Strong Background Noise Research Articles

Bearing fault diagnosis based on wavelet adaptive threshold filtering and multi-channel fusion cross-attention neural network.

In industrial applications, it is difficult to extract the fault feature directly when the rolling bearing works under strong background noise. In addition, single-channel vibration sensor data pose limitations in providing a comprehensive representation of bearing fault features; how to effectively fuse data of each channel and extract features is a challenge. To solve the above-mentioned problems, a fault diagnosis method based on wavelet adaptive threshold filtering and multi-channel fusion cross-attention neural network is proposed in this paper. First, the multi-scale discrete wavelet transform is applied to obtain the wavelet coefficients of each channel. Adaptive threshold filtering is conducted to filter out noise and extract symbolic features. The threshold updates with the training of the network. Then, the wavelet coefficients are reconstructed and the channel attention is performed to further extract the symbolic features of the fault signal. Finally, the multi-channel fault signals are fused by a cross-attention module. This module can fully extract the features of each channel and fuse multi-channel data. To improve the generalization ability of the network, residual connections are added. To verify the effectiveness of the proposed method, experiments are carried out on the rolling bearing datasets of Case Western Reserve University and Xi'an Jiaotong University. In addition, the gas turbine main bearing dataset is also applied to prove the reliability of this method.

Rolling bearing fault diagnosis method based on MSDCNN in strong noise environment

Aiming at the problems of poor noise resistance, high computational complexity and insufficient generalization performance of traditional bearing fault diagnosis methods based on deep learning, a rolling bearing fault diagnosis method based on multi-scale dynamic convolutional neural network (MSDCNN) is proposed. Firstly, the one-dimensional vibration signal of the rolling bearing is converted to the frequency domain by Fourier transform, and the features are further extracted by wide convolution kernel. Secondly, a multi-scale dynamic convolution structure is proposed, and the feature information extracted by convolution kernels of different sizes is given different weights by using an improved channel attention mechanism. Then, a self-calibrating spatial attention mechanism (SCSAM) is designed, and the extracted feature information is input into the spatial attention mechanism to capture the importance of different regions. Finally, the features are further extracted by small convolution kernels, and the fault category is classified by using Softmax classifier. The fault diagnosis performance of the proposed model is verified by three different data sets. The experimental results show that compared with other intelligent models, the proposed model has higher classification accuracy, better generalization ability and stronger robustness under strong noise background.

Adaptive feature mode decomposition method for bearing fault diagnosis under strong noise

In practical mechanical equipment operation, bearing vibration signals are challenging to analyze due to their non-stationarity and low signal-to-noise ratio for traditional detection methods. As a new signal decomposition method, the feature mode decomposition (FMD) method has been successfully applied to bearing fault diagnostics. However, FMD’s decomposition efficiency is easily influenced by the input parameter settings, and the efficiency of the decomposed signals is related to the number of initialized filter banks. For this reason, this paper proposes an adaptive parameterized feature mode decomposition method. Firstly, based bearing fault diagnosis method using a cuckoo search algorithm with logarithmic decline of nonlinear inertial weights and random adjustment discovery probability (DWCS), which is used to optimize the three parameters of the FMD. Then, a new approach to finding out the maximum feature fault frequency and its multiplicative feature frequency is proposed, called the feature frequency ratio (FFR), obtained from the envelope spectrum of bearing fault signal. Finally, this paper uses the DWCS based on the maximum FFR to adaptively select the best FMD parameter combination, which is verified by simulation, the results of the proposed AFMD can effectively extract bearing fault features under strong background noise with a signal-to-noise ratio of −15 dB, and actual experiments further verify the superiority of the method, which not only improves the accuracy of fault diagnosis under strong background noise, but also contributes to the overall maintenance and operational efficiency of the mechanical system.

PeriodNet: Noise-Robust Fault Diagnosis Method Under Varying Speed Conditions.

Rolling bearings are critical components in modern mechanical systems and have been extensively equipped in various rotating machinery. However, their operating conditions are becoming increasingly complex due to diverse working requirements, dramatically increasing their failure risks. Worse still, the interference of strong background noises and the modulation of varying speed conditions make intelligent fault diagnosis very challenging for conventional methods with limited feature extraction capability. To this end, this study proposes a periodic convolutional neural network (PeriodNet), which is an intelligent end-to-end framework for bearing fault diagnosis. The proposed PeriodNet is constructed by inserting a periodic convolutional module (PeriodConv) before a backbone network. PeriodConv is developed based on the generalized short-time noise resist correlation (GeSTNRC) method, which can effectively capture features from noisy vibration signals collected under varying speed conditions. In PeriodConv, GeSTNRC is extended to the weighted version through deep learning (DL) techniques, whose parameters can be optimized during training. Two open-source datasets collected under constant and varying speed conditions are adopted to assess the proposed method. Case studies demonstrate that PeriodNet has excellent generalizability and is effective under varying speed conditions. Experiments adding noise interference further reveal that PeriodNet is highly robust in noisy environments.

Operating modal parameter identification of railway vehicle bogie based on k-value optimization variational mode decomposition and second-order blind identification method

Due to the variable operating conditions and strong background noise, the accuracy of modal parameter identification of the railway vehicle bogie is low. This paper proposes a method for identifying operational modal parameters by combining the K-value optimization variational mode decomposition (KVMD) and second-order blind identification (SOBI) method (KVMD–SOBI method). Firstly, the variational mode decomposition (VMD) algorithm is used to adaptively decompose a single-channel vibration signal into intrinsic mode function (IMF) components with different scale characteristics. The feature IMF components are selected through the K-value optimization method based on energy ratio and stability graph. Then, the SOBI algorithm is used to separate the matrix composed of the feature IMF components. And the mode shape and single-mode signal are obtained. Finally, the natural frequency and damping ratio of each order mode are extracted by the single-mode parameter recognition method. Through simulation and experimental verification, it is shown that the proposed method has good noise robustness and can effectively identify the first three modal parameters of railway vehicle bogies in variable-speed operation. At the same time, the feature IMF components extracted by KVMD algorithm eliminate the sensitivity of sensor installation positions and operating conditions to detection results. Therefore, this method is an excellent operational mode identification method based on a single-channel signal with good noise and working condition robustness.

Multi-frequency weak signal detection based on Liu-like chaotic synchronization system and its hardware circuit implementation

Considering the shortcomings of traditional chaotic systems in weak signal detection methods, such as the high threshold sensitivity requirement and the narrow detection frequency domain. This study proposes a novel three-dimensional chaotic synchronization system, and the dynamical of the system are exhaustively characterized using equilibrium points, phase diagrams, Lyapunov exponential spectra, and bifurcation diagrams. This method involves weak signal detection by means of chaotic synchronization control. Synchronization of a chaotic system using a backstepping synchronization method is used to detect weak signals by analyzing the synchronization error after the introduction of weak signals in a strong noise background. The chaotic system is implemented by hardware circuits, and the simulation of chaotic synchronization control and detection of weak signals from the perspective of circuits is carried out by circuit simulation software. Additionally, the frequency range within which the system is capable of weak signal detection is tested through extensive simulation experiments. Finally, multi-frequency signals detection experiments are performed. The experimental results demonstrate that the system can accurately detect the frequency of weak signals address the limitations of narrow-band detection and multi-frequency signal detection is possible. Meanwhile, the circuit structure proposed in this paper is simple and has some value for engineering applications.

Application of a vibration resonance-assisted enhanced feedforward cascaded stochastic resonance system in bearing diagnostics

Extracting fault features from strong background noise is a key challenge in bearing diagnostics. Stochastic resonance (SR) has been widely applied in bearing diagnostics by utilizing noise to enhance useful signals. This paper proposes a feedforward cascaded system based on vibration resonance (VR) to assist in enhancing SR from the perspective of multi-system synergy, aiming to improve the performance of the SR system. Firstly, a matching steady-state potential function is designed to effectively alleviate the output saturation phenomenon. Then, by introducing a feedback term and VR, a feedforward cascaded system with band-pass characteristics is constructed, significantly enhancing the ability to suppress low-frequency interference. Finally, the proposed system is analyzed in the detection of simulated signals and actual bearing fault signals from the German PADERBORN dataset. The results demonstrate that the system outperforms two other SR systems and the Fast-Kurtogram method (FK), validating its performance and practicality.

Demonstration of shape analysis of neutron resonance transmission spectrum measured with a laser-driven neutron source

Laser-driven neutron sources (LDNSs) can generate strong short-pulse neutron beams, which are valuable for scientific studies and engineering applications. Neutron resonance transmission analysis (NRTA) is a nondestructive technique used for determining the areal density of each nuclide in a material sample using pulsed thermal and epithermal neutrons. Herein, we report the first successful NRTA performed using an LDNS driven by the Laser for Fast Ignition Experiment at the Institute of Laser Engineering, Osaka University. The key challenge was achieving a well-resolved resonance transmission spectrum for material analysis using an LDNS with a limited number of laser shots in the presence of strong background noise. We addressed this by employing a time-gated 6Li\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{6}{\ extrm{Li}}$$\\end{document}-glass scintillation neutron detector to measure the transmission spectra, reducing the impact of electromagnetic noise and neutron and gamma-ray flashes. Output waveforms were recorded for each laser shot and analyzed offline using a counting method. This approach yielded a spectrum with distinct resonances, which were attributed to 115In\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{115}\\,{\ extrm{In}}$$\\end{document} and 109Ag\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{109}{\ extrm{Ag}}$$\\end{document}, as confirmed through neutron transmission simulation. The spectrum was analyzed using the least-square nuclear-resonance fitting program, REFIT, demonstrating the possibility of using an LDNS for nondestructive areal-density material characterization.

Application of group decomposition and 1.5-dimensional spectra to acoustic composite fault diagnosis of rolling bearings

Abstract Under the background of strong noise and mutual interference coupling of each fault, the acoustic compound fault diagnosis of rolling bearing is very challenging。 For the composite fault features which are difficult to be extracted due to strong noise interference and uneven distribution of fault intensity, put forward the optimization of swarm decomposition combined with 1.5-dimensional(1.5-d) spectrum method of acoustic composite rolling bearing fault feature separation. The method firstly uses the composite index to iteratively search for the optimal group decomposition threshold value, the adaptive group decomposition of composite fault acoustic signals is realized by optimal parameter group disassembly, and then selects the sensitive components for the decomposed components, and then further analyzes the envelope signal of the sensitive components to reduce the redundancy components and the noise interference, and selects the 1.5-d spectrum to further analyze the envelope signal, thus realizes the effective separation of the composite faults of the rolling bearings acoustic faults. Rolling bearing simulations and experimental acoustic signals verify the validity of the proposed method, and this work gives a new tool for composite fault diagnosis of revolving machinery.

An early fault warning and diagnosis model based on a two-step analysis strategy for wind turbine bearings

Early bearing fault diagnosis plays a crucial role in ensuring the reliability and safety of wind turbines. This process encounters two primary challenges, that is, weak fault characteristics and strong background noise. Consequently, they restrict the effectiveness of conventional diagnosis methods. To address these difficulties, a two-step early fault diagnosis model is proposed based on an anomaly monitoring index [Formula: see text] and improved successive variational mode decomposition-fast spectral correlation (ISVMD-FSC). Initially, [Formula: see text] enables early anomaly detection. Subsequently, the fault type is further identified using ISVMD-FSC. The ISVMD adaptive decomposition of early warning signals can effectively reduce the noise and isolate the fault impact characteristics. Then, the fault characteristic frequency is extracted using spectral coherence analysis to achieve the purpose of fault diagnosis. This model is tested and validated on simulated data, laboratory accelerated bearing life span data, and real high-speed bearing fault data of wind turbines. Notably, the comparative study shows that [Formula: see text] is capable of detecting the fault earlier than the spectral L2/L1 norm, the sum of the weighted normalized square envelope, index 8, and index 9. Additionally, the fault feature extraction effect of ISVMD-FSC is better than many more advanced time-frequency analysis methods. These results emphasize the model’s excellent performance in early bearing fault diagnosis and its good generalization ability.

Research on ACMD-ICYCBD method for rolling bearing fault feature extraction

Aiming at the difficulty in obtaining the eigenfrequency of the vibration component of rolling bearing faults in a strong background noise environment and the problem of extraction efficiency, the adaptive chirp mode decomposition (ACMD) combined with Improved maximum second-order cyclostationary blind deconvolution (ICYCBD) fault feature extraction algorithm is proposed. Firstly, to improve the signal-to-noise ratio, the original signal is adaptively decomposed using the ACMD method, and the optimal components are selected based on the principle of maximizing the correlation gini coefficient index. Secondly, to improve the accuracy of parameter setting and extraction efficiency, an improved CYCBD method is proposed to estimate the cyclic frequency set of CYCBD using the proposed enhanced energy harmonic product spectrum (EEHPS) method for the optimal components, the envelope spectrum peak factor index is improved by proposing the envelope spectral period pulse factor (EPPF) index, and the filtering length of the CYCBD is selected adaptively using the step search to obtain the optimized filtered signal. Finally, the envelope spectrum analysis is carried out to extract the fault information accurately. The simulation signals and experimental data show that the method can quickly and accurately extract the fault characteristics of rolling bearings under strong background noise, and the comparison with other methods shows the effectiveness and superiority of the proposed method.

Multi-timescale attention residual shrinkage network with adaptive global-local denoising for rolling-bearing fault diagnosis

In actual engineering scenarios, bearing fault signals are inevitably overwhelmed by strong background noise from various sources. However, most deep-learning-based diagnostic models tend to broaden the feature extraction scale to extract rich fault features for bearing-fault identification under noise interference, with little attention paid to multi-timescale discriminative feature mining with adaptive noise rejection, which affects the diagnostic performance. Thus, a multi-timescale attention residual shrinkage network with adaptive global-local denoising (AMARSN) was proposed for rolling-bearing fault diagnosis by learning discriminative multi-timescale fault features from signals and fully eliminating noise components in the multi-timescale fault features. First, a multi-timescale attention learning module (MALMod) was developed to capture multi-timescale fault features and enhance their discriminability under noise interference. Subsequently, an adaptive global-local denoising module (AGDMod) was constructed to fully eliminate noise in multiscale fault features by constructing specific global-local denoising thresholds and designing an adaptive smooth soft thresholding function. Finally, end-to-end bearing fault diagnosis tasks were realized using a softmax classifier located at the end of the AMARSN. The AMARSN was validated using two bearing datasets. The extensive results demonstrated that the AMARSN can mine more effective fault features from signals and achieve average diagnostic accuracies of 85.24% and 80.09% under different noise with different levels.

High-Level Feature Fusion Deep Learning Model for Fault Detection in Handling Equipment in Dry Bulk Ports

The flexibility of handling equipment in dry bulk ports is poor, and frequent equipment fault induced by the high-load and high-power working conditions greatly impacts the overall port handling operations, making accurate fault detection play an important role in improving the efficiency and stability of dry bulk port operations. However, as we know, most fault detection methods for port handling equipment depend heavily on monitoring sensor data, which is not applicable in the dry bulk port due to high configuration and maintenance cost, as well as the high false alarm rate of monitoring sensors caused by strong background noise. To solve the problem, this study proposes a High-Level Feature Fusion Deep Learning Model, which uses different deep learning sub-models to extract features of structured and unstructured data. It fuses the extracted feature vectors to achieve fault detection in the handling equipment, establishing the mapping relationship between the fault (e.g., waiting for the pre-loading process, equipment fault, and others) and multi-source heterogeneous operation and maintenance data for the handling equipment, including reclaimers, belt conveyors, dumpers, and ship loaders. To verify the effectiveness of the proposed method, the actual data of a coal port in Northern China is employed as an example. The results show the deep learning model can achieve high prediction accuracy (over 86%) with high efficiency (0.5 s for each sample), which provides decision support for the fault detection in dry bulk port handling equipment.

Dual-Path Neural Network based on Deep Features and Transfer Learning for Bearing Fault Diagnosis

The safe operation of mechanical equipment is a basic requirement in the industrial manufacturing process. As an important component of mechanical equipment, rolling bearings are prone to unforeseen failures due to long-term operation under complex conditions. The occurrence of failures, no matter how big or small, will cause economic losses. Therefore, reliable diagnosis of rolling bearings is crucial. Aiming at the problem of insufficient feature recognition of convolutional neural networks under strong noise background, a deep feature extraction network integrating multiscale convolutional neural network (MSCNN) and bidirectional gated recurrent unit (BiGRU) is proposed. MSCNN and BiGRU are used to extract multiscale features and temporal features from noisy vibration signals respectively, and different weights are assigned to the fused features through the attention mechanism module to achieve important feature selection. Furthermore, in order to solve the problem of different feature distributions between the source domain and the target domain under variable working conditions, transfer learning is introduced in the proposed deep feature extraction network. The difference in feature distribution between the source domain and the target domain is measured by a multi-level distance formula, and the measurement result is added to the loss function. The back propagation of the loss is used to achieve the alignment of feature distribution between the source domain and the target domain. Finally, the model uses the SoftMax function as a classifier for rolling bearing fault diagnosis. Experimental comparison and analysis show that the proposed model has good migration ability and achieves a higher fault identification accuracy.

Noise reduction method for wind turbine gearbox vibration signals based on CVMD-DRDSAE

Abstract Wind turbine gearbox fault feature extraction is difficult due to strong background noise. To address this issue, a noise reduction method combining comprehensive learning particle swarm optimization-variational mode decomposition (CLPSO-VMD) and deep residual denoising self-attention autoencoder (DRDSAE) is proposed. Firstly, the proposed CLPSO-VMD algorithm is used to decompose the noisy wind turbine gearbox vibration signals. Subsequently, the intrinsic mode functions highly correlated with the original signals are selected through the Spearman correlation coefficient and utilized for signal reconstruction, thereby filtering out high-frequency noise outside the fault frequency band in the frequency domain characterization. Secondly, the improved DRDSAE is utilized to learn the latent representations of data in the first-level denoised signal, further reducing the residual noise within the fault frequency band while retaining important signal features. Finally, the envelope spectrum highlights the weak feature of the wind turbine gearbox vibration signal. Experimental results demonstrate the effectiveness of the proposed method in denoising wind turbine gearbox vibration signals under strong noise.

Shallow water source depth discrimination based on a vertical linear array using deep learning

This study aims to explore source depth discrimination in shallow water waveguides using a vertical line array. Due to the similarity in physical characteristics of features and environmental parameters, the feature distributions of shallow and deep sources overlap in feature space. This overlap is further exacerbated by strong background noise, reducing detection reliability. Therefore, a deep learning-based source depth discrimination (DL-SDD) scheme is proposed. Specifically, this scheme is based on a residual structure, embedding channel attention mechanisms into the deep structure to eliminate noise-related information gradually. Furthermore, a specially designed loss function considers inter-class and intra-class distances to achieve compact and mutually distant distributions of source features. When this loss function is applied, the overlap of source feature distributions is suppressed in end-to-end feature learning, leading to a high detection probability. The numerical simulations demonstrate that the proposed DL-SDD outperforms traditional method, achieving a 7 % increase in detection probability and a 15 % decrease in false alarm rate, even near the discrimination depth. Furthermore, the discrimination depth is reduced by nearly half. Experimental results from the South China Sea validate the effectiveness of the proposed DL-SDD.

The Self-Powered Thermoelectric System Driving an Electric Field for Raman-Spectrum Glucose Quantitative Analysis

For a long time, blood glucose monitoring has been a challenging issue in the field of medicine. Although it is possible to visit a hospital for regular blood tests to track blood sugar concentration, the physiological reactions triggered by blood collection might lead to feeling weak, dizzy, or even vomiting. Therefore, especially for diabetic patients, they need to frequently monitor their blood sugar concentration in their daily lives to ensure their health. Currently, the primary methods for blood glucose testing still rely on invasive techniques, such as using fingertip glucose meters and other methods which require skin puncture to obtain blood samples. Although these methods provide a quick assessment of current blood glucose concentration, they come with the pain and discomfort of blood collection process and the potential risk of infection. As a result, in recent years, many researchers have been dedicated to developing non-invasive blood glucose monitoring methods. However, accurately measuring glucose concentration inside the human body remains a challenging task. One of the primary issues is the presence of strong noise and background noise. To avoid this challenge, we have developed an innovative detection platform that is combines surface-enhanced Raman spectroscopy (SERS) and external electric field sensing technology. With the electromagnetic effect, we add external electric field in our platform to excite more the surface plasmon resonance. By using this method, we can achieve signal amplification to detect lower concentrations of glucose. Now, we can detect down to 0.001mM of glucose. We also exhibit self-powered electric field system, using the flexible themoelctric device to combine our platform. By the property of self-powered, we success to create a self-powered electric field system. We focus on driving themoelectric through the temperature different between body temperature and environment. The approach is aimed at generating power from the human body, providing a self-powered thermoelectric system for Raman-spectrum glucose sensors. This innovative approach holds promise for advancing non-invasive blood glucose monitoring and benefiting a larger population of diabetes patients in the future.

An adaptive feature mode decomposition-guided phase space feature extraction method for rolling bearing fault diagnosis

Abstract The extraction of fault features from rolling bearings is a challenging and highly important task. Since they have complex operating conditions and are usually under a strong noise background. In this study, a novel approach termed phase space feature extraction guided by an adaptive feature mode decomposition (AFMDPSFE) is proposed to detect subtle faults in rolling bearings. Initially, a new method using Kullback-Leiber divergence is introduced to automatically select the optimal mode number and filter length for the decomposition of vibration signals, facilitating the automatic extraction of optimal components and ensuring efficient screening. This eliminates the need for manual configuration of feature mode decomposition parameters. Furthermore, a criterion that could determine two crucial parameters to capture system dynamics characteristics in phase space reconstruction is embedded into AFMDPSFE algorithm. Subsequently, a series of high-dimensional independent components is derived. The envelope spectrum of the principal component exhibiting the highest kurtosis value is computed to achieve fault identification, consequently enhancing the separation of signal from noise. Both simulations and experimental results confirm the effectiveness of AFMDPSFE approach. A comparison analysis shows the excellent performance of AFMDPSFE in extracting fault features from significant noise interference.

Research on remote reference denoising method based on non-coaxial and non-coplanar tunnel NMR detection

Abstract The limited space within the tunnel constrains the size of the antenna for NMR detection, thereby significantly impacting the signal-to-noise ratio (SNR) of NMR signals. Insufficient SNR data poses substantial challenges to obtaining reliable NMR signals. The paper presents a novel approach to address the challenge of strong background noise in tunnel environments and low SNR data by incorporating the ground multi-channel remote reference denoising method into tunnel NMR advance detection. Specifically designed for narrow tunnels, a multi-channel non-coaxial and non-coplanar remote reference denoising method is proposed. Firstly, the effectiveness of the non-coaxial, non-coplanar remote reference denoising method is verified in the laboratory environment. Secondly, the correlation between the detector antenna and the reference antenna is calculated theoretically to ensure the significant correlation between the detector antenna and the reference antenna. Finally, two processing methods of reference denoising and non-reference denoising are carried out respectively by combining the tunnel detection data. By comparing the inversion results and the engineering construction results, the effectiveness of non-coaxial and non-coplanar remote reference denoising methods in tunnel NMR detection is proved, which provides relevant research support for expanding the application of tunnel NMR detection technology.

Enhanced Nanoparticle Recognition via Deep Learning-Accelerated Plasmonic Sensing.

Surface plasmon microscopy proves to be a potent tool for capturing interferometric scattering imaging data of individual particles at both micro and nanoscales, offering considerable potential for label-free analysis of bio-particles and bio-molecules such as exosomes, viruses, and bacteria. However, the manual analysis of acquired images remains a challenge, particularly when dealing with dense samples or strong background noise, common in practical measurements. Manual analysis is not only prone to errors but is also time-consuming, especially when handling a large volume of experimental images. Currently, automated methods for sensing and analysis of such data are lacking. In this paper, we develop an accelerated approach for surface plasmon microscopy imaging of individual particles based on combining the interference scattering model of single particle and deep learning processing. We create hybrid datasets by combining the theoretical simulation of particle images with the actual measurements. Subsequently, we construct a neural network utilizing the EfficientNet architecture. Our results demonstrate the effectiveness of this novel deep learning technique in classifying interferometric scattering images and identifying multiple particles under noisy conditions. This advancement paves the way for practical bio-applications through efficient automated particle analysis.