High precision and fast classification of different dimensions of Baijiu using an OptGSCV quadratic optimization network combined with AS-LIBS

Abstract

Aiming to tackle the problems of low classification accuracy, difficult data processing, long analysis time, and the influence of subjective consciousness of traditional Baijiu (Chinese liquor), this paper reports a method based on the combination of laser-induced breakdown spectroscopy with high-frequency ultrasonic atomization system (AS-LIBS) and optimized grid search cross validation (Opt-GSCV) for high-precision and rapid classification of Baijiu of different dimensions. Multi-dimensional high-precision classification of 23 Baijiu samples with different series, degrees, and production dates from Shanxi Xinghuacun Fen Wine Factory Co., Ltd was performed. By studying the growth and evolution mechanism of plasma generated by laser excitation, a new method of Baijiu spectral signal acquisition was realized by combining LIBS with Baijiu detection. Aiming at addressing the shortcomings of high breakdown thresholds and poor signal stability due to the direct breakdown of Baijiu with high ethanol content, a high-frequency atomization system based on ultrasonic technology is proposed to realize the efficient extraction of Baijiu plasma spectral signals. With regard to the selection of the number of nodes, weights and thresholds of the implicit layer of the back-propagation (BP) network, an optimization model based on the combination of the particle swarm optimization (PSO) algorithm combined with the genetic algorithm (GA) is proposed to realize the construction of the BP optimal network architecture. By constructing a secondary optimization seeking network model based on OptGSCV, the technical problem of mismatch between optimal parameter selection and optimal network construction is solved, and a new network integrating an intelligent qualitative analysis algorithm and a traditional classification algorithm is realized. The experimental results show that the proposed OptGSCV quadratic optimization network combined with the AS-LIBS multidimensional Baijiu high-precision classification model not only introduces an innovative and high-efficiency detection method for the Baijiu industry, but also provides important technical support and new ideas for the key areas of food safety production and Baijiu quality control.

---

1. Introduction

With the acceleration of globalization and consumers' pursuit of high quality of life, food safety and product authenticity have become a focus of social attention.¹ In the field of food science, especially in the study of Chinese Baijiu, which has deep cultural and economic value, accurate quality testing and classification have become increasingly important topics.² As one of the world's six largest distilled alcoholic beverages, Baijiu is not only an alcoholic beverage, but also an important part of traditional Chinese culture, carrying deep historical and cultural value and unique technology. Unlike other alcoholic beverages, although the main components of Baijiu are water and alcohol, it also contains more than 300 organic compounds such as esters, acids, aldehydes, phenols, etc.^3,4 Although the volume fraction of these compounds is less than 2%, they greatly increase the complexity and diversity of Baijiu. Therefore, a precise classification of Baijiu from a scientific point of view not only enhances our understanding of its complexity, but also helps protect the interest of consumers and industry. Especially for a brand with a long history and cultural heritage like Fenjiu, it is important to understand the uniqueness and value of its different varieties.^5,6 The economic value of Fenjiu is to a large extent closely related to its production process, choice of raw materials, vintage of brewing and storage conditions. These factors together determine the quality of Fenjiu, which in turn affects its market value. Therefore, there is an urgent need for an advanced technology that can quickly and accurately differentiate different Fenjiu varieties, as well as effectively assess their vintage and authenticity, to be applied in the industrial production of liquor and its circulation.

At present, researchers have more methods to test and analyze the quality of alcoholic beverages, but due to the lower specific heat capacity of ethanol in Baijiu, so that in the transformation of liquid molecules into gas molecules do not need to absorb too much heat, and ethanol molecules of small molecular weight, the molecular spacing between molecules is large, and therefore can be relatively easy to detach from the surface of the liquid into the gaseous state, and these reasons determines the nature of the ethanol volatility. As the ethanol content of Baijiu is higher than that of other alcoholic beverages (wine, whisky, etc.), it makes Baijiu analysis methods more complex.^7–9 At present, Baijiu quality testing methods mainly include sensory oral evaluation rating method,¹⁰ inductively coupled plasma mass spectrometry (ICP-MS),^11,12 phase liquid chromatography (PLC),¹³ gas-phase ion mobility spectrometry (GIMS),¹⁴ and bionic identification method based on the electronic nose, electronic tongue.¹⁵ As one of the more common methods for determining the authenticity of Baijiu in daily production and life, the sensory taste rating method has a high degree of detection flexibility and can directly and quickly reflect the comprehensive quality of Baijiu, including color, smell, taste and other aspects, and can capture the subtle differences in Baijiu. However, the results of this method mainly depend on the experience of the taster, are subjective, and are easily affected by the external environment and the taster's personal preferences and other factors, and it is difficult to quantitatively describe the different levels of the original Baijiu, and its accuracy fluctuates greatly.¹⁶ ICP-MS technology can accurately analyze the content of trace elements in the Baijiu, and it has a higher analytical precision, but also has the advantages of fast detection speed and the determination of a variety of elements. It also has the advantages of fast detection speed and determination of multiple elements. But in the detection process there is a need to add a certain amount of fluorescent substances to the Baijiu so that the sample is contaminated. Not only this, the method also requires professional equipment and technical support and the detection cost is higher.¹⁷ PLC technology provides fast and accurate analysis of organic substances such as esters, aldehydes, ketones and acids in Baijiu. However, the sensitivity is low, and the ability to analyze complex samples is limited.¹⁸ GIMS technology has the advantages of speed and sensitivity and is suitable for the detection of volatile organic compounds. However, for Baijiu, which has large sugar content and is not easy to volatilize, there are limitations in its detection ability.¹⁹ Electronic nose and electronic tongue based on some of the bionic identification methods belong to the sensory simulation technology and can quickly identify the varieties and origins of Baijiu and analyze faster. However, the technology used in the sensor is non-specific mainly because the higher degree of similarity of the mixture makes it difficult to distinguish correctly, and the detection cost is higher. Therefore, the above means of detection in practical applications have certain limitations.

Laser induced breakdown spectroscopy (LIBS)^20,21 is a plasma atomic emission spectroscopy technique based on high power pulses, which utilizes an intense pulsed laser focused on the surface of the target sample to generate a plasma, realizing the quantitative and qualitative analysis of the sample composition. As an emerging means of chemical analysis, LIBS not only has the advantages of rapid, multi-element detection, but it also allows non-contact in situ detection,²² which opens up new possibilities for real-time on-line detection, and it has been widely used in many fields, such as food safety,^23,24 substance detection,²⁵ geological exploration,²⁶ and Mars exploration.^27,28 In recent years, most of the chemometric methods for LIBS spectral data analysis have combined advanced machine learning models based on multivariate statistical techniques and various intelligent optimization algorithms.²⁹ Xinmeng Luo et al.³⁰ used a particle swarm optimized (PSO) back-propagation (BP) neural network combined with LIBS data to achieve rapid detection of heavy metal content in Pinus sylvestris. Jie Ren et al.³¹ optimized the BP network using genetic algorithm (GA) and combined it with PSO model to achieve the detection of soil Cd content under double-pulse LIBS.

The above examples show that advanced machine learning algorithmic modeling of LIBS experimental data can reduce and correct the spectral intensity fluctuations caused by emission source noise and matrix effects, thus improving the measurement accuracy.³² In this regard, this paper designs a multi-dimensional Baijiu high-precision characterization system based on an optimized grid search and cross-validation (OptGSCV) quadratic optimization network combined with LIBS based on high-frequency ultrasonic atomization system (AS-LIBS), which achieves high-precision real-time on-line inspection of the wavelength range of the internal plasma of the Baijiu, the fluid properties, and the compositional constituents.

• By studying the growth and evolution mechanism of plasma generated by laser excitation, LIBS combines the advantages of fast analysis rate, on-line in situ detection, no sample pretreatment, and micro-destructive detection of samples to be tested with traditional Baijiu detection technology and creates a new method for the detection and analysis of trace elements in Baijiu.

• Independent design and development of a high-frequency atomization system based on ultrasonic technology, breaking through the Baijiu due to high ethanol content, and plasma spectroscopy signal acquisition distortion, to achieve the Baijiu characteristics of plasma spectroscopy signal acquisition with high efficiency.

• An intelligent optimization qualitative analysis algorithm with hyperplane segmentation constraints and a quadratic optimization network model based on OptGSCV are innovatively proposed. It realizes the individual optimization of each qualitative analysis index and the overall optimization and solves the technical problem of the mismatch between the selection of optimal parameters and the construction of an optimal network.

• The breakthrough proposes a new method of multi-dimensional high-precision rapid detection and qualitative analysis of Baijiu using a machine learning optimization model combined with AS-LIBS, which realizes high-precision online in situ analysis of Baijiu with multiple dimensions identified at one time.

2. Experimental part

2.1 Experimental system setup

The multidimensional Baijiu characterization system based on a high-frequency ultrasonic atomization system called AS-LIBS is jointly composed of LIBS spectral detection module, high-frequency ultrasonic atomization auxiliary module, spectral data preprocessing module and characterization network optimization module, as shown in Fig. 1. The LIBS spectral detection module consists of a laser, a spectrometer, an ICCD camera, an optical module and electronic controller module. The laser is a Nd:YAG nanosecond laser (LItron Lasers Ltd., Litron Nano LG, UK) with an operating wavelength of 1064 nm, a pulse duration of 8 ns, a repetition frequency of 10 Hz, and an output pulse energy range of 0–270 mJ. The dual-channel spectrometer employs two spectrometers with different characteristics to cover a wider spectral range and ensure the accuracy and comprehensiveness of the experimental data. One spectrometer is an Avantes ULS2048L designed for high-sensitivity measurements under low-light conditions, with a spectral range of 180–700 nm, and the other is an Avantes ULS4096L with a spectral range of 700–900 nm. In the experiment, by fine-tuning the parameters of each spectrometer, we were able to ensure that the data from both could be seamlessly integrated to maximize the performance advantages of each. The ICCD camera delay is adjustable from 300 ns to 1500 ns, and the integration gate width is adjustable from 900 ns to 30000 ns. The self-designed ultrasonic atomization auxiliary device consists of a piezoelectric ceramic oscillator with a resonance frequency of 1.2 ± 0.05 MHz, an atomization volume of 400 ml h⁻¹, a spherical air-carrying device, an airflow stabilizer, and a DC 12 V power supply module. At the beginning of the experiment, the Baijiu sample was poured into the columnar sample reaction chamber to completely submerge the piezoelectric ceramic oscillator. When the instrument is activated, the sample is atomized into droplet aerosol particles, and the vapor enters into the spherical air carrier device through the rubber air guide tube. Through the cushioning effect of the spherical air carrier device, a uniform and stable column of Baijiu vapor is formed at the outlet of the atomization system. After the vapor column is stabilized, the nanosecond pulse laser is turned on for Baijiu spectral data acquisition. In the electronic timing controller under the action of the laser sequentially through the optical isolator and focusing optical lens, focusing on the ball-shaped gas carrier device at the exit position, on the stabilization of the Baijiu vapor column to break through. Finally, the Baijiu plasma spectral signal is coupled into an optical fiber through a beam collection lens and transmitted to a dual-channel fiber optic spectrometer. The ICCD camera at the probe of the dual-channel fiber optic spectrometer collects the plasma breakdown image in real time and finally transmits the plasma spectral signal and plasma breakdown image to a computer. The data analysis and processing module consists of a ChemLogix-based NIST³³ spectral database and Matlab software (Matlab®, v R2020a, the Mathworks, Inc. Natick, USA), which enables the selection and construction of a quadratic optimized network architecture. In the team's previous work, the optimization of the optical path of the AS-LIBS system was achieved by adjusting the position of the spectral signal collection optical path, comparing the intensity and enhancement factor of the Baijiu plasma spectra under the lateral optical path and the rearward optical path, and finally determining to adopt the lateral path as the experimental optical path of the detection system.

Fig. 1 Diagram of a multi-dimensional Baijiu authentication system based on a high-frequency ultrasonic atomization system.

Fenjiu, a typical representative of clear-flavored Baijiu, was chosen as the research object based on the characteristics of the local economy and people's health.^34,35 The 23 Fenjiu samples used in the experiment were provided by Shanxi Xinghuacun Fenjiu Factory Co., Ltd.,³⁶ which guaranteed the authority of the samples and the authenticity of the experimental data. The samples were selected to cover all the best-selling products of Fenjiu, including 10 different series, 14 different brands and 7 different degrees. In order to investigate the effects of different production dates on the differences in trace element contents of Fenjiu, 10 different years of red-capped glass Fenjiu from 2013 to 2023 were selected as the representatives of the experiment. Table 1 lists the differences in the series to which the different series belong, the differences in alcohol content, and the differences in the production dates of the different samples.

Table 1 Experimental samples of Baijiu with different series, different brand names, different alcohol contents and different production dates

Belonging series	Brand name	Alcohol content	Production date	Sample
Glass Fenjiu series	Glass Fenjiu with a red lid	42°	October 20, 2013	Sample 1
	Glass Fenjiu with a red lid	42°	November 10, 2014	Sample 2
	Glass Fenjiu with a red lid	42°	June 15, 2015	Sample 3
	Glass Fenjiu with a red lid	42°	August 8, 2016	Sample 4
	Glass Fenjiu with a red lid	42°	June 16, 2017	Sample 5
	Glass Fenjiu with a red lid	42°	July 7, 2018	Sample 6
	Glass Fenjiu with a red lid	42°	June 30, 2020	Sample 7
	Glass Fenjiu with a red lid	42°	August 28, 2021	Sample 8
	Glass Fenjiu with a red lid	42°	May 19, 2022	Sample 9
	Glass Fenjiu with a red lid	42°	March 18, 2023	Sample 10
	Yellow cap glass Fenjiu	53°	September 28, 2020	Sample 11
Fenyang King series	Fenyang King 10	45°	June 19, 2021	Sample 12
	Fenyang King 25	50°	October 24, 2022	Sample 13
Export Fenjiu series	Opalescent glass Fenjiu	48°	March 4, 2019	Sample 14
	Export of ceramic Fenjiu	53°	September 14, 2018	Sample 15
Blue and white porcelain Fenjiu series	Blue and white porcelain Fenjiu 20	53°	March 17, 2022	Sample 16
	Blue and white porcelain Fenjiu 30	53°	April 18, 2021	Sample 17
Bamboo leaf green Fenjiu series	Bamboo leaf green Fenjiu	38°	May 11, 2020	Sample 18
Rose Fenjiu series	Rose Fenjiu	40°	August 21, 2019	Sample 19
White jade Fenjiu series	White jade Fenjiu	40°	December 2, 2019	Sample 20
Panama Fenjiu series	Panama 1915 Black tan Fenjiu	42°	December 2, 2022	Sample 21
Old white Fenjiu series	Old white Fenjiu 10	53°	July 7, 2022	Sample 22
Base liquor	Fenjiu base Baijiu	65°	January 20, 2020	Sample 23

---

Experimental samples were taken directly from unopened bottles to avoid the influence of airborne microbial populations on the experimental data. Each LIBS spectrum was obtained from a single test. For 23 different Fenjiu, each sample was sampled six times, and 150 ml of each sample was placed in an ultrasonic atomization system. In order to obtain stable spectral signals, the LIBS detection system was run after a stable gas flow column appeared at the outlet of the spherical carrier gas device for 2 min. The laser pulse energy was set at 50 mJ and the repetition frequency at 10 Hz. The laser was focused through a 300 mm focusing lens to the outlet of the nebulizer to break down the aerosol of the Fenjiu droplets, and the signals were coupled to the spectrometer through the laser probe and the optical fiber (core diameter of 1000 mm and a numerical aperture of 0.22). In order to prevent the effect of toughness radiation, a digital delay generator (Stanford model DG535) was utilized to trigger the detector with a delay of 1 μs between the laser pulse and the collected plasma radiation. 100 spectral datasets were collected for each experiment, and a total of 13 800 (23 × 6 × 100) 7846-dimensional Fenjiu LIBS spectral datasets were collected for the 23 samples. To ensure the credibility of the experimental results, 70% of the Fenjiu spectral data (9660) were randomly selected as the training set, 15% as the validation set (2070), and 15% as the test set (2070). Each spectral dataset is independent of each other and consists of different kinds of Fenjiu spectral signals randomly.

2.2 Methodology for data analysis

2.2.1 PSO external optimization. PSO is an optimization-type algorithm based on group intelligence, which is usually used to solve the problem of parameter optimal search.³⁷ On the basis of behavioral observation of animal cluster activities, the algorithm makes use of the sharing of information by individuals in the group, so that the movement of the whole group produces an evolution process from disorder to order in the problem solution space and thus finds the optimal solution. In PSO, each individual is considered to be a mutually independent particle, and each particle has an initial position and velocity. The initial position represents the current state of the solution and the velocity indicates the direction and speed of change of the current solution. The optimal solution is sought by constantly updating the position and velocity of the particles.

Assuming a particle swarm consisting of N particles in a D-dimensional target search space, where each particle is a D-dimensional vector, the spatial location of the ith particle can be expressed as eqn (1).

X_i = (x_i1, x_i2, …, x_iD), i = 1, 2, 3, …, N (1)

The spatial position of the particle is a solution to the objective optimization problem, which can be calculated by substituting it into the fitness function, and the merit of the particle is measured according to the size of the fitness value. The flight velocity of the ith particle can be expressed as eqn (2).

V_i = (v_i1, v_i2, …, v_iD), i = 1, 2, 3, …, N (2)

The particle's position and velocity averages are randomly generated within a given range. The update of the position and velocity of the i particle in generation t as it evolves to generation t + 1 can be represented by eqn (3) and (4).

X_ij(t + 1) = x_ij(t) + v_ij(t + 1) (3)

V_ij(t + 1) = ωv_ij(t) + C₁r₁(p_ij(t) − x_ij(t)) + C₂r₂[g_j(t) − x_ij(t)] (4)

where i denotes the ith particle, j denotes the jth dimension of the particle, t denotes the current number of iterations, p denotes the position experienced by the particle with the optimal fitness value known as the individual historical optimal position, g denotes the optimal position experienced by the whole swarm of particles known as the global historical optimal position, and r is a random number in the range of [0, 1]. ω is the inertia weight that determines the inertia of the particle's motion, which gives it a tendency to expand the search space. C₁ is the individual learning coefficient and C₂ is the global learning coefficient, and the two coefficients determine the weights of the statistical acceleration term of the particle moving towards the optimal position.

The general randomly generated initialized population makes the algorithm convergence assessment limited due to the uneven distribution of individuals, as shown in Fig. 2(a). So this paper chooses to introduce circle turbid mapping to the population for the initialization operation, as shown in Fig. 2(b), to improve the convergence speed of the algorithm while obtaining a more homogeneous and diverse initial population structure. The sample selection can be represented by eqn (5).

The num_i denotes the number of the ith turbid sequence, and mod(a, b) denotes the remainder operation of a to b. In order to make the particles fit the optimal point faster, an optimization scheme of adaptive inertia weights is proposed to make ω larger in the early stage to ensure the global search capability and smaller in the later stage to enhance the local search capability, as shown in Fig. 2(c). The step size is dynamically adjusted according to the change of weights to better adapt to different optimization processes and improve the convergence speed of the PSO algorithm while enhancing robustness and adaptability, as shown in Fig. 2(d). In this paper, the PSO algorithm is chosen to optimize the external structural parameters of the neural network and to optimize the number of nodes in the implicit layer by simulating the foraging behavior of the bird flock, so as to improve the performance of the model structure.

Fig. 2 PSO schematic diagram, (a) randomly generating inhomogeneous initialized populations, (b) introducing circle turbidity mapping to obtain homogeneous and diverse initialized population structure, (c) introducing adaptive inertia weights to improve the efficiency of model fitting and (d) optimal model fitting.

2.2.2 GA-BP internal optimization. Because of its good self-learning ability, BP is widely used in data processing, but the model is easy to fall into local optimization. The GA is good to make up for the defects of traditional BP because of its global optimality searching properties. So, the GA is usually considered an internal optimization function to achieve the selection of the optimal number of nodes, weights, and thresholds of the BP neural network through evaluation, selection, crossover, and mutation. The optimization principle is shown in Fig. 3. In this paper, the BP is internally optimized by the GA to find the optimal weights and thresholds to improve the performance of the model.

Fig. 3 Schematic diagram of the GA-BP principle.

2.2.3 OptGSCV secondary optimization. As a simple and convenient hyper-parameter search algorithm, GSCV is widely used in the optimization search of various datasets.³⁸ The method can not only realize the global optimal solution and determine the optimal combination of parameters effectively, but also enhance the overall matching degree of the model and improve the overall performance.

In view of the ability of the network search model to find the parameter with the highest accuracy within the specified parameter range, an optimized network search method (OptGSCV) is proposed to traverse each possible parameter combination involved with the qualitative analysis of the model's judgement index as the goal and the parameter range and search step size as constraints. By continuously shrinking the size and step size of the grid, the approximate location of the optimal parameter combination is quickly localized and gradually approached to that point. The method improves the efficiency of parameter optimization while avoiding the interference of local optimal solutions and realizes the quadratic parameter optimization of the network. In this paper, OptGSCV is utilized to perform secondary optimization of the hyperparameters of the model, which ensures the matching of the optimal network structure with the optimal parameters, thus further improving the model performance. The specific steps of OptGSCV are as follows:

• Spectral data are input into the module, the ranges of values of the two parameters to be searched, σ and λ, determined using the BP training function trainscg, are set [2^−M, 2^M], respectively, and the step size N of the initial search framework is set to obtain a coarse search network, where the nodes in the network are all the possible combinations of the parameters that can be obtained within the given range.

• The value of k is set in the k-fold cross-validation and assign all the parameters are assigned to the model, respectively, after the k-fold cross-validation method of the discriminative performance of the model evaluation, to find the parameter combination with the smallest mean squared error (σ_i, λ_j), and the k-fold cross-validation model diagram is shown in Fig. 4.

Fig. 4 k-Fold cross-validation.

• The network between four nodes (σ_i+N, λ_j+N), (σ_i+N, λ_j−N), (σ_i−N, λ_j−N) and (σ_i−N, λ_j+N) around the parameter combination (σ_i, λ_j) is selected as a new search range and the search step size N/2 is set to build a fine-grained network structure, which is again cross-validated to find a new parameter combination (σ_m, λ_n) with the smallest mean square error.

• If the analytical metrics obtained from the parameter combination (σ_m, λ_n) in the k-fold cross validation have met the experimental requirements, then (σ_m, λ_n) is stored into the PSO-GABP model and the analytical metrics are output in the form of a table. If it does not meet the requirements, then return to the third step and the parameters are optimized again until the accuracy requirements are met.

We use the confusion matrix to calculate the accuracy, precision, recall, and F1 value and then use these four metrics to evaluate the effectiveness of the classifier's performance. Accuracy is defined as the ratio of the number of correctly classified samples to the total number of samples, as shown in eqn (6), where TP is the correct prediction of the positive class, TN is the correct prediction of the negative class, FP is the false prediction of the positive class, and FN is the false prediction of the negative class.

Precision is defined as the ratio of the number of samples correctly predicted to be in the positive category to the number of samples predicted to be in the positive category and is usually used to indicate the degree of prediction of the correct sample results. Its formula is shown in eqn (7).

Recall is the ratio of the number of correct samples retrieved to the number of all correct samples in the database and is used to measure the ability of the classifier to recognize correct samples. Its formula is shown in eqn (8).

The F1 value is the reconciled average of precision and recall and is commonly used to evaluate the performance of machine learning models on unbalanced datasets. Its formula is shown in eqn (9).

3. Results and analysis

3.1 AS-LIBS spectral line analysis and feature extraction

Water and ethanol in Baijiu account for 98–99% of the total mass, and more than 300 organic compounds such as esters, acids, aldehydes, phenols, etc. account for less than 2% of the volume fraction.³⁹ It is the presence of a small number of trace elements to promote the aging of the Baijiu body, so that the Baijiu in the taste and smell of the differences between the formation of different Baijiu between the sensory characteristics of each style.⁴⁰ The metallic trace elements in Baijiu, such as Ca, K, Ni, Fe, Mn, etc. determine the taste of Baijiu.

According to Fig. 5, small amounts of metallic trace elements such as Mn, Zn, Cu, Fe, Ca and Mg exist in Baijiu, and although their contents are very small, their absence determines the quality and taste of Baijiu. These trace elements are involved in the aging process of the Baijiu in the vat, which promotes the maturation and aging of the Baijiu. Mn, Cu, Fe and other elements are involved in the redox reaction of the Baijiu, which promotes the evolution of the organic matter in the Baijiu, and makes each Baijiu have unique sensory characteristics. Not only this, the presence of trace elements increases the chemical complexity of Baijiu, making it richer and more diverse. Their interaction with organic components produces a variety of chemical reactions that form the unique aroma and flavor of the Baijiu.⁴¹ Therefore, the plasma spectrometry technique is used to determine the content of trace elements in the Baijiu body, so as to identify the different series, degrees and production dates of Fenjiu, and its analytical accuracy and precision are reliable.

Fig. 5 LIBS spectral signal plots of 23 different Fenjiu experimental samples.

Since the experiments were carried out in bare air, there are some high contents of C, O, N, and H in the full spectral information of Baijiu. Because the electronic excitation energy level of C is high, the plasma spectral lines of C are usually in the shorter wavelength range. The plasma spectral lines of H are distributed in a longer wavelength range because the ground state electron energy level of H is lower. In addition to this, N and O are more abundant in the visible wavelength band of 700–900 nm. The existence of these elements will not only affect the other spectral data with noise, but also lead to an increase in the subsequent calculations wasting a lot of time. So it is very necessary to perform feature extraction on the full-spectrum signal on the basis of keeping the main features. Considering the small differences in elemental composition and high similarity of spectral information between different Fenjiu from the same manufacturer, the principal component analysis (PCA) method was chosen to be used for feature extraction of the full spectrum information.

In order to avoid obvious order-of-magnitude differences in the input variables, these signals are min–max normalized, and the calculation formula is shown in eqn (10), where max{x_j} is the maximum value of the sample data and min{x_j} is the minimum value of the sample data. Min–max standardization is a linear transformation of the original data so that the resultant values are mapped between [0, 1], eliminating the effects of the variables and the range of variation, while preserving the relationship between the data to the greatest extent possible.

By constructing the cumulative contribution curves of different principal components, the number of principal components that can be used for subsequent analysis is selected. In this paper, we adopt the method based on the change amplitude of the cumulative contribution curve in selecting the number of principal components, i.e., we observe the change trend of the cumulative contribution curve, look for an “inflection point” that represents the significant change of the growth rate of the cumulative contribution rate with an increase in the number of principal components, and select the number of principal components at this inflection point as the final number of principal components. The number of principal components at this inflection point is chosen as the final number of principal components. Fig. 6 shows the cumulative contribution curves of the first 10 principal components after PCA treatment, and the contribution rates of the first 6 PCs are 46.5%, 29.89%, 10.04%, 2.17%, 0.75%, and 0.66%, respectively, and the cumulative contribution rate reaches 90.01%, which indicates that the first PCs carry most of the information of the original variables, so the 6 PCs are selected as the representative of the entire sample variable representing the characteristics of the whole sample for subsequent analysis.

Fig. 6 Contributions and cumulative contribution rates of principal components.

3.2 OptGSCV-PSO-GA-BP

Each optimization model has its own unique algorithmic principles and application scenarios. The team previously used the GA model to optimize the BP network and evaluate the model recognition rate corresponding to different numbers of hidden layer neurons to achieve the qualitative analysis of Baijiu by constructing the optimal architecture. In this paper, on the basis of the research of the group, we explore the ability of the emerging bionic intelligence optimization model to find the optimal overall network architecture, as well as the effect of the optimized GSCV model on the overall performance of the model by hyperparameter secondary optimization to improve the overall performance of the model. Therefore, in this paper, we construct a quadratic optimized PSO-GABP network model based on OptGSCV to process the AS-LIBS Fenjiu spectral data in order to improve the performance of the model for qualitative analysis and to realize the on-line in situ high precision classification study of Fenjiu samples in a short period of time. The construction of the OptGSCV-PSO-GABP network is divided into the following three main modules:

First, the PSO algorithm is used to simulate the foraging behavior of bird flocks and search for optimal solutions in the solution space in a disordered-to-ordered manner. Through the strategy of adaptive inertia weights, PSO dynamically adjusts the search direction and speed of each particle during the search process to find the optimal number of nodes of the hidden layer in the global solution space β. The randomness and regularity of the initial search population are enhanced by the circle turbid mapping to ensure the diversity of exploration.

Second, after the PSO has determined the more desirable network structure parameters, the network is further internally optimized using a GA. The GA screens the individuals (i.e., network configurations) in the initial population based on their performance (fitness) by mimicking the natural selection mechanism and generates a new generation of network configurations by crossover and mutation operations. This process is optimized for the weight matrix and bias terms of the BP neural network, aiming to improve the training effectiveness and generalization ability of the network.

Finally, the grid search algorithm with an optimization step is introduced to carry out the secondary optimization search for the two hyperparameters that affect the model evaluation performance, and the overall analysis performance of the model is evaluated by k-fold cross-validation with the model accuracy, precision, recall, and F1 value as reference indices, so as to achieve the optimal performance of the overall model under the optimization of a single parameter, and the model framework of OptGSCV-PSO-GA-BP is shown in Fig. 7.

Fig. 7 Overall model diagram of OptGSCV-PSO-GA-BP.

PSO is mainly used to optimize the external structural parameter of the neural network, i.e., the number of nodes in the hidden layer (β). We optimize the hidden layer parameter β as a free parameter to find the optimal network structure. This parameter does not limit the number of hidden layers or the number of nodes, but indicates the number of nodes in the whole hidden layer. PSO finds the optimal number of nodes in the hidden layer by simulating the foraging behavior of a flock of birds to optimize the performance of the whole network architecture. The GA is used for the optimization of weights and thresholds inside the neural network, which is achieved by evaluating, selecting, crossover, and mutating the weight matrix and bias terms of the network in order to obtain the optimal network architecture. On the premise of obtaining the optimal network architecture, OptGSCV is used to conduct secondary optimization search again for the two hyperparameters learning rate (σ) and regularization coefficient (λ), which have a greater impact on the performance of the model, to ensure that the optimal network architecture matches with the optimal parameters and to further improve the performance of the model. Fig. 7 presents the mismatch of network architectures with different colors, and the network model with different colors from the beginning is eventually optimized to a model with consistent colors to illustrate the optimal parameter selection and the construction of the optimal network architecture.

In order to classify the 23 samples with 13 800 × 6 different Fenjiu spectral data with high accuracy, the sample data were first Z-score normalized to reduce the potential adverse effects of anomalous samples on the model performance. In the process of optimizing the network structure, the PSO algorithm was used to dynamically adjust the number of nodes in the hidden layer. The initial number of particles is set to 30, the maximum number of iterations is 10, and 3 × 3 initialized parameters to be optimized are generated according to the boundary position, including the number of nodes in the implied layer β and the two hyper-parameters σ and λ of the quantized connected gradient BP training function trainscg. The number of nodes in the input layer 6 depends on the dimensions of the input data. For the three parameters to be optimized, the search ranges are set to be 1 < β < 20, 10⁻⁸ < σ < 10⁻², and 10⁻⁷ < λ < 10⁻², and the BP neural network is externally optimized using iterative adaptation as a metric.

In the initialization phase of PSO, we use eqn (11) to randomly generate a collection of candidate solutions, where E denotes the number of nodes in the implicit layer, U is the number of nodes in the input layer, B is the number of nodes in the output layer, and a is a random integer ranging from 1–20 to introduce diversity in the initial solutions. This equation is not a direct way to finalize the number of nodes, but rather provides a diverse starting point for the PSO algorithm. Subsequently, these initial solutions are evaluated and optimized based on the fitness function. The PSO operation boundaries are randomly generated according to the number of input and output layers, the maximum particle flight velocity V_max = 6, the individual learning coefficient C₁ = 2, the global learning coefficient C₂ = 2, the adaptive inertia weights ω range from 0.6–0.9, and the parameters a = 0.5 and b = 0.2 in the circle mapping in order to ensure the optimal number of nodes of the hidden layer after the PSO external optimization.

The BP neural network under specific conditions is constructed based on the optimal number of hidden layer nodes found by external optimization, followed by determining the optimal structure and hyperparameter settings of the BP by the GA. The number of populations is set to 50, the maximum number of iterations is 80, the variation rate is 0.1, the crossover rate is 0.2, the optimal weights corresponding to the network architecture are outputted to be 400, and the number of optimal thresholds is 29. Then, traincg is used as the optimizer of weights and biases to make fine adjustments, and the maximum number of training times of the network is set to be 200, and the learning rate is set to be 0.2, with the target error being 1 × 10⁻⁸, and the network is quickly and efficiently adjusted by using the conjugate gradient method for fast and effective weight adjustment of the network. traincg meticulously adjusts the weights and biases to minimize the error function based on the network structure and hyperparameters determined by the GA. Efficient network optimization is achieved through global search by the GA and local fine tuning by traincg to output the correctness of the training set and the trained neural network (net) at this point.

The optimal fitness 99.957 exists when β = 17 and σ = 0.0071, and λ = 0.0031 is finally selected through the stochastic search of the network and assigned to the BP neural network, at which time the network error can reach the required value of 1 × 10⁻⁸. The parameter optimization process of the model is carried out through several iterations, each of which evaluates the performance of the current parameter combination. With the PSO and GA algorithms, we were able to search efficiently over a wide parameter space, while OptGSCV further fine-tuned the parameters based on these optimizations. Eventually, the parameter combination with the best performance on the validation set is selected as the final parameters of the model. The optimal network architecture is output to the PSO, and random slicing is chosen to slice the dataset to prevent the large deviation between the model performance and the real metrics caused by the fixed slicing dataset. On the premise of finding the optimal network architecture with the model optimal fitness as the criterion, OptGSCV is used to conduct the secondary optimization search for the two hyperparameters σ and λ, which have a large influence. Setting the search range σ (10⁻⁵, 100) and λ (10⁻⁵, 100) the search step is set to 1 × 10⁻⁵. 10⁵ × 10⁵ hyperparameter combinations are searched and the search results are shown in Fig. 8.

Fig. 8 Results of OptGSCV's secondary optimization search for the two hyperparameters σ and λ.

After traversing 1 × 10¹⁰ different spatial search points, the optimal hyper-parameter combinations σ = 0.0076 and λ = 0.0083 corresponding to the optimal network architecture were found, and the qualitative analysis accuracy of the model at this time was 99.984%. In order to conduct a comprehensive assessment of the model to reduce the risk of model overfitting, the 10-fold cross-validation method was used to assess the generalization of the OptGSCV-PSO-GA-BP model in the qualitative analysis of Fenjiu's multidimensional data, and the model's qualitative analysis accuracy was used as the evaluation index. As shown in Fig. 9, the highest analysis accuracy of the model is 99.998% and the lowest is 99.945%, which further validates the generalization performance of the quadratic optimization model for high-precision qualitative analysis of Baijiu.

Fig. 9 10-fold cross-validation results.

The qualitative analysis performance of the optimal model is described by the confusion matrix as shown in Fig. 10 and 11. From the confusion matrix, it can be seen that the qualitative classification accuracy is 99.987% in the training set and 99.983% in the test set. The vast majority of the predicted and true labels of Fenjiu are the same, both in the test set and in the prediction set. There are 14 Fenjiu samples with prediction errors in the training set and 4 Fenjiu samples with prediction errors in the test set. In order to further illustrate the reasons for the prediction errors of a small number of Fenjiu samples, the accuracy, precision, recall, and F1 values corresponding to the 23 different Fenjiu samples in the training and test sets were respectively calculated, as shown in Table 2.

Fig. 10 Confusion matrix for the training set of 23 Baijiu samples.

Fig. 11 Confusion matrix for the test set of 23 Baijiu samples.

Table 2 Performance indices of different kinds of Baijiu after GSCV-PSO-GA-BP treatment

Sample	Data set	GSCV-PSO-GA-BP sample performance evaluation index (%)
		Accuracy	Precision	Recall	F1-Score
Sample 1	Training set	99.979	99.518	100	99.758
	Test set	99.903	98.824	98.824	98.824
Sample 2	Training set	99.990	100	99.762	99.881
	Test set	99.903	98.876	98.876	98.876
Sample 3	Training set	100	100	100	100
	Test set	100	100	100	100
Sample 4	Training set	100	100	100	100
	Test set	100	100	100	100
Sample 5	Training set	99.979	99.515	100	99.757
	Test set	100	100	100	100
Sample 6	Training set	99.990	100	99.758	99.879
	Test set	100	100	100	100
Sample 7	Training set	99.990	100	99.764	99.882
	Test set	100	100	100	100
Sample 8	Training set	99.990	100	99.761	99.881
	Test set	100	100	100	100
Sample 9	Training set	99.990	99.761	100	99.880
	Test set	100	100	100	100
Sample 10	Training set	99.979	100	99.531	99.765
	Test set	100	100	100	100
Sample 11	Training set	99.979	99.765	99.765	99.765
	Test set	99.952	98.837	100	99.415
Sample 12	Training set	99.990	99.768	100	99.884
	Test set	99.952	100	98.913	99.454
Sample 13	Training set	99.990	100	99.770	99.885
	Test set	100	100	100	100
Sample 14	Training set	100	100	100	100
	Test set	100	100	100	100
Sample 15	Training set	99.990	99.762	700	99.881
	Test set	100	100	100	100
Sample 16	Training set	99.979	99.518	100	99.758
	Test set	99.952	98.947	100	99.471
Sample 17	Training set	99.990	100	99.757	99.878
	Test set	100	100	100	100
Sample 18	Training set	100	100	100	100
	Test set	100	100	100	100
Sample 19	Training set	99.979	99.524	100	99.761
	Test set	99.952	100	99.010	99.502
Sample 20	Training set	99.979	100	99.523	99.761
	Test set	100	100	100	100
Sample 21	Training set	99.969	99.525	99.762	99.643
	Test set	100	100	100	100
Sample 22	Training set	99.979	100	99.525	99.762
	Test set	100	100	100	100
Sample 23	Training set	100	100	100	100
	Test set	100	100	100	100

---

The above results show that there is a certain connection between the 23 samples and the trace elements that determine the taste of Fenjiu of the same series, the brand and degree are the same, and the differences are small, which makes it difficult to guarantee the accuracy of the traditional Baijiu qualitative analysis technique. The combination of AS-LIBS and the OptGSCV quadratic optimization network with its high analytical performance provides a new data processing solution for multidimensional Fenjiu detection and provides a new data processing scheme. To further validate the analytical performance of the model, it was compared with some commonly used qualitative analysis models (KNN and SVM) and unoptimized BP models (BP, GA-BP, and PSO-GA-BP). Based on the team's previous research experience, k was set to be 8 in the KNN model, and a linear kernel function was selected for the SVM, with the parameters C = 0.5 and g = 0.005. The parameter settings for the BP and GA-BP models were the same as those in GSCV-GA-BP. The data after PCA dimensionality reduction was selected to be trained sequentially, and the results are shown in Table 3.

Table 3 Comparison of evaluation performance of different models

Data set	Performance evaluation index	Performance evaluation of different qualitative analysis models (%)
		KNN	SVM	BP	GA-BP	PSO-GA-BP	GSCV-PSO-GA-BP
Training set	Accuracy	80.238	78.665	84.523	90.507	98.809	99.987
	Precision	68.625	78.665	79.902	90.509	98.810	99.855
	Recall	73.958	78.663	85.000	90.508	98.808	99.855
	F1-Score	70.307	78.665	80.740	90.509	98.809	99.855
Test set	Accuracy	78.621	78.551	83.617	90.483	98.762	99.983
	Precision	65.214	78.049	80.549	89.863	98.547	99.807
	Recall	71.567	78.050	85.043	89.865	98.546	99.808
	F1-Score	68.328	78.049	81.064	89.864	98.547	99.807

---

The results show that the PSO-GA-BP network optimized based on the OptGSCV quadratic optimization network exhibits strong qualitative analysis capability in the multi-dimensional Fenjiu detection problem. Compared with the other five algorithms, its analytical performance is more significantly improved both on the training set and the test set, with the qualitative accuracy as high as 99.987% on the training set and 99.983% on the test set. The proposed model simultaneously realizes the identification of Fenjiu's different series, different degrees and different production dates in multiple dimensions at one time, which provides a new method for the high-precision and fast classification of multi-dimensional Baijiu.

4. Conclusion

This paper starts from the issues of new methods of Baijiu detection, safeguarding people's life and property safety, and takes the local economic characteristics as the main research goal. Fenjiu, the model of Chinese clear Baijiu, was selected as the research object, and a multi-dimensional Fenjiu high-precision rapid classification method based on the OptGSCV quadratic optimization network combined with AS-LIBS was proposed in response to the problems of low analytical precision, complex pre-processing, long identification time and large influence of subjective consciousness of traditional liquor detection technology. By combining chemometrics, quadratic optimization algorithms and other related technologies, it can realize high-precision composition detection and qualitative analysis of different series, brands, degrees and production dates of Fenjiu.

• Through the research of laser-induced plasma spectroscopy intensity and particle number coupling models, an LIBS online testing system is constructed to solve the problems of low precision of traditional Baijiu identification technology, complicated pre-processing, long analyzing time, and subjective consciousness of the sommelier. An elemental component detection method for Fenjiu was proposed using new generation LIBS technology to realize high-precision detection and analysis of Baijiu.

• By studying the volatility of ethanol, the main component of Fenjiu, and the plasma spectral enhancement scheme of high-frequency ultrasonic atomization, we constructed a plasma spectral signal detection system for Baijiu and put forward the AS-LIBS high-frequency atomization scheme, which realizes a new process of plasma signal detection for Baijiu.

• Through the study of intelligent qualitative classification algorithm implicit layer nodes, weights, and hyperparameter iterative model, the PSO intelligent optimization method is adopted to make the best collocation between the linkage parameters under the premise that each parameter of the qualitative categorization model reaches the optimal value, to solve the technical problem of mismatch between the selection of the optimal parameters and the construction of the optimal network and to realize the fusion of the intelligent qualitative classification algorithm and the new network of the traditional classification algorithm.

• Through the research of optimal intelligent qualitative analysis model parameter adaptation and the optimal parameter extraction method of OptGSCV, we constructed a multi-dimensional plasma spectral information database of Fenjiu, built a multi-dimensional detection and analysis system of Fenjiu based on the quadratic optimization network, solved the bottlenecks of the Baijiu testing industry, such as low precision, high difficulty, time-consuming, etc., and realized the new platform of on-line qualitative analysis of Baijiu in situ with high precision and high efficiency.

AS-LIBS combined with the OptGSCV-PSO-GABP Baijiu testing and analyzing system provides a brand new qualitative classification and authenticity identification method for the Baijiu production and identification industry and plays an important role in the fields of food safety, anti-counterfeiting detection, analysis and testing. In particular, it is of great significance in combating counterfeit and shoddy Baijiu and protecting the legitimate interest of consumers and enterprises and lays the foundation for realizing multi-dimensional high-precision component detection by major famous Baijiu enterprises and classification of counterfeit and shoddy Baijiu in the market.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This paper was supported by the National Natural Science Foundation of China (No. 52075504), the Shanxi Province key research and development Projects (No. 2023021501016), the Sponsored by the Fund for Shanxi ‘1331 Project’ Key Subject Construction, the State Key Laboratory of Quantum Optics and Quantum Optics Devices (No. KF202301), the Shanxi Key Laboratory of Advanced Semiconductor Optoelectronic Devices and System Integration (No. 2023SZKF11), and Postgraduate Scientific Research Innovation Project of Shanxi Province in 2023 (No.2023KY584) for financial support. Thanks to Shanxi Xinghuacun Fen Wine Factory Co., Ltd for their strong support of the experiment.

References

1. A. Shruti, N. Bage and P. Kar, Nanomaterials based sensors for analysis of food safety, Food Chem., 2023, 433, 137284,  DOI:10.1016/j.foodchem.2023.137284.
2. W. Dong, R. Guo, X. Sun, H. Li, M. Zhao, F. Zheng, J. Sun, M. Huang and J. Wu, Assessment of phthalate ester residues and distribution patterns in Baijiu raw materials and Baijiu, Food Chem., 2019, 283, 508–516,  DOI:10.1016/j.foodchem.2019.01.069.
3. X. Duan, Y. Li, H. Li and Y. Wu, Accurate ethanol determination in Chinese Baijiu based on red-emitted carbon quantum dots (CQDs) and a simple pH correction, Food Chem., 2023, 428, 136733,  DOI:10.1016/j.foodchem.2023.136733.
4. M. Fan, S. Yuan, L. Li, J. Zheng, D. Zhao, C. Wang, H. Wang, X. Liu and J. Liu, Application of Terpenoid Compounds in Food and Pharmaceutical Products, Fermentation, 2023, 9, 119,  DOI:10.3390/fermentation9020119.
5. A. Shen and G. A. Antonopoulos, ‘No banquet can do without liquor’: alcohol counterfeiting in the People's Republic of China, Trends Organ. Crime, 2017, 20, 273–295,  DOI:10.1007/s12117-016-9296-x.
6. Y. Wang, J. Liu, Y. Xiong, X. Liu and X. Wen, Food Fraud Vulnerability Assessment in the Chinese Baijiu Supply Chain, Foods, 2023, 12, 516,  DOI:10.3390/foods12030516.
7. S. Moncayo, J. Rosales, R. Lzquierdo-Hornillos, J. Anzano and J. Caceres, Classification of red wine based on its protected designation of origin (PDO) using Laser-induced Breakdown Spectroscopy (LIBS), Talanta, 2016, 158, 185–191,  DOI:10.1016/j.talanta.2016.05.059.
8. J. Bockova, A. Roldan, J. Yu and P. Veis, Potential use of surface-assisted LIBS or determination of strontium in wines, Appl. Opt., 2018, 57, 8272–8278,  DOI:10.1364/AO.57.008272.
9. J. Bockova, Y. Tian, H. Yin, N. Delepine-Gilon, Y. Chen, P. Veis and J. Yu, Determination of Metal Elements in Wine Using Laser-Induced Breakdown Spectroscopy (LIBS), Appl. Spectrosc., 2017, 71, 1750–1759,  DOI:10.1177/0003702817708337.
10. Y. He, S. Chen, K. Tang, M. Qian, X. Yu and Y. Xu, Sensory characterization of Baijiu pungency by combined time-intensity (TI) and temporal dominance of sensations (TDS), Food Res. Int., 2021, 147, 110493,  DOI:10.1016/j.foodres.2021.110493.
11. Q. Xiong, Y. Lin, W. Wu, J. Hu, Y. Li, K. Xu, X. Wu and X. Hou, Chemometric intraregional discrimination of Chinese liquors based on multi-element determination by ICP-MS and ICP-OES, Appl. Spectrosc. Rev., 2021, 56, 115–127,  DOI:10.1080/05704928.2020.1742729.
12. Y. Zhang, J. Gu, C. Ma, Y. Wu, L. Li, C. Zhu, H. Gao, Z. Yang, Y. Shang and C. Wang, Flavor classification and year prediction of Chinese Baijiu by time-resolved fluorescence, Appl. Opt., 2021, 60, 5480–5487,  DOI:10.1364/AO.424015.
13. T. Watanabe, A. Yamamoto, S. Nagai and S. Terabe, Micellar electrokinetic chromatography as an alternative to high-performance liquid chromatography for separation and determination of phenolic compounds in Japanese spirituous liquor, J. Chromatogr. A, 1998, 793, 409–413,  DOI:10.1016/S0021-9673(97)00931-X.
14. X. Xie, X. Liu, X. Zhang, F. Zheng, D. Yu, C. Li, S. Zheng, B. Chen, X. Liu, M. Ma and G. Xu, In-depth profiling of carboxyl compounds in Chinese Baijiu based on chemical derivatization and ultrahigh-performance liquid chromatography coupled to high-resolution mass spectrometry, Food Chem.: X, 2022, 15, 110440,  DOI:10.1016/j.fochx.2022.100440.
15. L. Vera, L. Acena, R. Boque, J. Guasch, M. Mestres and O. Busto, Application of an electronic tongue based on FT-MIR to emulate the gustative mouthfeel “tannin amount” in red wines, Anal. Bioanal. Chem., 2010, 397, 3043–3049,  DOI:10.1007/s00216-010-3852-z.
16. W. Cai, Y. Xue, Y. Wang, W. Wang, N. Shu, H. Zhao, F. Tang, X. Yang, Z. Guo and C. Shan, The Fungal Communities and Flavor Profiles in Different Types of High-Temperature Daqu as Revealed by High-Throughput Sequencing and Electronic Senses, Front. Microbiol., 2021, 12, 784651,  DOI:10.3389/fmicb.2021.784651.
17. B. Li, S. Chua, D. Yu, S. Chan and A. Li, Determination and Characterization of Gold Nanoparticles in Liquor Using Asymmetric Flow Field-Flow Fractionation Hyphenated with Inductively Coupled Plasma Mass Spectrometry, Molecules, 2024, 29, 248,  DOI:10.3390/molecules29010248.
18. Y. Li, S. Fan, A. Li, G. Liu, W. Lu, B. Yang, F. Wang, X. Zhang, X. Gao, Z. Lu, N. Su, G. Wang, Y. Liu, X. Ji, P. Xin, G. Li, D. Wang, F. Lu and Q. Zhong, Vintage analysis of Chinese Baijiu by GC and H-1 NMR combined with multivariable analysis, Food Chem., 2021, 360, 129937,  DOI:10.1016/j.foodchem.2021.129937.
19. E. I. Geană, C. T. Ciucure, C. Apetrei and V. Artem, Application of Spectroscopic UV-Vis and FT-IR Screening Techniques Coupled with Multivariate Statistical Analysis for Red Wine Authentication: Varietal and Vintage Year Discrimination, Molecules, 2019, 24, 4166,  DOI:10.3390/molecules24224166.
20. X. Hao, Y. Yang, X. Liu, R. Sun, Y. Liu and P. Sun, Atomic emission dual-spectrum thermometry for laser-induced Cu plasma temperature, Optik, 2021, 242, 167077,  DOI:10.1016/j.ijleo.2021.167077.
21. X. Hao, T. Tang and X. Hu, Detection sensitivity improvement study of LIBS by combining Au-nanoparticles and magnetic field, Spectrosc. Spectral Anal., 2019, 39, 1599–1603,  DOI:10.3964/j.issn.1000-0593(2019)05-1599-05.
22. Y. Yang, X. Hao, L. Zhang and L. Ren, Application of scikit and Keras libraries for the classification of iron ore data acquired by laser-induced breakdown spectroscopy (LIBS), Sensor, 2020, 20, 1–11,  DOI:10.3390/s20051393.
23. N. Gyftokostas, E. Nanou, D. Stefas, V. Kokkinos, C. Bouras and S. Couris, Classification of Greek olive oils from different regions by machine learning-aided laser-induced breakdown spectroscopy and absorption spectroscopy, Molecules, 2021, 26, 1241,  DOI:10.3390/molecules26051241.
24. E. Bellou, N. Gyftokostas, D. Stefas, O. Gazeli and S. Couris, Laser-induced breakdown spectroscopy assisted by machine learning for olive oils classification: the effect of the experimental parameters, Spectrochim. Acta, Part B, 2020, 163, 105746,  DOI:10.1016/j.sab.2019.105746.
25. R. Gaudiuso, E. Ewusi-Annan, N. Melikechi, S. Sun, B. Liu, L. F. Campesato and T. Merghoub, Using LIBS to diagnose melanoma in biomedical uids deposited on solid substrates: limits of direct spectral analysis and capability of machine learning, Spectrochim. Acta, Part B, 2018, 146, 106–114.
26. W. Hao, X. Hao, Y. Yang, X. Liu, Y. Liu, P. Sun and R. Sun, Rapid classification of soils from different mining areas by laser-induced breakdown spectroscopy (LIBS) coupled with a PCA-based convolutional neural network, J. Anal. At. Spectrom., 2021, 36, 2509–2518,  10.1039/d1ja00078k.
27. P. Yang, G. Fu, J. Wang, Z. Luo and M. Yao, A tutorial review on methods of agricultural product sample pretreatment and target analysis by laser-induced breakdown spectroscopy, J. Anal. At. Spectrom., 2022, 37, 1948–1960,  10.1039/d2ja00149g.
28. Y. Zhang, T. Zhang and H. Li, Application of laser-induced breakdown spectroscopy (LIBS) in environmental monitoring, Spectrochim. Acta, Part B, 2021, 181, 106218,  DOI:10.1016/j.sab.2021.106218.
29. P. Yan, S. Shang, C. Zhang, N. Yin, X. Zhang, G. Yang, Z. Zhang and Q. Sun, Research on the Processing of Coal Mine Water Source Data by Optimizing BP Neural Network Algorithm With Sparrow Search Algorithm, IEEE Access, 2021, 9, 108718–108730,  DOI:10.1109/ACCESS.2021.3102020.
30. X. Luo, R. Chen, M. H. Kabir, F. Liu, Z. Tao, L. Liu and W. Kong, Fast Detection of Heavy Metal Content in Fritillaria thunbergii by Laser-Induced Breakdown Spectroscopy with PSO-BP and SSA-BP Analysis, Molecules, 2023, 28, 3360,  DOI:10.3390/molecules28083360.
31. J. Ren, Z. Yang, Y. Zhao and K. Yu, Collinear double-pulse laser-induced breakdown spectroscopy based Cd profiling in the soil, Opt. Express, 2022, 30, 37711–37726,  DOI:10.1364/OE.471563.
32. Q. Wu, P. Li, Z. Chen and Z. Tao, A clustering-optimized segmentation algorithm and application on food quality detection, Sci. Rep., 2023, 13, 9069,  DOI:10.1038/s41598-023-36309-8.
33. National Institute of Standards and Technology, NIST Chemistry WebBook. https://webbook.nist.gov/chemistry/, accessed February 5, 2023.
34. T. Guan, X. Tian, J. Xu, J. Luo, Z. Peng, H. Yang, X. Zhao and J. Zhang, Investigation and risk assessment of ethyl carbamate in Chinese Baijiu, LWT--Food Sci. Technol., 2021, 152, 112340,  DOI:10.1016/j.lwt.2021.112340.
35. H. A. Sakandar, R. Hussain, Q. F. Khan and H. Zhang, Functional microbiota in Chinese traditional Baijiu and Mijiu Qu (starters): A review, Food Res. Int., 2021, 138, 109830,  DOI:10.1016/j.foodres.2020.109830.
36. Shanxi Xinghuacun Fen Wine Factory Co., Ltd., Enterprise Search, https://www.fenjiu.com.cn/, accessed August 26, 2023.
37. T. M. Shami, A. A. EI-Saleh, M. Alswaitti, Q. Al-Tashi, M. A. Summakieh and S. Mirjalili, Particle Swarm Optimization: A Comprehensive Survey, IEEE Access, 2023, 10, 10031–10061,  DOI:10.1109/ACCESS.2022.3142859.
38. G. O. Anyanwu, C. I. Nwakanma, J. M. Lee and D. S. Kim, Optimization of RBF-SVM Kernel Using Grid Search Algorithm for DDoS Attack Detection in SDN-Based VANET, IEEE Internet Things J., 2023, 10, 8477–8490,  DOI:10.1109/JIOT.2022.3199712.
39. J. Zhao, Y. Yang, L. Chen, J. Zheng, X. Lv, D. Li, Z. Fang, C. Shen, V. Mallawaarachchi and Y. Lin, Quantitative metaproteomics reveals composition and metabolism characteristics of microbial communities in Chinese liquor fermentation starters, Front. Microbiol., 2023, 13, 1098268,  DOI:10.3389/fmicb.2022.1098268.
40. B. Wang, Q. Wu, Y. Xu and B. Sun, Specific Volumetric Weight-Driven Shift in Microbiota Compositions With Saccharifying Activity Change in Starter for Chinese Baijiu Fermentation, Front. Microbiol., 2018, 9, 2349,  DOI:10.3389/fmicb.2018.02349.
41. C. Fang, W. Lu, Q. Liu, Y. Chen, W. Jia and Y. Xu, Comparative study between the effects of aged and fresh Chinese baijiu on gut microbiota and host metabolism, Food Biosci., 2022, 49, 101859,  DOI:10.1016/j.fbio.2022.101859.

---