the specific properties of the soil particles such as the specific surface area and the pore structure, the intuitive characteristics of the adhesion and the infiltration are converted into quantitative indexes to determine the water affinity parameter.

[0090] Step 202: According to the particle pore, the water affinity parameter and the soil particle material parameter of the to-be-detected agricultural soil, the initial state data of the plasma when it is formed is generated by the laser simulation module of the digital twin simulation model and the energy propagation process of the laser beam in different soil region units.

[0091] In this step, the initial state data refers to a related data set when the plasma is just formed by simulating the energy propagation process of the laser beam in different soil region units.

[0092] Step 203: According to the temperature and humidity data, the existence duration and energy distribution parameter of the plasma in the initial state data are adjusted to generate the final state data.

[0093] In this step, the existence duration of the plasma refers to the time length from the formation to the disappearance of the plasma, which is a key parameter for characterizing the stability of the plasma.

[0094] The energy distribution parameter refers to a quantitative index for characterizing the propagation and distribution of the plasma energy in space.

[0095] The final state data refers to the plasma state data obtained after adjusting the existence duration and the energy distribution parameter in the initial state data according to the temperature and humidity data.

[0096] In the embodiments of the present application, according to the temperature and humidity data, the influence of temperature and humidity on the stability of plasma and energy propagation is analyzed, and according to the correlation between temperature and humidity and the existence time and energy distribution of plasma, the existence time and energy distribution parameters of plasma in the initial state data are adjusted to generate final state data more suitable for the actual environment.

[0097] For example, in the scene of detecting heavy metals in agricultural soil in region A, the influence of temperature and humidity on plasma is analyzed according to the temperature and humidity data of the soil to be detected. Specifically, when the temperature is high, the particle motion rate inside the plasma increases, the collision frequency increases, which leads to the plasma being more likely to dissipate, and the existence time is correspondingly shortened; when the humidity is at a moderate level, the water vapor in the soil to be detected will slightly hinder the energy propagation of the plasma, so that the energy is more likely to gather in the local area, and the energy distribution is relatively concentrated.

[0098] According to the above analysis results, combined with the preset correlation between temperature and humidity and the existence time and energy distribution of plasma, and the initial state data, for the case of high temperature, the existence time of plasma in the initial state data is appropriately shortened to make it more consistent with the dissipation law under high temperature environment. For the case of moderate humidity, adjust the energy distribution parameters to reduce the unnecessary diffusion of energy to the periphery and enhance the concentration of energy in the core area. The preset correlation determines the adjustment range of the existence time corresponding to different temperature intervals and the concentration adjustment coefficient of the energy distribution corresponding to different humidity levels.

[0099] Through the similar above-mentioned and actual temperature and humidity working condition precise matching targeted adjustment mode, the initial state data is corrected to the final state data more suitable for the current temperature and humidity environment in region A.

[0100] Step 204: According to the final state data, adjust the preset initial laser energy parameter and the preset initial acquisition parameter until the target laser energy parameter and the target acquisition parameter are obtained when the radiation intensity simulation data of the plasma meets the stable state condition.

[0101] In this step, the preset initial laser energy parameter refers to the initial parameter related to the laser energy for exciting the soil to generate plasma which is set in advance before the simulation starts. The initial laser energy parameter includes initial energy density, spot diameter, etc.

[0102] The preset initial acquisition parameter refers to the initial parameter related to the acquisition of plasma radiation signal which is set in advance before the simulation starts. The initial acquisition parameter includes acquisition time window, sensitivity, etc.

[0103] The radiation intensity simulation data refers to the intensity related data of the plasma radiation light signal simulated by the digital twin simulation model.

[0104] The stable state condition refers to a preset criterion for judging whether the preset radiation intensity simulation data is stable.

[0105] In the embodiments of the present application, the preset initial laser energy parameter and the preset initial acquisition parameter can be called, and the size of the initial laser energy parameter and the setting of the initial acquisition parameter can be gradually adjusted according to the plasma state reflected by the final state data. The radiation intensity simulation data is updated synchronously every time the adjustment is made, until the radiation intensity simulation data reaches the stable state condition with small fluctuation amplitude and high stability, at which time the corresponding adjusted parameters are the target laser energy parameter and the target acquisition parameter.

[0106] Optionally, in step 202, initial state data when plasma is formed is generated by a laser simulation module of the digital twin simulation model and a laser-induced breakdown spectroscopy simulation laser beam energy propagation process in different soil region units according to the particle pore, moisture affinity parameter and soil particle material parameter of the agricultural soil to be detected, specifically including the following steps:

[0107] In step 211, an influence parameter set is determined according to the particle pore, moisture affinity parameter and soil particle material parameter of the agricultural soil to be detected, and the influence parameter set includes soil regions of different void levels, connected path directions, moisture affinity region distributions and absorption coefficients of different soil particle types to laser energy.

[0108] In this step, the influence parameter set refers to a set of key factors affecting laser energy propagation determined based on the particle pore, moisture affinity parameter and soil particle material parameter.

[0109] The soil region of the void level refers to a soil region formed after different levels are divided according to the size of the void between soil particles.

[0110] The moisture affinity region distribution refers to the distribution of regions with different interaction capabilities with water in the soil in space.

[0111] In the embodiments of the present application, the specific influence of each parameter on laser energy propagation can be analyzed one by one based on the particle pore, moisture affinity parameter and soil particle material parameter of the agricultural soil to be detected, to determine the influence parameter set. Specifically, different soil regions of different void levels can be divided according to the actual size of the particle pore, the connected path direction of the internal void of the soil can be determined in combination with the distribution of the particle void of the agricultural soil to be detected, the range of the moisture affinity region distribution formed by the combination of the agricultural soil to be detected and water can be determined according to the moisture affinity parameter, the absorption coefficient of laser energy of different types of soil particles can be calculated according to the material characteristics of the soil particles, and finally these key factors are integrated to form a complete influence parameter set.

[0112] Wherein, the absorption coefficient of different soil particle types to laser energy is obtained by multiplying the quantitative parameter of the basic absorption capacity of soil particle material and the correction coefficient of particle size, the quantitative parameter of the basic absorption capacity of soil particle material refers to the quantitative value of the inherent absorption degree of the composition of soil particles to the specific wavelength laser, and the correction coefficient of particle size refers to the adjustment ratio determined according to the actual particle size of soil particles, which is used to calibrate the influence of the difference of the acting area of particles and laser under different particle sizes on the absorption effect.

[0113] Step 212: setting initial laser parameters of the laser beam according to the laser pulse characteristics of laser-induced breakdown spectroscopy.

[0114] In this step, the initial laser parameters refer to the initial parameters of the laser beam set according to the laser pulse characteristics of laser-induced breakdown spectroscopy, and the initial laser parameters include initial energy density, spot diameter, pulse duration, etc.

[0115] In the embodiments of the present application, the laser pulse characteristics of laser-induced breakdown spectroscopy are analyzed, including the duration of the pulse, the energy release law, etc., and then the initial laser parameters of the laser beam are set in combination with the actual needs of soil detection to provide initial laser conditions for simulating the propagation of laser energy.

[0116] Step 213: simulating the energy propagation process of the laser beam in different soil area units according to the spatial structure of the agricultural soil to be detected through the laser simulation module of the digital twin simulation model according to the influence parameter set and the initial laser parameters, and generating an energy data table, wherein the energy data table includes the residual energy density of each soil area unit.

[0117] In this step, the energy data table refers to a structured data set recording the residual energy density of each soil area unit after the laser energy propagation.

[0118] The residual energy density refers to the ratio of the residual energy of the laser beam after propagating in the soil area unit to the volume of the area unit.

[0119] Step 214: correcting the preset ionization energy of the target soil substance in the agricultural soil to be detected according to the energy data table to obtain the corrected ionization energy, and selecting the soil area unit with a residual energy density greater than or equal to the corrected ionization energy as the initial formation area of the plasma.

[0120] In this step, the target soil substance refers to a specific substance in the agricultural soil to be detected, which needs to detect the content of heavy metals, and the target soil substance includes heavy metal elements and substances related to detection in soil matrix.

[0121] The preset ionization energy refers to a preset minimum energy threshold required for ionization of a target soil material, and is a basic criterion for determining whether a soil region unit can form a plasma.

[0122] The initial formation region refers to a soil region unit with a residual energy density greater than or equal to the corrected ionization energy, and is the starting position when the plasma is just formed.

[0123] In the embodiments of the present application, the residual energy density of each soil region unit is extracted from the energy data table, and the preset ionization energy of the target soil material in the agricultural soil to be detected is retrieved. Then, the preset ionization energy of the target soil material is adjusted according to the residual energy density of each soil region unit in combination with the characteristics of the soil region unit, to obtain the corrected ionization energy adapted to the actual energy environment of each region unit. Then, the soil region unit with a residual energy density greater than or equal to the corrected ionization energy is selected as the initial formation region of the plasma.

[0124] Step 215: According to the region characteristics of each initial formation region, the initial ionization degree of the soil material in each initial formation region is calculated to form the initial state data when the plasma is formed.

[0125] In this step, the region characteristics refer to the relevant attribute information of each initial formation region, and the region characteristics include the spatial coordinate range, the residual energy density peak, the starting time of plasma formation, and the energy accumulation rate.

[0126] The initial ionization degree refers to the degree of ionization of the soil material in the initial formation region when the plasma is just formed, and is a key parameter for representing the initial state of the plasma formation, which is used to construct the initial state data.

[0127] In the embodiments of the present application, the region characteristics of each initial formation region are extracted, and then the degree of ionization of the soil material in each initial formation region is calculated according to the energy data in the region characteristics and the characteristics of the target soil material, to obtain the initial ionization degree. Finally, the region characteristics and the initial ionization degree of each initial formation region are integrated to form the initial state data when the plasma is formed.

[0128] Optionally, in step 213, according to the set of influence parameters and the initial laser parameters, the energy propagation process of the laser beam in different soil region units is simulated according to the spatial structure of the agricultural soil to be detected by a laser simulation module of a digital twin simulation model, to generate an energy data table, and the energy data table includes the residual energy density of each soil region unit, and specifically includes the following steps:

[0129] Step 221: According to the spatial structure of the agricultural soil to be detected, the agricultural soil to be detected is divided into multiple soil region units.

[0130] In the embodiments of the present application, according to the actual spatial structure of the agricultural soil to be detected, the soil is divided into a plurality of soil area units with clear boundaries and no overlap, and each soil area unit corresponds to a specific void level, a moisture affinity state or a soil particle type.

[0131] Step 222: set the propagation step length of the laser beam, simulate the energy propagation process of the laser beam in each soil area unit by the laser simulation module of the digital twin simulation model according to the propagation step length of the laser beam, until the remaining energy of the laser beam is lower than the preset energy or the laser beam propagates to the preset depth, and generate an energy data table.

[0132] In this step, the initial energy refers to the initial energy when the laser beam enters the surface of the agricultural soil to be detected, that is, the total energy corresponding to the initial energy density in the initial laser parameter.

[0133] In the embodiments of the present application, the propagation step length of the laser beam in the soil is set according to the accuracy requirement of the laser propagation simulation. Then, according to the propagation step length, the energy propagation process of the laser beam in each soil area unit is simulated in turn by the laser simulation module of the digital twin simulation model, and the energy loss of the laser beam in each area unit is calculated in real time. When the remaining energy of the laser beam is lower than the preset energy threshold or the propagation depth reaches the preset depth, the simulation is stopped, and finally the residual energy density data of each soil area unit is summarized to generate an energy data table.

[0134] The embodiments of the present application improve the pertinence and rationality of parameter setting, lay a foundation for subsequent accurate capture of plasma signal and improvement of heavy metal detection accuracy, and adapt to complex detection scenes in the field.

[0135] The present application provides a specific embodiment, step 105, converting the corrected waveband signal into a digital signal, performing baseline correction and abnormal signal rejection processing on the digital signal to obtain a corrected digital signal, specifically including the following steps:

[0136] Step 501: determine a preset sampling frequency that adapts to the corrected waveband signal and corresponds to the characteristic waveband range of the soil heavy metal, and perform an analog-to-digital conversion operation on the corrected waveband signal according to the preset sampling frequency to obtain a digital signal.

[0137] In this step, the characteristic waveband range of the soil heavy metal refers to a specific spectral waveband interval that can reflect the characteristics of the heavy metal substance determined based on the known characteristic radiation wavelength of the target heavy metal.

[0138] In the embodiment of the present application, firstly, the characteristic band range of the soil heavy metal is determined, and the signal characteristics such as the signal transmission speed and the signal change rule of the correction band signal are analyzed, the preset sampling frequency is determined according to the basic principle of signal sampling, in combination with the demand of the signal capture in the characteristic band range. Then, the continuously changing correction band signal is converted into discrete digital signal through the analog-digital conversion device according to the preset sampling frequency, so as to ensure that the converted digital signal can accurately retain the characteristic information related to the soil heavy metal in the original correction band signal.

[0139] Step 502: Baseline analysis is performed on the digital signal, target data segments meeting the signal fluctuation condition in the digital signal are identified, and a reference value is calculated based on the target data segments.

[0140] In this step, the signal fluctuation condition refers to the standard for judging the stability of the digital signal data segment, which can be that the signal amplitude change amplitude in the data segment is small, there is no obvious mutation and it is in the preset stable interval.

[0141] The reference value refers to the arithmetic mean value calculated based on all signal values in the target data segment, which is a unified calibration standard for eliminating the baseline drift of the digital signal.

[0142] In the embodiment of the present application, the digital signal is analyzed segment by segment to screen out the continuous data segment meeting the preset signal fluctuation condition as the target data segment. Then, all signal values in the target data segment are extracted, the arithmetic mean value of these values is calculated, and the mean value is determined as the reference value.

[0143] Step 503: The digital signal is adjusted according to the reference value to obtain an adjusted signal.

[0144] In the embodiment of the present application, the signal value of each data point in the digital signal is subtracted by the reference value respectively, the baseline drift interference in the digital signal is eliminated through this difference calculation, the signal reference tends to be consistent, and the adjusted signal is obtained, which creates more optimal conditions for subsequent abnormal data point identification.

[0145] Step 504: According to the characteristic band range, the signal amplitude range is set, and the data points with amplitude values exceeding the signal amplitude range in the adjusted signal are marked as abnormal data points.

[0146] In this step, the abnormal data point refers to the signal data point with the amplitude value exceeding the set signal amplitude range in the adjusted signal.

[0147] In the embodiment of the present application, the signal amplitude range is set reasonably by referring to the signal intensity interval corresponding to the characteristic waveband range of soil heavy metals and combining the accuracy requirement of the detection scene. Then each data point in the adjusted signal is checked one by one to determine whether the amplitude value is within the set signal amplitude range, and the signal data points exceeding the signal amplitude range are marked as abnormal data points.

[0148] Step 505: Calculate the numerical average of the normal data points adjacent to the abnormal data points in the adjusted signal, replace the signal value of the abnormal data point in the adjusted signal with the numerical average, and obtain a corrected digital signal.

[0149] In this step, the normal data point refers to the signal data point in the adjusted signal whose amplitude value is within the set signal amplitude range, and the signal value thereof can reflect the true characteristics of soil heavy metals.

[0150] In the embodiment of the present application, the normal data points adjacent to each abnormal data point are found, and the signal values of these normal data points are extracted to calculate the arithmetic average of these signal values. Then the average value is used to replace the signal value of the corresponding abnormal data point in the adjusted signal to eliminate the interference of abnormal data on the whole signal, so as to obtain a corrected digital signal that is purer and can reflect the true characteristics of soil heavy metals, thereby providing reliable input for subsequent multi-waveband feature extraction.

[0151] The embodiment of the present application can solve the problems of signal baseline drift and abnormal interference, and ensure that the corrected digital signal can accurately reflect the characteristics of heavy metals, thereby improving the accuracy and reliability of the detection result.

[0152] The present application provides a specific embodiment, step 106, using a quantitative analysis model to extract a multi-waveband feature set from the corrected digital signal, cross-validating the multi-waveband feature set to obtain correlation feature information, specifically including the following steps:

[0153] Step 601: Obtain standard spectrum data of soil heavy metal standard samples with different concentration gradients, and establish a corresponding relationship between each standard spectrum data and the concentration value of the corresponding soil heavy metal standard sample.

[0154] In this step, the standard spectrum data refers to the spectrum data obtained by performing spectrum collection on soil heavy metal standard samples with different concentration gradients by a laser-induced breakdown spectroscopy detection device, and the standard spectrum data includes characteristic spectral line information of the target heavy metal.

[0155] The corresponding relationship refers to the mapping relationship between the standard spectrum data and the known concentration value in the soil heavy metal standard sample.

[0156] In the embodiments of the present application, first, different concentration gradient soil heavy metal standard samples covering the preset low, medium and high concentration intervals are selected, and the spectrum of each selected standard sample is collected by a laser-induced breakdown spectroscopy detection device to obtain standard spectral data including heavy metal characteristic information. Then, the standard spectral data of each standard sample is corresponded to its known concentration value, and a corresponding relationship between the standard spectral data and the concentration value of the corresponding soil heavy metal standard sample is established.

[0157] Step 602: According to the standard spectral data, the characteristic peak position corresponding to the target heavy metal in the agricultural soil to be detected is identified, and the characteristic wave band corresponding to the target heavy metal and the wavelength range of each characteristic wave band are determined in combination with the detection scene demand of the agricultural soil to be detected.

[0158] In this step, the characteristic wave band refers to a specific wavelength interval that can stably reflect the characteristics of the target heavy metal, and each characteristic wave band corresponds to one or more spectral characteristics of the target heavy metal.

[0159] In the embodiments of the present application, the standard spectral data is analyzed by spectral line analysis to identify the characteristic peak position corresponding to the target heavy metal in the agricultural soil to be detected. Then, in combination with the detection scene demand of the agricultural soil to be detected, the characteristic peaks that can stably reflect the characteristics of the target heavy metal are screened out, the characteristic wave band corresponding to the target heavy metal is determined according to the distribution of the characteristic peaks, and the wavelength range of each characteristic wave band is determined to ensure that the characteristic wave band can accurately reflect the spectral information of the target heavy metal.

[0160] Step 603: According to the characteristic wave band and the wavelength range of each characteristic wave band, the multi-wave band feature set is extracted from the corrected digital signal by using the quantitative analysis model.

[0161] In the embodiments of the present application, the corrected digital signal is input into the quantitative analysis model, and then the signal peak position, signal peak width, signal peak value intensity and other key features in the corresponding characteristic wave band in the corrected digital signal are extracted by the model according to the set wavelength range, and these key features are classified and integrated according to the corresponding characteristic wave band to form a multi-wave band feature set.

[0162] Step 604: The multi-wave band feature sets corresponding to different detection batches are cross-validated, the feature weights of the wave band features in the multi-wave band feature set passing the cross-validation are determined according to the detection accuracy demand of the target heavy metal, the corresponding association rules between the wave band features in the multi-wave band feature set passing the cross-validation and the soil heavy metal concentration are determined in combination with the multi-wave band feature set passing the cross-validation and the corresponding relationship, all the corresponding association rules are integrated, and the association feature information is obtained.

[0163] In this step, the multi-band feature set that has passed cross-validation refers to the multi-band feature set with good repeatability and strong stability that has been selected after being compared and repeatedly validated among different detection batches.

[0164] The feature weight of a band feature refers to the importance coefficient assigned to each band feature in the multi-band feature set that has passed cross-validation, based on the detection accuracy requirements of the target heavy metal. The higher the weight, the greater the influence of that band feature on the concentration analysis.

[0165] Correspondence association rules refer to the inherent mapping rules between each band feature and the soil heavy metal concentration in a multi-band feature set established based on the correspondence between standard spectral data and concentration values.

[0166] In this embodiment of the application, firstly, the multi-band feature sets corresponding to different batches of detection information are determined. These multi-band feature sets are cross-validated by mutual comparison and repeated verification. Feature sets with poor repeatability and insufficient stability are eliminated, and multi-band feature sets that pass cross-validation are retained.

[0167] Next, based on the required detection accuracy of the target heavy metals, the importance of each band feature in the cross-validated multi-band feature set was determined, and the feature weights of each band feature were obtained. Subsequently, combining the cross-validated multi-band feature set with the previously established correspondence, the intrinsic relationship between each band feature and soil heavy metal concentration was analyzed, and the corresponding association rules between each band feature and soil heavy metal concentration were determined. Finally, all corresponding association rules were integrated to obtain complete association feature information.

[0168] The embodiments of this application enhance the characteristic signals of target heavy metals, reduce the influence of interference factors, and improve the accuracy and stability of quantitative analysis, providing core technical support for the accurate detection of soil heavy metal concentrations.

[0169] This application provides a specific embodiment. Step 104 involves separating the target band signal from the optical signal and correcting the target band signal to obtain a corrected band signal. This specifically includes the following steps:

[0170] Step 401: Based on the known characteristic radiation wavelengths of the target heavy metal in the agricultural soil to be detected, determine the target wavelength range corresponding to the target heavy metal.

[0171] In this step, the known characteristic radiation wavelength refers to the unique wavelength exhibited by the target heavy metal during plasma radiation, which is the spectral identifier of the target heavy metal.

[0172] The target band range refers to a specific spectral range defined by taking the known characteristic radiation wavelengths of the target heavy metal as the core, and combining anti-interference requirements and spectral resolution requirements.

[0173] In the embodiments of the present application, first, the type of the target heavy metal in the agricultural soil to be detected is determined, and then the known characteristic radiation wavelength of the target heavy metal in the process of plasma radiation is extracted from the spectral characteristic data of the target heavy metal. Second, in combination with the anti-interference requirement and the spectral resolution requirement of the detection scene, the basic spectral interval including the known characteristic radiation wavelength is determined, and then by analyzing the radiation wavelength distribution of other substances in the interval, the interference wavelength band easily overlapping with the signal of the target heavy metal is removed, and finally the target wavelength range capable of effectively distinguishing the interference of other substances and accurately covering the target heavy metal is determined.

[0174] Step 402: Extracting the target wavelength signal in the target wavelength range from the light signal.

[0175] In the embodiments of the present application, the light signal is filtered by the wavelength screening technology. Specifically, the signal component in the target wavelength range is retained, and the irrelevant interference signal beyond the range is removed, and the target wavelength signal including only the characteristic information of the target heavy metal is extracted.

[0176] Step 403: According to the preset interference relationship table, querying the target influence proportion corresponding to the temperature and humidity data, the target influence proportion including the signal gain proportion and the signal attenuation proportion.

[0177] In this step, the preset interference relationship table refers to a structured table for storing the corresponding relationship between different temperature and humidity conditions and signal influence proportions, which includes the temperature and humidity data, the signal gain proportion, the signal attenuation proportion and other related information.

[0178] The target influence proportion refers to the signal influence quantitative index obtained by querying the preset interference relationship table according to the temperature and humidity data.

[0179] In the embodiments of the present application, the preset interference relationship table is called, and the signal influence proportion corresponding to the current temperature and humidity data in the interference relationship table is searched as the target influence proportion.

[0180] Step 404: Based on the target influence proportion, correcting the target wavelength signal to obtain a preliminary correction signal.

[0181] In this step, the preliminary correction signal refers to the signal obtained by intensity correction of the target wavelength signal based on the target influence proportion, and the signal has been corrected for the intensity fluctuation caused by temperature and humidity.

[0182] In the embodiments of the present application, the intensity of each data point of the target wavelength signal is corrected by multiplying the signal original intensity by the signal gain proportion, and then multiplying by the difference of 1 minus the signal attenuation proportion, and the preliminary correction signal is obtained.

[0183] Step 405: According to the target waveband range, the preliminary correction signal is wavelength calibrated to correct the spectral line drift caused by the temperature and humidity data, to obtain a corrected waveband signal.

[0184] In this step, the spectral line drift caused by the temperature and humidity data refers to the phenomenon that the actual wavelength of the target waveband signal deviates from the standard wavelength position of the target waveband range due to temperature and humidity changes.

[0185] In the embodiments of the present application, the actual wavelength distribution of the preliminary correction signal can be compared with the standard wavelength position of the determined target waveband range, and the spectral line drift caused by the temperature and humidity data can be identified. Then, by adjusting the wavelength coordinates of the preliminary correction signal, the drifted spectral line is corrected to the standard wavelength position of the target waveband range, to obtain a corrected waveband signal with accurate wavelength and reliable intensity, so as to eliminate the interference of temperature and humidity on the spectral line position.

[0186] The embodiments of the present application can specifically eliminate the interference of temperature and humidity, improve the wavelength accuracy and intensity reliability of the signal, and guarantee the accuracy and stability of heavy metal detection.

[0187] Figure 3 A specific implementation structure diagram of an agricultural soil heavy metal detection system based on laser-induced breakdown spectroscopy provided by the embodiments of the present application is shown in Figure 3 The system can include:

[0188] A first acquisition module 21 is configured to acquire temperature and humidity data, water content data, and corresponding detection batch information of the agricultural soil to be detected.

[0189] An analog module 22 is configured to simulate the process of generating plasma in the agricultural soil to be detected by using a digital twin simulation model and laser-induced breakdown spectroscopy according to the temperature and humidity data and the water content data, to generate target laser energy parameters and target acquisition parameters.

[0190] A second acquisition module 23 is configured to acquire a light signal of plasma radiation generated when a target laser beam acts on the agricultural soil to be detected, and the emission parameters of the target laser beam are determined based on the target laser energy parameters and the target acquisition parameters.

[0191] A correction module 24 is configured to separate a target waveband signal from the light signal, correct the target waveband signal, and obtain a corrected waveband signal.

[0192] A processing module 25 is configured to convert the corrected waveband signal into a digital signal, perform baseline correction and abnormal signal rejection processing on the digital signal, and obtain a corrected digital signal.

[0193] The generating module 26 is configured to extract a multi-band feature set from the corrected digital signal by using a quantitative analysis model, cross-verify the multi-band feature set, obtain correlation feature information, and generate a heavy metal detection result of the agricultural soil to be detected in combination with the detection batch information.

[0194] The laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection system according to the embodiments of the present application is used to implement the foregoing laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method, and therefore the specific embodiments of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection system can be seen from the foregoing embodiments of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method, and the specific embodiments can be referred to the descriptions of the corresponding embodiments, which will not be described herein again.

[0195] The present application also provides an electronic device, comprising a memory for storing a computer program, and a processor for executing the computer program to implement the steps of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method.

[0196] The present application also provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the steps of the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method.

[0197] In an exemplary embodiment, the computer readable storage medium can include, but is not limited to, a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0198] The embodiments of the present application also provide a computer program product, and the computer program product includes a computer program, and the computer program is executed by a processor to implement the steps in the laser-induced breakdown spectroscopy-based agricultural soil heavy metal detection method.

[0199] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the foregoing description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0200] The above describes in detail the agricultural soil heavy metal detection method and system based on laser-induced breakdown spectroscopy provided by the present application. In this paper, specific examples are applied to explain the principles and implementation modes of the present application, and the above examples are only used to help understand the method of the present application and its core idea. It should be pointed out that for ordinary skilled persons in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy, characterized in that, include: Collect temperature and humidity data, moisture content data, and corresponding batch information of the agricultural soil to be tested; Based on the temperature and humidity data and the moisture content data, the process of stimulating plasma generation in the agricultural soil to be tested is simulated using a digital twin simulation model and laser-induced breakdown spectroscopy to generate target laser energy parameters and target acquisition parameters. The optical signal of plasma radiation generated when the target laser beam acts on the agricultural soil to be tested is collected. The emission parameters of the target laser beam are determined based on the target laser energy parameters and the target acquisition parameters. The target band signal is separated from the optical signal, and the target band signal is corrected to obtain the corrected band signal; The corrected band signal is converted into a digital signal, and the digital signal is subjected to baseline correction and abnormal signal removal processing to obtain a corrected digital signal. A multi-band feature set is extracted from the corrected digital signal using a quantitative analysis model. The multi-band feature set is cross-validated to obtain associated feature information. Combined with the detection batch information, the heavy metal detection results of the agricultural soil to be tested are generated. Based on the temperature and humidity data and the moisture content data, a digital twin simulation model and laser-induced breakdown spectroscopy are used to simulate the process of exciting plasma in the agricultural soil to be tested, in order to generate target laser energy parameters and target acquisition parameters, including: Based on the moisture content data, determine the particle porosity and water affinity parameters of the agricultural soil to be tested; Based on the particle porosity, water affinity parameters, and soil particle material parameters of the agricultural soil to be tested, the energy propagation process of the laser beam in different soil regional units is simulated through the laser simulation module and laser-induced breakdown spectrum of the digital twin simulation model to generate initial state data when plasma is formed. Based on the temperature and humidity data, the plasma existence duration and energy distribution parameters in the initial state data are adjusted to generate the final state data; Based on the final state data, adjust the preset initial laser energy parameters and preset initial acquisition parameters until the target laser energy parameters and target acquisition parameters are obtained when the simulated radiation intensity data of the plasma meets the steady-state conditions.

2. The method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to claim 1, characterized in that, Based on the particle porosity, water affinity parameters, and soil particle material parameters of the agricultural soil to be tested, the energy propagation process of the laser beam in different soil unit areas is simulated using the laser simulation module and laser-induced breakdown spectrum of the digital twin simulation model, generating initial state data for plasma formation, including: Based on the particle porosity, water affinity parameters, and soil particle material parameters of the agricultural soil to be tested, a set of influencing parameters is determined. The set of influencing parameters includes soil regions with different porosity levels, the orientation of the connectivity path, the distribution of water affinity regions, and the absorption coefficient of different soil particle types to laser energy. Based on the laser pulse characteristics of the laser-induced breakdown spectrum, the initial laser parameters of the laser beam are set. Based on the set of influencing parameters and the initial laser parameters, the laser simulation module of the digital twin simulation model simulates the energy propagation process of the laser beam in different soil regional units according to the spatial structure of the agricultural soil to be tested, and generates an energy data table, which includes the remaining energy density of each soil regional unit. According to the energy data table, the preset ionization energy of the target soil material in the agricultural soil to be detected is corrected to obtain the corrected ionization energy, and soil region units with remaining energy density greater than or equal to the corrected ionization energy are selected as the initial formation region of plasma. Based on the regional characteristics of each initial formation region, the initial degree of ionization of soil material within each initial formation region is calculated to form initial state data during plasma formation.

3. The method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to claim 2, characterized in that, Based on the set of influencing parameters and the initial laser parameters, the laser simulation module of the digital twin simulation model simulates the energy propagation process of the laser beam in different soil unit areas according to the spatial structure of the agricultural soil to be tested, generating an energy data table. The energy data table includes the remaining energy density of each soil unit, including: Based on the spatial structure of the agricultural soil to be tested, the agricultural soil to be tested is divided into multiple soil regional units; Set the propagation step size of the laser beam, and simulate the energy propagation process of the laser beam in each soil area unit through the laser simulation module of the digital twin simulation model according to the propagation step size of the laser beam, until the remaining energy of the laser beam is lower than the preset energy or the laser beam propagates to the preset depth, and generate an energy data table.

4. The method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to claim 1, characterized in that, The corrected band signal is converted into a digital signal, and the digital signal is subjected to baseline correction and outlier removal processing to obtain a corrected digital signal, including: A preset sampling frequency is determined that is compatible with the correction band signal and corresponds to the characteristic band range of heavy metals in the soil. The correction band signal is then subjected to analog-to-digital conversion according to the preset sampling frequency to obtain a digital signal. Baseline analysis is performed on the digital signal to identify target data segments in the digital signal that meet the signal fluctuation conditions, and a reference value is calculated based on the target data segments. The digital signal is adjusted according to the reference value to obtain the adjusted signal; Based on the characteristic band range, a signal amplitude range is set, and data points whose amplitude values ​​in the adjusted signal exceed the signal amplitude range are marked as abnormal data points. Calculate the average value of normal data points adjacent to abnormal data points in the adjusted signal, and replace the signal values ​​of abnormal data points in the adjusted signal with the average value to obtain the corrected digital signal.

5. The method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to claim 1, characterized in that, A multi-band feature set is extracted from the corrected digital signal using a quantitative analysis model. Cross-validation is then performed on the multi-band feature set to obtain associated feature information, including: Obtain standard spectral data of soil heavy metal standard samples with different concentration gradients, and establish the correspondence between each standard spectral data and the corresponding concentration values ​​of soil heavy metal standard samples; Based on the standard spectral data, the characteristic peak positions corresponding to the target heavy metals in the agricultural soil to be tested are identified. Combined with the testing scenario requirements of the agricultural soil to be tested, the characteristic bands corresponding to the target heavy metals and the wavelength range of each characteristic band are determined. Based on the characteristic bands and the wavelength range of each characteristic band, the quantitative analysis model is used to extract a multi-band feature set from the corrected digital signal; Cross-validate the multi-band feature sets corresponding to different batches of detection information, and determine the feature weights of each band feature in the cross-validated multi-band feature set according to the detection accuracy requirements of the target heavy metal. Combine the cross-validated multi-band feature set and the corresponding relationship to determine the corresponding association rules between each band feature in the cross-validated multi-band feature set and the soil heavy metal concentration. Integrate all corresponding association rules to obtain the associated feature information.

6. The method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy according to claim 1, characterized in that, The target band signal is separated from the optical signal, and the target band signal is corrected to obtain the corrected band signal, including: Based on the known characteristic radiation wavelengths of the target heavy metal in the agricultural soil to be detected, the target wavelength range corresponding to the target heavy metal is determined. Extract the target band signal within the target band range from the optical signal; According to the preset interference relationship table, the target influence ratio corresponding to the temperature and humidity data is queried. The target influence ratio includes the signal gain ratio and the signal attenuation ratio. Based on the target influence ratio, the target band signal is corrected to obtain a preliminary corrected signal; Based on the target band range, the preliminary correction signal is wavelength-calibrated to correct the spectral drift caused by the temperature and humidity data, thereby obtaining the corrected band signal.

7. A heavy metal detection system for agricultural soil based on laser-induced breakdown spectroscopy, used to perform the heavy metal detection method for agricultural soil based on laser-induced breakdown spectroscopy as described in claim 1, characterized in that, include: The first data acquisition module is used to collect temperature and humidity data, moisture content data, and corresponding batch information of the agricultural soil to be tested. The simulation module is used to simulate the process of exciting the agricultural soil to be tested to generate plasma based on the temperature and humidity data and the moisture content data using a digital twin simulation model and laser-induced breakdown spectroscopy, so as to generate target laser energy parameters and target acquisition parameters. The second acquisition module is used to acquire the optical signal of plasma radiation generated when the target laser beam acts on the agricultural soil to be tested. The emission parameters of the target laser beam are determined based on the target laser energy parameters and the target acquisition parameters. The correction module is used to separate the target band signal from the optical signal, correct the target band signal, and obtain the corrected band signal. The processing module is used to convert the corrected band signal into a digital signal, perform baseline correction and abnormal signal removal processing on the digital signal, and obtain a corrected digital signal. The generation module is used to extract a multi-band feature set from the corrected digital signal using a quantitative analysis model, perform cross-validation on the multi-band feature set to obtain associated feature information, and combine it with the detection batch information to generate the heavy metal detection results of the agricultural soil to be tested.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements the method for detecting heavy metals in agricultural soil based on laser-induced breakdown spectroscopy as described in any one of claims 1 to 6.

Images