Trilinear Decomposition Research Articles

Fast Determination and Source Apportionment of Eight Polycyclic Aromatic Hydrocarbons in PM10 Using the Chemometric-Assisted HPLC-DAD Method

Polycyclic aromatic hydrocarbons (PAHs) are a group of organic compounds that are both toxic and hazardous to human health and ecological systems. In recent work, a novel analytical strategy based on the chemometric-assisted HPLC-DAD method was proposed for the quantification and source apportionment of eight PAHs in PM10 samples. Compared to traditional chromatographic methods, this approach does not require the purification of complex PM10 samples. Instead, it utilizes a mathematical separation method to extract analytes’ profiles from overlapping chromatographic peaks, enabling precise quantification of PAHs in PM10. Firstly, 40 PM10 samples collected in Loudi city during two sampling periods were used for analysis. Subsequently, the second-order calibration method based on alternating trilinear decomposition (ATLD) was employed to handle the three-way HPLC-DAD data. Finally, the pollution sources of PAHs were analyzed by the feature component analysis method according to the obtained relative concentration matrix. For the validation model, the average recoveries of eight PAHs were between (88.8 ± 7.6)% and (105.6 ± 7.5)%, and the root-mean-square errors of prediction ranged from 0.03 μg mL−1 to 0.47 μg mL−1. The obtained limits of quantification for eight PAHs were in the range of 0.0050 μg mL−1 to 0.079 μg mL−1. For actual PM10 samples, results of the feature component analysis indicated that the main source of PAHs in PM10 may be traffic emissions and coal combustion. In summary, the proposed method provided a new and rapid analysis method for the accurate determination and source apportionment of PAHs in atmospheric aerosols.

Geographical origin identification of cinnamon using HPLC-DAD fingerprints and chemometrics

The commercial value of cinnamon depends largely on its geographical origin. This study proposed high-performance liquid chromatography-diode array detection (HPLC-DAD) fingerprints combined with chemometrics to identify geographical origins of cinnamons. Alternating trilinear decomposition (ATLD) algorithm was applied to extract meaningful chemical information from the HPLC fingerprint data. Then, non-targeted HPLC fingerprints and chemical component information were used as chemical descriptors to distinguish geographical origins using principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA)/N-way partial least squares-discriminant analysis (NPLS-DA). Satisfactory classification results were acquired for each discriminant model and the correct classification rates (CCRs) of training and test sets were 100%. Moreover, four characteristic variables were screened by variable importance in projection (VIP) method, three of which were reasonably identified as cinnamic acid, cinnamaldehyde and 2-methoxycinnamaldehyde, which can be regarded as potential markers to distinguish cinnamon from different geographical origins. The results indicated that HPLC-DAD fingerprints combined with chemometrics can be a promising means to identify the geographical origins and screen potential markers of cinnamon samples, especially in the absence of high-cost and sophisticated analytical instrumentals (e.g., high-resolution mass spectrometry and nuclear magnetic resonance spectroscopy). This work can help to prevent cinnamon fraud practices and maintain the normal order of the cinnamon market.

PARAFAC under non-negativity constraint is adapted to recover the underlying Beer-Lambert law of the excitation-emission fluorescence matrix measurements acquired from analyte-triggered semiconductor QDs photoluminescence modulation. When and why?

BackgroundAnalyte-triggered semiconductor quantum dots (QDs) modulation in the presence of non-consistently responsive fluorescent species represents a challenging analytical issue in concrete multi-way data handling. QDs with heterogeneous sizes and/or uneven distribution of functional moieties on their surfaces exhibit significant fluctuations in the fluorescent response components, known as chemical rank, across different excitation/emission modes. This phenomenon may lead to a substantial deviation from the proportionality prescribed by Beer-Lambert law. Nonetheless, even in the presence of such deviation, a multi-way model may be successfully selected after determining a proper chemical rank in a QDs system. ResultsWe show that in a valid PARAllel FACtor (PARAFAC) model under properly determined chemical rank, meaningfully resolved pure spectral profiles can be reached for each fluorescent responsive constituent in the original excitation-emission fluorescence matrix (EEFM) measurements. This was thoroughly illustrated by applying PARAFAC trilinear decomposition of a three-way data array of two distinct datasets acquired from semiconductor QDs sensing systems with low-rank trilinear assumption. The first dataset, presented here for the first time, comprises EEFM measurements of the ligand-driven quenching of thiomalic acid (TMA)-capped AgInS2 (AIS) QDs by vomitoxin. The second dataset, employed for illustrative purposes, comprises EEFM measurements of the quenching, via cation bridging, of glutathione (GSH)-capped CdTe QDs by Pb(II). The results of this study enabled the determination of vomitoxin at a ppb level in real samples of fish feeds, showcasing the efficacy of the PARAFAC model in resolving spectral signatures (loadings) and pure concentration profiles (scores). SignificancePARAFAC under a properly examined chemical rank can be easily adapted for retrieval the underlying Beer-Lambert law of the original EEFM measurements with a low-rank trilinear structure through the chemically meaningful information either when (i) no deviation of Beer-Lambert law was observed as deeply discussed in connection with the dataset acquired from vomitoxin-driven molecular sensing through TMA-capped AIS QDs, or when (ii) substantial deviations of the Beer-Lambert law are evident, as discussed in connection with the dataset collected from sensing ionic species through Pb(II) bridging of GSH-capped CdTe QDs.

Green Approach for Simultaneous Quantitative Analysis of Fluoroquinolones in the Environment using Three-Dimensional Fluorescence Coupled with Second-Order Calibration Method.

In this study, a straightforward and quick analytical technique based on the self-weighted alternating trilinear decomposition (SWATLD) algorithm in conjunction with excitation-emission matrix (EEM) fluorescence for the simultaneous determination of the antibiotics levofloxacin (LVFX) and ciprofloxacin (CIP) in environmental waters and sediments was developed. This approach completely utilizes the "second-order advantage" and inherits the great sensitivity of classic fluorescence. It replaces or improves the conventional "physical/chemical separation" with "mathematical separation", enabling direct and quick quantification of the target analytes even in the presence of unknown interferences, greatly streamlining sample preparation procedures, consuming less solvent, and speeding up analysis time, and allows successful and environmentally friendly solution of overlapping fluorescence spectra of multiple components in complicated environmental matrices without cumbersome pretreatment steps and complex and expensive instrumentation. The limits of detection varied between 0.34 and 0.67 ng mL- 1, and the average spiking recoveries of LVFX and CIP in water and sediment ranged from 97.6 to 107.7% with relative standard deviations lower than 6.6%. The developed method shows the reliability of the technology and the ability to quickly detect trace antibiotics in lake water even in the presence of unidentified interferents.

Data-Driven Version of Multiway Soft Independent Modeling of Class Analogy (N-Way DD-SIMCA): Theory and Application.

One-class classification (OCC) is discussed in the framework of the measurement and processing of multiway data. Data-driven soft independent modeling of class analogy (DD-SIMCA) is applied in the following formats: (1) multiblock and (2) Tucker 3 N-way SIMCA, which are shown to be useful tools for solving classification tasks. A new decision rule for N-way DD-SIMCA is adopted based on the conventional two-way DD-SIMCA model. Multiblock SIMCA is shown to be useful for variable selection, and Tucker 3 SIMCA to select the optimal model complexity when applying multiway data decomposition and to assess the role of individual samples in the classification model. Both approaches, together with the two-way DD-SIMCA version applied to the unfolded data, are compared regarding the analysis of an experimental data set including genuine and adulterated blueberry extract samples. The latter were employed to produce matrix spectral-time data matrices per sample within a flow injection system, taking advantage of the spectral changes in the sample constituents as a function of the pH of the carrier phase. The need to employ the Tucker 3 model instead of a trilinear decomposition is supported by a discussion on the lack of the trilinearity property of the studied data.

Core consistency blind source separation based on compressive sensing trilinear decomposition

In the traditional fault diagnosis method based on blind source separation, the mechanical signals are required to meet some additional conditions in the process of estimation of mixed matrix and separation of source signals, which may cause many problems in practice application because the mechanical signals do not suffer to these additional conditions. Additionally, the trilinear model has complex construction, high computational complexity, and large storage capacity in the fault source blind separation model with trilinear parallel factors. Based on the above deficiencies, a core consistency blind source separation method is proposed based on compressive sensing trilinear decomposition. In the proposed method, the core consistency diagnostic (CORCONDIA) is used to fit the tensor model, and determine the number of fault sources. The mixed matrix of the observed signal is estimated by the load matrices. Then, the minimum norm method is used to solve the underdetermined blind separation of fault sources. The simulation and experiment results show that the proposed method is superior the traditional core consistency blind separation method based on trilinear parallel factor analysis in the underdetermined blind separation. The proposed method can effectively solve the problem that the number of sources is unknown and the number of observed signals is less than the number of source signals.

Determination of phenolic compounds in water using a multivariate statistical analysis method combined with three-dimensional fluorescence spectroscopy.

Phenolic compounds are toxic chemical pollutants present in water. Three-dimensional fluorescence spectroscopy analysis is an effective and rapid method for real-time phenol monitoring in aquatic environments. However, similar chemical structures of phenols result in highly overlapping three-dimensional fluorescence spectra. Therefore, it is extremely difficult to analyze and quantify the concentration of components in a mixture system that includes two or more phenolic compounds. In this article, we study the mixed phenol system containing phenol, o-cresol, p-cresol, m-cresol, catechol, and resorcinol combined with excitation-emission matrix (EEM) fluorescence data. A multivariate statistical method called best linear unbiased prediction (BLUP) is proposed to analyze the spectra with the aim to achieve quantitative results and a trilinear decomposition algorithm called parallel factor analysis (PARAFAC) was used for comparison. Two experiments with different calibration samples were set to validate the effectiveness of BLUP through recovery, ARecovery (Average Recovery), AREP (Average Relative Error of Prediction), and RMSE (Root Mean Square Error). Overall, the average recovery of each component in experiment 1 and experiment 2 ranged from 95.91% to 111.62% and 82.91% to 129.02%, respectively. Based on the results of the experiments, the concentration of phenolic compounds in water can be quantitatively determined by combining three-dimensional fluorescence spectroscopy with the BLUP method.

Rapid identification of high-temperature Daqu Baijiu with the same aroma type through the excitation emission matrix fluorescence of maillard reaction products

The flavor of high-temperature Daqu Baijiu with the same aroma type is similar but the quality and price are very different, so that the high-quality Baijiu is often pretended as low-quality Baijiu. In this study, excitation-emission matrix fluorescence spectrometry combined with chemometrics was proposed for rapid identification of high temperature Daqu Baijiu with the same aroma type. The method was based on seven components screened by alternating trilinear decomposition (ATLD), and the discriminant model was established by data-driven-soft independent modeling of class analogy (DD-SIMCA), with the discriminant rate of 100%. Some Maillard reaction products, such as pyrazine and furfural compounds, with correlation >0.8 were obtained by GC-MS combined with Pearson analysis, and the prediction accuracy between 90.32% and 96.77%. This method has the characteristics of high sensitivity, fast detection and no pre-treatment, which provides a new strategy for rapid discrimination of high-temperature Daqu Baijiu with the same aroma type.

Rapid detection and quantification of adulteration in saffron by excitation-emission matrix fluorescence combined with multi-way chemometrics.

Saffron has gained people's attention and love for its unique flavor and valuable edible value, but the problem of saffron adulteration in the market is serious. It is urgent for us to find a simple and rapid identification and quantitative estimation of adulteration in saffron. Therefore, excitation-emission matrix (EEM) fluorescence combined with multi-way chemometrics was proposed for the detection and quantification of adulteration in saffron. The fluorescence composition analysis of saffron and saffron adulterants (safflower, marigold and madder) were accomplished by alternating trilinear decomposition (ATLD) algorithm. ATLD and two-dimensional principal component analysis combined with k-nearest neighbor (ATLD-kNN and 2DPCA-kNN) and ATLD combined with data-driven soft independent modeling of class analogies (ATLD-DD-SIMCA) were applied to rapid detection of adulteration in saffron. 2DPCA-kNN and ATLD-DD-SIMCA methods were adopted for the classification of chemical EEM data, first with 100% correct classification rate. The content of adulteration of adulterated saffron was predicted by the N-way partial least squares regression (N-PLS) algorithm. In addition, new samples were correctly classified and the adulteration level in adulterated saffron was estimated semi-quantitatively, which verifies the reliability of these models. ATLD-DD-SIMCA and 2DPCA-kNN are recommended methods for the classification of pure saffron and adulterated saffron. The N-PLS algorithm shows potential in prediction of adulteration levels. These methods are expected to solve more complex problems in food authenticity. © 2023 Society of Chemical Industry.

Aqueous two-phase systems coupled with chemometrics-enhanced HPLC-DAD for simultaneous extraction and determination of flavonoids in honey.

In this study, an accurate, rapid, green, and environment friendly method for the extraction and quantitative analysis of flavonoids in honey was established by using the aqueous two-phase extraction combined with the chemometrics-assisted HPLC-DAD. The first purpose of this study was to extract seven flavonoids in five different types of honey using alcohol/salt aqueous two-phase system (ATPS). The system with 2.82mL sodium citrate (30%), 1.58mL water, and 3.10mL isopropanol, showed the highest flavonoids extraction yields in the top phase (87.66-101.50%). Additionally, the three-way array of honey samples based on HPLC-DAD was decomposed mathematically by the alternating trilinear decomposition (ATLD) algorithm to obtain reasonable chromatograms, spectra, and concentration profiles for each analyte. Compared with the traditional solid-phase extraction method, the ATPS-ATLD-based method showed satisfactory spiked recoveries, lower limit of detection, and higher sensitivity, further verifying its accuracy and stability.

Chemometrics-assisted excitation-emission matrix fluorescence spectroscopy for real-time migration monitoring of multiple polycyclic aromatic hydrocarbons from plastic products to food simulants

Polycyclic aromatic hydrocarbons (PAHs), as a class of organic pollutants that have attracted much attention, are likely to be formed with the production and processing of plastic products, and they may migrate from contaminated plastic products to food, causing the risk of poisoning or cancer. In this study, migration tests were carried out on disposable plastic products for food contact, and a novel strategy that combines excitation-emission matrix (EEM) fluorescence spectroscopy with the advanced second-order calibration method based on the three-direction resection alternating trilinear decomposition (TDR-ATLD) algorithm was used to monitor the migration of three PAHs anthracene (ANT), pyrene (PYR), and phenanthrene (PHE) from the plastic products to food simulants in real-time. With the “second-order advantage”, even if the fluorescence spectra of the target analytes overlapped seriously, and other unknown substances migrated from the plastic products to food simulants, accurate qualitative and quantitative results were still obtained by the proposed method. In the static system, the coefficient of determination (R2) of the three PAHs within the calibration range were all greater than 0.99, and the average spiked recoveries were 99.5-107.1%, with the standard deviation lower than 8.9%. The figures of merit (FOMs) and intra- or inter-day precision also showed good feasibility and reliability of the method. In the simulation study of the migration kinetic process, three PAHs can be quantified in real-time in complex matrix, then the related migration equations were established. The results indicate that the proposed method can be used for real-time migration quantitative monitoring of PAHs, providing a potential and available method for the study of the migration kinetics of hazardous substances from food contact materials to food or food simulants.

Determination of the Carbamate Pesticides Fuberidazole and Pirimicarb in Crayfish (Procambarus Clarkii) Using Excitation-Emission Fluorescence Matrices with Self-Weighted Alternating Trilinear Decomposition (SWATLD)

Fuberidazole and pirimicarb are carbamate pesticides for insect control. There is pesticide accumulation through the food chain that may ultimately migrate into the human body through the crayfish. Here a rapid method for simultaneous determination of fuberidazole and pirimicarb in crayfish is reported. The excitation-emission fluorescence matrix was obtained by simple pretreatment of crayfish meat using acetonitrile, and qualitative and quantitative information was separated from overlapping spectra using the self-weighted alternating trilinear decomposition. The detection limits of the pesticides for fuberidazole and pirimicarb were 0.068 mg/kg and 0.160 mg/kg, respectively, and the limits of quantitation were 0.207 mg/kg and 0.485 mg/kg. The average recoveries were 90.79% and 98.77%. The results show that a combination of excitation-emission fluorescence matrix and alternating trilinear decomposition can be used for the determination of fuberidazole and pirimicarb in crayfish with simple pretreatment using acetonitrile.

Three-Dimensional Near-Field Passive Localization Considering Propagation Attenuation Based on Trilinear Decomposition

Three-Dimensional Near-Field Passive Localization Considering Propagation Attenuation Based on Trilinear Decomposition

Fluorophores-Assisted Excitation Emission Matrix Fluorescence Method for the Origin Traceability of Lily

In this work, a fluorophores-assisted excitation/emission matrix (EEM) fluorescence method was proposed to trace the origin of lily in the Chinese market. There are few active components in lilies that have fluorescent signals, and too few characteristic variables may lead to unsatisfactory accuracy in the subsequent classification. Therefore, three fluorophores, 2-Aminoethyl diphenylborinate (DPBA), o-Phthalaldehyde (OPA) and Rhodamine B (RB), were used to enrich the information of the fluorescent fingerprint of lily, which can improve the classification accuracy. The lily samples were characterized by using EEM fluorescence coupled with the alternating trilinear decomposition (ATLD) algorithm, which was able to extract information of various fluorophores in lily samples. Two chemical pattern recognition methods, principal component analysis-linear discriminant analysis (PCA-LDA) and partial least squares-discrimination analysis (PLS-DA), were used to model and trace the origin of different lilies. When the fluorophores were added, the accuracy of the test set and prediction set obtained by the classification model increased from 71.4% to 92.9% and 66.7% to 100%, respectively. The proposed method combined fluorophores-assisted EEM fluorescence with multi-way chemometric methods to extract comprehensive information on the samples, which provided a potential method for the origin traceability of traditional Chinese medicine.

Anti-interference and non-destructive identification of textile fabrics using front-face excitation-emission matrix fluorescence spectroscopy combined with multi-way chemometrics

The identification of trace textile fabrics discovered at crime scenes plays a crucial role in the case of forensic investigations. Additionally, in practical situations, fabrics may be contaminated, making identification more challenging. To address the aforementioned issue and promote the application of fabrics identification in forensic analysis, front-face excitation-emission matrix (FF-EEM) fluorescence spectra coupled with multi-way chemometric methods were proposed for the interference-free and non-destructive identification of textile fabrics. Common commercial dyes in the same color range under different materials (cotton, acrylic, and polyester) that cannot be visually distinguished were investigated, and several binary classification models for the identification of dye were established using partial least squares discriminant analysis (PLS-DA). The identification of dyed fabrics in the presence of fluorescent interference was also taken into consideration. In each kind of pattern recognition model mentioned above, the classification accuracy (ACC) of the prediction set was 100%. The alternating trilinear decomposition (ATLD) algorithm was executed to separate mathematically and remove the interference, and the classification model based on the reconstructed spectra attained an accuracy of 100%. These findings indicate that FF-EEM technology combined with multi-way chemometric methods has broad prospects for forensic trace textile fabric identification, especially in the presence of interference.

Kernelizing: A way to increase accuracy in trilinear decomposition analysis of multiexponential signals

The unmixing of multiexponential decay signals into monoexponential components using soft modelling approaches is a challenging task due to the strong correlation and complete window overlap of the profiles. To solve this problem, slicing methodologies, such as PowerSlicing, tensorize the original data matrix into a three-way data array that can be decomposed based on trilinear models providing unique solutions. Satisfactory results have been reported for different types of data, e.g., nuclear magnetic resonance or time-resolved fluorescence spectra. However, when decay signals are described by only a few sampling (time) points, a significant degradation of the results can be observed in terms of accuracy and precision of the recovered profiles.In this work, we propose a methodology called Kernelizing that provides a more efficient way to tensorize data matrices of multiexponential decays. Kernelizing relies on the invariance of exponential decays, i.e., when convolving a monoexponential decaying function with any positive function of finite width (hereafter called “kernel”), the shape of the decay (determined by the characteristic decay constant) remains unchanged and only the preexponential factor varies. The way preexponential factors are affected across the sample and time modes is linear, and it only depends on the kernel used. Thus, using kernels of different shapes, a set of convolved curves can be obtained for every sample, and a three-way data array generated, for which the modes are sample, time and kernelizing effect. This three-way array can be afterwards analyzed by a trilinear decomposition method, such as PARAFAC-ALS, to resolve the underlying monoexponential profiles. To validate this new approach and assess its performance, we applied Kernelizing to simulated datasets, real time-resolved fluorescence spectra collected on mixtures of fluorophores and fluorescence-lifetime imaging microscopy data. When the measured multiexponential decays feature few sampling points (down to fifteen), more accurate trilinear model estimates are obtained than when using slicing methodologies.

Trilinear decomposition based near-field source localization with MIMO velocity vector sensor arrays

This paper studies the near-field (NF) parameter estimation problem for a multiple-input multiple-output (MIMO) array system, which employs multiple pairs of orthogonal velocity sensors at both the transmitter and the receiver. A trilinear decomposition method is proposed to estimate the four-dimensional (4-D) parameters, including the direction of departure (DOD), the range from transmitter to target (RFTT), the direction of arrival (DOA), and the range from target to receiver (RFTR). Firstly, the output of the matched filter at the receiver is formulated in a third-order parallel factor (PARAFAC) model; secondly, the initial coarse estimates of DOD, RFTT, DOA and RFTR embedded in the velocity vector sensors are obtained through trilinear decomposition, and then more accurate estimates of DOD, RFTT, DOA and RFTR are achieved from the steering vector. The proposed method is search-free and has a close form, with automatically paired results. Its performance is demonstrated via numerical examples.

A novel second-order calibration algorithm for processing fluorescence data with scattering: Three-direction resection ATLD

The light scattering is common during fluorescence measurements, such as first- and second-order Rayleigh and Raman scatterings, which are not conducive to the decomposition of the multi-way data because it destroys the multi-linear structure. This paper proposed two second-order calibration algorithms, including three-direction resection alternating trilinear decomposition (TDR-ATLD) and three-direction resection self-weight alternating trilinear decomposition (TDR-SWATLD), which are based on an improved trilinear decomposition model for directly decomposition of the three-way fluorescence data with scattering regions removed. Through the simulation data, the proposed algorithms were compared with other three algorithms (ATLD, Interpolation-ATLD, TDR-PARAFAC). The calculation results of three algorithms (TDR-PARAFAC, TDR-ATLD and TDR-SWATLD) are better at small or large scattering widths. Both TDR-ATLD and TDR-SWATLD are insensitive to the number of components and the performance of TDR-ATLD had little fluctuation with different noise levels among the five algorithms. Additionally, TDR-ATLD and TDR-SWATLD are better than other algorithms when dealing with collinear data. The proposed algorithms were also used in two sets of real data, compared with other two algorithms (Interpolation-ATLD and TDR-PARAFAC) for dealing with scattering, the proposed algorithms could deal with fluorescence data with scattering and provided satisfactory results and TDR-ATLD has faster convergence. In general, TDR-ATLD has greater potential and performs better in processing fluorescence data with scattering.

An upgrade of MVC2, a MATLAB graphical user interface for second-order multivariate calibration: Beyond trilinear models

In the present report, an upgrade of a MATLAB graphical user interface (GUI) toolbox for implementing second-order multivariate calibration models is presented, named Multivariate Calibration 2 (MVC2). Due to the advances in the chemometric field, this software has undergone numerous changes since its last release in 2009. The new freely available MVC2 incorporates novel features and models that make it a very practical tool for data processing. Overall, the current version presents different types of algorithms, designed to provide specific degrees of model flexibility for different multi-way data scenarios: in addition to trilinear decomposition (TLD) and residual bilinearization (RBL) models, which were previously included in the 2009 version, parallel factor analysis 2 (PARAFAC2) and multivariate curve resolution (MCR) methods are now available. All these chemometric methodologies are integrated in the software for quantitative purposes, and if needed, they permit to exploit the so-called analytical second-order advantage. The program offers practical tools for data loading and visualization, model development and validation, and prediction in unknown samples. The software output also includes a comprehensive report on the analytical figures of merit (AFOMs) applicable to second-order calibration models. Specifically, a tool was incorporated for the diagnosis of the remaining rotational ambiguity in the MCR solutions. In the present report, the new features of MVC2 are illustrated through simulated examples and real calibration datasets.

Front-face excitation-emission matrix fluorescence spectroscopy combined with interpretable deep learning for the rapid identification of the storage year of Ningxia wolfberry

Ningxia wolfberry stored for many years may be disguised as fresh wolfberry by unscrupulous traders and sold for huge profits. In this work, the front-face excitation-emission matrix (FF-EEM) fluorescence spectroscopy coupled with interpretable deep learning was proposed to identify the storage year of Ningxia wolfberry in a lossless, fast and accurate way. Alternating trilinear decomposition (ATLD) algorithm was used to decompose the three-way data array obtained by Ningxia wolfberry samples, extracting the chemically meaningful information. Meanwhile, a convolutional neural network (CNN) model for the identification of the storage year of Ningxia wolfberry, called EEMnet, was proposed. The model successfully classified wolfberry samples from different storage years by extracting the subtle feature differences of the spectra, and the correct classification rate of the training set, test set and prediction set was more than 98%. In addition, a series of interpretability analyses were implemented to break the “black box” of the deep learning model. These results indicated that the method based on FF-EEM fluorescence spectroscopy combined with EEMnet could quickly and accurately identify the year of Ningxia wolfberry in a green way, providing a new idea for the identification of the storage years of Chinese medicinal materials.