CN-121027029-B - Trace naphthylacetic acid detection method and system based on terahertz spectrum detection technology

CN121027029BCN 121027029 BCN121027029 BCN 121027029BCN-121027029-B

Abstract

The invention provides a trace naphthalene acetic acid detection method and system based on a terahertz spectrum detection technology, wherein the method comprises the steps of determining target parameters of a target metamaterial sensor, and performing simulation based on the target parameters to obtain a target terahertz material; the method comprises the steps of obtaining target terahertz material, carrying out spectrum acquisition on a naphthylacetic acid sample to obtain target spectrum data, training an initial model by adopting a target feature selection method, carrying out dimension reduction treatment on the target spectrum data based on the trained initial model to obtain dimension reduction spectrum data, carrying out parameter optimization on model parameters of the initial model, carrying out optimization on the initial model based on the optimized parameters and the dimension reduction spectrum data to obtain a target detection model, and realizing trace naphthylacetic acid residue detection by the target detection model.

Inventors

HU JUN
MAO XIAODONG
XIAO ZICHENG
Shi Haoqi
CHEN YANG
XIAO GUOQING

Assignees

华东交通大学

Dates

Publication Date: 20260505
Application Date: 20251030

Claims (7)

1. The trace naphthylacetic acid detection method based on the terahertz spectrum detection technology is characterized by comprising the following steps of: Determining target parameters of a target metamaterial sensor based on the characteristic peak frequency of the naphthylacetic acid, and performing simulation based on the target parameters to obtain a target terahertz material; Preparing a naphthalene acetic acid sample, and performing spectrum acquisition on the naphthalene acetic acid sample through the target terahertz material to obtain target spectrum data; constructing an initial model, training the initial model by adopting a target feature selection method, and performing dimension reduction processing on the target spectrum data based on the trained initial model so as to obtain dimension reduction spectrum data; performing parameter optimization on model parameters of the initial model based on a target parameter optimizing method to obtain optimized parameters, and optimizing the initial model based on the optimized parameters and the dimension reduction spectrum data to obtain a target detection model; Residual detection of trace naphthylacetic acid is realized through the target detection model; The target terahertz material is provided with an L-shaped composite double-peak structure, wherein the target parameters comprise a period P 1 :58 mu m, a structure edge rectangle length L 1 :32 mu m, an L-shaped structure long side length L 2 :22 mu m, an L-shaped structure short side length L 3 :17 mu m, an interval L 4 :14 mu m between long sides of different L-shaped structures, an interval L 5 :4 mu m between short sides of different L-shaped structures, a structure edge rectangle width W 1 :2 mu m and an L-shaped structure width W 2 :4 mu m, silicon with a refractive index of 3.335 is selected as a substrate material of the target terahertz material, and a surface metal layer is gold; The step of training the initial model by adopting a target feature selection method and performing dimension reduction processing on the target spectrum data based on the trained initial model to obtain dimension reduction spectrum data comprises the following steps of: initializing the target spectrum data into a feature set, wherein the feature set comprises a plurality of feature subsets, and inputting the feature set into the initial model for prediction output to obtain a predicted value : ; In the formula, For the feature weights of the j-th feature subset, An ith feature of the jth feature subset, In order to be an intercept of the beam, Is the number of features in the feature subset; Sequentially inputting the feature subsets into an initial model for training, outputting feature weights of each feature in each training round, arranging the features in the feature subsets in a descending order according to the absolute values of the feature weights, and eliminating the features with the lowest absolute values of the feature weights to obtain a new feature subset; Re-inputting the new feature subset into an initial model, and repeatedly executing the processes of feature weight output and feature elimination until the condition of the dimension reduction requirement so as to obtain dimension reduction spectrum data; the step of optimizing the model parameters of the initial model based on the target parameter optimizing method to obtain optimized parameters, and optimizing the initial model based on the optimized parameters and the dimension reduction spectrum data to obtain a target detection model comprises the following steps: initializing model parameters of the initial model to a target population, and initializing each individual in the target population: ; In the formula, In order for the individual to be initialized, 、 The variables are respectively the lower bound and the upper bound, Is a uniform random vector; simulating periodic deviation behaviors of oat stems under wind power drive through the initial individuals and guiding each individual to swing towards the current optimal solution: ; In the formula, As the oscillation amplitude control factor, For the frequency of oscillation, In the event of a phase disturbance, For the current globally optimal individual, 、 The individuals after the t and t+1 times of iteration are respectively; Increasing disturbance update on an individual basis after the t+1st iteration To obtain updated individuals: ; In the formula, 、 Two different individuals selected at random, Is a disturbance regulating factor; in the iterative process, carrying out boundary correction on each updated individual, comparing the fitness of each updated individual before and after iteration, and reserving the individual with smaller fitness; and repeatedly executing the iterative process until the iterative stopping condition is met, outputting an optimal individual to obtain an optimization parameter, and optimizing the initial model based on the optimization parameter and the dimensionality reduction spectrum data to obtain a target detection model.
2. The method for detecting trace amounts of naphthalene acetic acid based on terahertz spectrum detection technology as claimed in claim 1, wherein the step of determining target parameters of the target metamaterial sensor based on characteristic peak frequencies of naphthalene acetic acid, and performing simulation based on the target parameters to obtain the target terahertz material comprises: acquiring the characteristic peak frequency of the naphthalene acetic acid, and determining a target peak frequency point of a target metamaterial sensor according to the characteristic peak frequency of the naphthalene acetic acid; Calculating the quality factor and the sensitivity of a target terahertz material through a simulation spectrum, and carrying out parameter scanning on a preset database according to the target peak frequency point, the quality factor and the sensitivity to obtain target parameters of a target metamaterial sensor; and selecting a single periodic structure as a modeling object, and performing modeling simulation according to the target parameters to obtain the target terahertz material.
3. The method for detecting trace amounts of naphthylacetic acid based on terahertz spectrum detection technique according to claim 2, wherein the target peak frequency points include 2.19 THz and 2.70 THz.
4. The method for detecting trace amounts of naphthalene acetic acid based on terahertz spectrum detection technique as claimed in claim 1, wherein the step of preparing a naphthalene acetic acid sample, and performing spectrum acquisition on the naphthalene acetic acid sample through the target terahertz material to obtain target spectrum data comprises: The method comprises the steps of selecting NAA standard solution as reference solution, quantitatively diluting the reference solution by using high-precision pipettor to obtain ultrapure deionized water, mixing each sample for 3 minutes at a rotating speed of 3000rpm through a vortex oscillator to ensure that solute molecules are uniformly distributed in the solvent, transferring a plurality of groups of obtained concentration gradient samples to a brown volumetric flask for sealing and storing, continuously introducing dry air into an optical cavity of a spectrometer through an air compressor in advance, monitoring through a temperature and humidity sensor to ensure that the relative humidity in the cavity is less than or equal to 10%, stabilizing the temperature within a range of 25+/-0.5 ℃, sequentially quantitatively absorbing concentration gradient samples with different concentration gradients from low concentration to high concentration by using a pipetting gun according to 20 mu l/time after the environment is stable, and dripping the concentration gradient samples onto the surface of the metamaterial sensor by adopting a compound sampling mode of repeated measurement of 5 points multiplied by 10 times to obtain target spectrum data.
5. A trace amount naphthalene acetic acid detecting system based on terahertz spectrum detecting technology, which adopts the trace amount naphthalene acetic acid detecting method based on terahertz spectrum detecting technology as claimed in claim 1, characterized in that the system comprises: the material module is used for determining target parameters of the target metamaterial sensor based on the characteristic peak frequency of the naphthylacetic acid, and performing simulation based on the target parameters to obtain a target terahertz material; The acquisition module is used for preparing a naphthalene acetic acid sample, and carrying out spectrum acquisition on the naphthalene acetic acid sample through the target terahertz material so as to obtain target spectrum data; The dimension reduction module is used for constructing an initial model, training the initial model by adopting a target feature selection method, and carrying out dimension reduction processing on the target spectrum data based on the trained initial model so as to obtain dimension reduction spectrum data; The optimization module is used for carrying out parameter optimization on the model parameters of the initial model based on a target parameter optimization method to obtain optimization parameters, and optimizing the initial model based on the optimization parameters and the dimension reduction spectrum data to obtain a target detection model; And the detection module is used for realizing the residual detection of trace naphthylacetic acid through the target detection model.
6. A computer comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method for detecting trace amounts of naphthylacetic acid based on the terahertz spectrum detection technique as claimed in any one of claims 1 to 4 when executing the computer program.
7. A storage medium having stored thereon a computer program which, when executed by a processor, implements the method for detecting trace amounts of naphthylacetic acid based on the terahertz spectrum detection technique according to any one of claims 1 to 4.

Description

Trace naphthylacetic acid detection method and system based on terahertz spectrum detection technology Technical Field The invention belongs to the technical field of pesticide residue detection, and particularly relates to a trace naphthalene acetic acid detection method and system based on a terahertz spectrum detection technology. Background Naphthalene acetic acid (1-NAPHTHALENEACETIC ACID, NAA) is an artificially synthesized plant growth regulator, and is widely applied to agricultural production, and is mainly used for promoting plant growth, regulating physiological metabolic process and improving crop yield. NAA belongs to auxin substances, and has the action mechanism of promoting plant cell elongation and inducing adventitious root formation and delaying fruit shedding to a certain extent mainly by simulating the function of natural auxin (IAA). In fruit tree cultivation, NAA is often used for protecting fruits, promoting rooting of cutting seedlings, prolonging harvesting period and the like, and is beneficial to improving commodity rate and storability of crops. However, NAA is used as an artificially synthesized plant hormone, and if the NAA is improperly used or the application dosage is too large, residues are easily generated on the surfaces or tissues of fruits and vegetables, and the NAA can affect the health of a human body after being eaten for a long time. Although NAA is relatively low in toxicity, studies have shown that it may possess potential reproductive toxicity, endocrine disrupting effects, and adverse effects on liver and kidney function. Therefore, many countries and regions make strict regulations on the maximum residual limit of NAA in agricultural products to ensure food safety. For example, the maximum residual quantity of NAA in apples is 0.1 mg/kg, which is specified in GB 2763-2021 Standard for maximum residual quantity of pesticides in foods in China. Since NAA is usually present at low concentrations on the surface of fruits and vegetables, its detection faces technical challenges of insufficient sensitivity and selectivity. Therefore, the NAA residue detection technology which is simple, quick, accurate and high in sensitivity is developed, and has important practical significance for guaranteeing food safety and public health. Terahertz waves generally refer to electromagnetic waves with frequencies ranging from 0.1 to 10 THz, are located between microwaves and infrared, and are not yet fully developed and utilized in the electromagnetic spectrum, and are therefore also referred to as "terahertz blank areas" or "junctions of electronics and photonics". With the development of material science, microelectronic technology and laser devices, terahertz technology has been rapidly developed in recent years, and has demonstrated great application potential in the fields of nondestructive testing, biological imaging, security inspection, communication, spectroscopic analysis, and the like. The method has the remarkable advantages in the characteristics of high sensitivity, non-contact, strong penetrability and the like, and becomes one of the important research directions of the current front-edge crossing technology. Compared with visible light and infrared light, the terahertz wave has good penetrability to nonpolar materials (such as plastics, paper, wood, ceramics, fabrics and the like), and can realize imaging and detection of the internal structure of nonmetallic or low-moisture materials. Therefore, the terahertz technology has unique advantages in nondestructive testing, safety inspection, cultural relic protection and the like. However, the pesticide residue tends to be very trace, and when the trace pesticide residue is detected by the traditional terahertz spectrum technology, the terahertz wave and the substance have weak interaction, so that the signal is weak, and the detection sensitivity is not high enough. In addition, terahertz spectra are sensitive to moisture and other biological components, which may introduce interfering signals that affect accurate identification of trace species. Thus, although terahertz spectra show good resolution in the detection of some samples, in the case of trace species, the resolution and selectivity of conventional terahertz techniques still have certain limitations. Therefore, for the detection of trace species, conventional terahertz spectroscopy often needs to incorporate other techniques to improve its sensitivity and accuracy. Metamaterials are a class of artificial materials that achieve specific electromagnetic responses by artificially designing and periodically building microstructure elements, whose characteristics are not derived from the intrinsic properties of the constituent materials, but rather are determined by their geometry and arrangement. Compared with natural materials, the metamaterial can show a series of unconventional optical phenomena such as negative refractive index, superlens effect, electrom