Time Frequency Domain Features Research Articles

Joint Feature Selection of Power Load in Time Domain and Frequency Domain Based on Whale Optimization Algorithm

Accurate classification of power load is the premise of demand-side response capability evaluation, which is of great significance for demand-side management. Therefore, this study proposes a selection method of joint characteristics of power load in time domain and frequency domain based on whale optimization algorithm. The electrical measurement data of six typical electrical equipment are selected to form the original power load identification dataset. Firstly, the time domain features and frequency domain features of the power load data are extracted, and then the joint characteristics of the power load are screened by Whale optimization algorithm (WOA). Finally, the selected feature information is used as the input to verify the performance of WOA on the joint characteristics of power load under back propagation neural network (BP), extreme learning machine (ELM), support vector machine (SVM), k-nearest neighbor (KNN), decision tree (DT), and naive Bayesian (NB) classifiers. The results show that WOA can effectively screen out the 15 most helpful feature attributes for power load identification, which can not only improve the accuracy of power load identification but also effectively reduce the time cost of the algorithm, which is of reference value for further demand-side response strategy. At the same time, it is of great significance for intelligent dispatching of power system and improving the economic operation of the power system.

A novel fault diagnosis method for early faults of PMSMs under multiple operating conditions

Early fault diagnosis method under multiple operating conditions (OCs) is very important to improve the reliability of the permanent magnet synchronous motors (PMSMs) system. In this paper, a novel early fault diagnosis method for irreversible demagnetization fault and inter-turn short fault under multiple OCs in PMSM is proposed. Multi-layer self-attention mechanism is constructed to capture the long time dependency from the sensor signal, which can fuse the features of individual OCs to diagnose faults under unseen OCs. Furthermore, local feature extraction structure with end-to-end deep learning features and time–frequency domain features is proposed to extract effective features from early faults. The proposed method uses only two-phase current signals as the inputs to diagnose the faults. By comparing 1D convolutional neural network structure with multi-variables and universal machine learning methods, the experimental results show that the proposed method achieves high accuracy for early fault diagnosis of PMSM in unseen OCs.

Series Dc arc fault detection and location in wind-solar-storage hybrid system based on variational mode decomposition

As the main new energy, the wind-solar-storage hybrid system is widely used because of its excellent complementarity. However, due to the complexity of the system, series DC arc faults are prone to occur. This paper presents a method for series DC arc fault detection and location in wind-solar-storage hybrid system, which works mainly on the principle of threshold comparison in the time-frequency domain. Therein, an algorithm called variational mode decomposition (VMD) is employed. The voltage signal of the identical DC load under different operating conditions is decomposed by VMD, and the time-frequency domain characteristics of the normal operating conditions are used as thresholds. The series DC arc faults in different subsystems are detected and located by comparing the time-frequency domain features under different operating conditions with the threshold features. Compared with classical decomposition algorithms, Empirical Mode Decomposition (EMD) and Wavelet Transform (WT), the results show that VMD can eliminate the influence of noise and extract the fault signals of different systems more accurately, ensuring the precise localization of series DC arc faults.

Machine Learning Based Real-Time Diagnosis of Mental Stress Using Photoplethysmography

Mental stress is a natural response to life activities. However, acute and prolonged stress may cause psychological and heart diseases. Heart rate variability (HRV) is considered an indicator of mental stress and physical fitness. The standard way of obtaining HRV is using electrocardiography (ECG) as the time interval between two consecutive R-peaks. ECG signal is collected by attaching electrodes on different locations of the body, which need a proper clinical setup and is costly as well; therefore, it is not feasible to monitor stress with ECG. Photoplethysmography (PPG) is considered an alternative for mental stress detection using pulse rate variability (PRV), the time interval between two successive peaks of PPG. This study aims to diagnose daily life stress using low-cost portable PPG devices instead of lab trials and expensive devices. Data is collected from 27 subjects both in rest and in stressed conditions in daily life routine. Thirty-six time domain, frequency domain, and non-linear features are extracted from PRV. Multiple machine learning classifiers are used to classify these features. Recursive feature elimination, student t-test and genetic algorithm are used to select these features. An accuracy of 72% is achieved using stratified leave out cross-validation using K-Nearest Neighbor, and it increased up to 81% using a genetic algorithm. Once the model is trained with the best features selected with the genetic algorithm, we used the trained weights for the real-time prediction of mental stress. The results show that using a low-cost device; stress can be diagnosed in real life. The proposed method enable the regular monitoring of stress in short time that help to control the occurrence of psychological and cardiovascular diseases.

Effect of audiovisual stimulation on adult memory performance based electroencephalography wavelet analysis

Human memory stores various information and events that can be retrieved when needed. Many factors can influence memory performance which either provide positive or negative feedback. This research investigates the effect of audiovisual stimulation on adult memory based on electroencephalography (EEG) analysis. Sixty college students are participating in this experimental study. They must memorize visual assessment at two different levels in Mozart’s Sonata music and white noise stimulation. During memorizing duration, the EEG machine records brain electrical activity based on 10–20 electrode placement. The collected raw brain signals are processed using the wavelet-based method. The stationary wavelet transform (SWT) is used for artifact elimination, whereas discrete wavelet transform (DWT) is applied to obtain alpha, beta, theta, and gamma rhythms. The time–frequency domain features are collected from the EEG signals to discover the influence of audiovisual stimulation. The findings showed a different increasing and decreasing trend of mean, standard deviation, and peak-to-peak EEG signal amplitude before and after audiovisual stimulation exposure. The theta and alpha rhythms showed the most influence with the highest relative power. Suppression of relative gamma and beta power is vital for improving visual information processing and attention level. Memorizing in audio stimulation has suppressed the relative alpha, beta, theta, and gamma power, leading to better visual memorizing ability. The white noise stimulation provides more influence on adult visual memory.

Multi-Sensor GA-BP Algorithm Based Gearbox Fault Diagnosis

To address the problem of the low recognition rate of time-frequency domain methods gearbox fault identification, a method featuring decision-level fusion of DS evidence theory and GA-BP algorithm was proposed in the present study. Firstly, the fault data of each state of the gearbox was classified, based on which the time-frequency domain features were extracted and 19 significant features have been selected. Secondly, the accuracy of the traditional BP algorithm was compared with that of the GA-BP algorithm. On this basis, it has been concluded that the GA-BP algorithm is highly accurate, and the local diagnostic results obtained by the GA-BP algorithm have been used as the basic probability. Finally, the DS evidence theory is currently used to fuses with the GA. In addition, the final fault identification of the gearbox can be achieved by using the DS evidence theory and the multi-sensor local diagnosis results obtained by the GA-BP algorithm for decision fusion. The results of the simulations and experiments showed that the method proposed has improved accuracy over a single algorithm for fault identification of gearboxes, respectively.

Modeling Brain Volume Using Deep Learning-Based Physical Activity Features in Patients With Dementia.

There is a proven correlation between the severity of dementia and reduced brain volumes. Several studies have attempted to use activity data to estimate brain volume as a means of detecting reduction early; however, raw activity data are not directly interpretable and are unstructured, making them challenging to utilize. Furthermore, in the previous research, brain volume estimates were limited to total brain volume and the investigators were unable to detect reductions in specific regions of the brain that are typically used to characterize disease progression. We aimed to evaluate volume prediction of 116 brain regions through activity data obtained combining time-frequency domain- and unsupervised deep learning-based feature extraction methods. We developed a feature extraction model based on unsupervised deep learning using activity data from the National Health and Nutrition Examination Survey (NHANES) dataset (n = 14,482). Then, we applied the model and the time-frequency domain feature extraction method to the activity data of the Biobank Innovations for chronic Cerebrovascular disease With ALZheimer’s disease Study (BICWALZS) datasets (n = 177) to extract activity features. Brain volumes were calculated from the brain magnetic resonance imaging of the BICWALZS dataset and anatomically subdivided into 116 regions. Finally, we fitted linear regression models to estimate each regional volume of the 116 brain areas based on the extracted activity features. Regression models were statistically significant for each region, with an average correlation coefficient of 0.990 ± 0.006. In all brain regions, the correlation was > 0.964. Particularly, regions of the temporal lobe that exhibit characteristic atrophy in the early stages of Alzheimer’s disease showed the highest correlation (0.995). Through a combined deep learning-time-frequency domain feature extraction method, we could extract activity features based solely on the activity dataset, without including clinical variables. The findings of this study indicate the possibility of using activity data for the detection of neurological disorders such as Alzheimer’s disease.

A Deep Ensemble-Based Wireless Receiver Architecture for Mitigating Adversarial Attacks in Automatic Modulation Classification

Deep learning-based automatic modulation classification (AMC) models are susceptible to adversarial attacks. Such attacks inject specifically crafted wireless interference into transmitted signals to induce erroneous classification predictions. Furthermore, adversarial interference is transferable in black box environments, allowing an adversary to attack multiple deep learning models with a single perturbation crafted for a particular classification model. In this work, we propose a novel wireless receiver architecture to mitigate the effects of adversarial interference in various black box attack environments. We begin by evaluating the architecture uncertainty environment, where we show that adversarial attacks crafted to fool specific AMC DL architectures are not directly transferable to different DL architectures. Next, we consider the domain uncertainty environment, where we show that adversarial attacks crafted on time domain and frequency domain features to not directly transfer to the altering domain. Using these insights, we develop our Assorted Deep Ensemble (ADE) defense, which is an ensemble of deep learning architectures trained on time and frequency domain representations of received signals. Through evaluation on two wireless signal datasets under different sources of uncertainty, we demonstrate that our ADE obtains substantial improvements in AMC classification performance compared with baseline defenses across different adversarial attacks and potencies.

Measurement and prediction of wear volume of the tool in nonlinear degradation process based on multi-sensor information fusion

Excessive tool wear seriously affects the surface quality of the workpiece and reduces the processing efficiency. Therefore, real-time monitoring of the tool wear status is very important. This paper proposes a new tool wear prediction method based on multi-sensor hybrid domain information fusion. First, the collected multi-sensor signals are decomposed by wavelet packet to extract the energy values of 16 frequency bands; in addition, the energy values are combined with the time domain and frequency domain features to construct a hybrid domain feature set; then, the gated recurrent unit model is used to adaptively explore the internal relationship between the hybrid domain features and tool wear, which overcomes the low efficiency of manual feature fusion monitoring; finally, the wear milling cutter data is used to verify the superiority of the proposed method. The results show that the prediction accuracy of tool wear based on multi-sensor feature fusion is significantly better than that based on a single sensor. Also, compared with traditional wear prediction methods, it again verifies the advancement of the proposed method in predicting tool wear.

Analysis of Dynamic System of Exercise Load Condition Monitoring Based on Characteristic Parameters

According to the algorithm of time difference and threshold value, this paper selects the more valuable data for motion state recognition and selects the characteristics, respectively selects the data of the combined acceleration and the combined angular velocity, and uses the data of the pitch angle and the roll angle more novelly. In the aspect of data preprocessing, the sliding segmentation window method is used for feature processing, and the time domain and frequency domain features of the data are extracted. A total of 108 dimensional features are extracted. In order to improve the calculation performance, PCA technology is used for data dimensionality reduction. In this paper, we collected data on changes in physiological parameters of 24 experimenters before and after exercise, collected 14 self-evaluated severely fatigued volunteers and self-evaluated severely stressed volunteers’ resting heart rate and blood pressure data as unhealthy data samples, and collected physiological data of 14 healthy experimenters as unhealthy data samples. For healthy samples, three sets of experiments were set up to analyze the changes of exercise heart rate, exercise blood pressure and exercise body temperature, and the effectiveness of fusion of physiological data to improve the performance of exercise recognition and the analysis of the health status of physiological parameters that introduce exercise interference. The experimental results show that during exercise, monitoring changes in systolic blood pressure is more meaningful than monitoring changes in diastolic blood pressure; it verifies the effectiveness of improving the performance of exercise recognition by fusion of physiological parameters. The addition of physiological data can effectively improve the recognition rate of exercise. The recognition rate has been increased from 93.7% to 96.3%; the effectiveness and applicability of the algorithm in this paper are analyzed through design experiments, and the results show that the recognition accuracy of the algorithm in this paper is above 87%. This result has a good classification recognition rate for a small sample.

Combined Remaining Life Prediction of Multiple Bearings Based on EEMD-BILSTM

To improve the accuracy of a symmetrical structural rolling bearing life prediction under noise interference, a multi-bearing life prediction method combining Ensemble Empirical Mode Decomposition (EEMD) and Bi-directional Long Short-Term Memory (BiLSTM) is proposed. First, EEMD is proposed to decompose the original vibration signal to obtain a finite number of Intrinsic Mode Function (IMF), and the IMFs are further filtered by combining the correlation criterion and kurtosis criterion. Then, the time domain features and frequency domain features of the reconstructed signal are filtered by monotonicity index to obtain the set of features containing key information. Finally, the BiLSTM network is trained on the filtered features set, and the method is proved to accurately predict the remaining life of rolling bearings under different operating conditions through rolling bearing full-life experiments, and the effectiveness of the method is verified by comparing the prediction results with several main recurrent neural networks.

Acoustic Signal Classification Using Symmetrized Dot Pattern and Convolutional Neural Network

The classification of sound signals can be applied to the fault diagnosis of mechanical systems, such as vehicles. The traditional sound classification technology mainly uses the time-frequency domain characteristics of signals as the basis for identification. This study proposes a technique for visualizing sound signals, and uses artificial neural networks as the basis for signal classification. This feature extraction method mainly uses a principle to convert a time domain signal into a coordinate symmetrized dot pattern, and presents it in the form of snowflakes through signal conversion. To verify the feasibility of this method to classify different noise characteristic signals, the experimental work is divided into two parts, which are the identification of traditional engine vehicle noise and electric motor noise. In sound measurement, we first use the microphone and data acquisition system to measure the noise of different vehicles under the same operating conditions or the operating noise of different electric motors. We then convert the signal in the time domain into a symmetrized dot pattern and establish an acoustic symmetrized dot pattern database, and use a convolutional neural network to identify vehicle types. To achieve a better identification effect, in the process of data analysis, the effect of the time delay coefficient and weighting coefficient on the image identification effect is discussed. The experimental results show that the method can be effectively applied to the identification of traditional engine and electric vehicle classification, and can effectively achieve the purpose of sound signal classification.

Machine learning-based automatic sleep apnoea and severity level classification using ECG and SpO2 signals

Sleep apnoea is a potentially serious sleep disorder that is characterised by repetitive episodes of breathing interruptions. Traditionally, sleep apnoea is commonly diagnosed in an attended sleep laboratory setting using polysomnography (PSG). The manual diagnosis of sleep apnoea using PSG is, however complex, and time-consuming, as many physiological variables are usually measured overnight using numerous sensors attached to patients. In PSG sleep laboratories, an expert human observer is required to work overnight, and the diagnosis accuracy is dependent on the physician’s experience. A quantitative and objective method is required to improve the diagnosis efficacy, decrease the complexity and diagnosis time and to ensure a more accurate diagnosis. The purpose of this study was then to develop an automatic sleep apnoea and severity classification using a simultaneously recorded electrocardiograph (ECG) and saturation of oxygen (SpO2) signals based on a machine learning algorithm. Different ECG and SpO2 time domain and frequency domain features were extracted for training different machine learning algorithms. For sleep apnoea classification, an accuracy of 99.1%, specificity of 98.1% and sensitivity of 100% were achieved using a support vector machine (SVM) based on combined ECG and SpO2 features. Similarly, for severity classification, an 88.9% accuracy, 90.9% specificity and 85.7% sensitivity have been obtained. For both apnoea and severity classification, using the combined features was found to be more accurate, and this is typically important when either channel is poor quality, the system can make an analysis based on the other channel and achieve good accuracy.

Acoustic Emission Intelligent Identification for Initial Damage of the Engine based on Single Sensor

Detecting the initial damage of the engine precisely is conducive to finding out the failure symptom timely and ensuring the reliable operation of the engine for avoiding malignant accidents. This article proposes a single-sensor acoustic emission (AE) diagnosis method based on deep learning (DL) for structural health monitoring (SHM) and initial damage intelligent identification of the engine. Firstly, the time-frequency domain features of AE signals collected from the engine test bench under different initial damages are extracted by continuous wavelet transform (CWT), and two-dimensional time-frequency images are constructed using the obtained results. After that, the proposed deep convolutional neural networks (DCNN) is employed to directly extract damage features from the converted two-dimensional time-frequency images via a multi-layer fusion and perform initial damage intelligent identification. In addition, the white Gaussian noise with different signal-to-noise ratio (SNR) is added to the raw AE signal, and the robustness of the proposed DCNN method is verified. Finally, to evaluate the performance of the proposed DCNN, the constructed test accuracy Euclidean distance and other indexes including Kappa, microF1 and macroF1 are introduced to compare with other commonly machine learning methods. The results demonstrated the effectiveness of the proposed DCNN model, which lays the foundation for the application of the single-sensor AE diagnosis method based on DCNN in SHM and initial damage intelligent identification of the engine.

Objective Recognition of Tinnitus Location Using Electroencephalography Connectivity Features.

Purpose: Tinnitus is a common but obscure auditory disease to be studied. This study will determine whether the connectivity features in electroencephalography (EEG) signals can be used as the biomarkers for an efficient and fast diagnosis method for chronic tinnitus.Methods: In this study, the resting-state EEG signals of tinnitus patients with different tinnitus locations were recorded. Four connectivity features [including the Phase-locking value (PLV), Phase lag index (PLI), Pearson correlation coefficient (PCC), and Transfer entropy (TE)] and two time-frequency domain features in the EEG signals were extracted, and four machine learning algorithms, included two support vector machine models (SVM), a multi-layer perception network (MLP) and a convolutional neural network (CNN), were used based on the selected features to classify different possible tinnitus sources.Results: Classification accuracy was highest when the SVM algorithm or the MLP algorithm was applied to the PCC feature sets, achieving final average classification accuracies of 99.42 or 99.1%, respectively. And based on the PLV feature, the classification result was also particularly good. And MLP ran the fastest, with an average computing time of only 4.2 s, which was more suitable than other methods when a real-time diagnosis was required.Conclusion: Connectivity features of the resting-state EEG signals could characterize the differentiation of tinnitus location. The connectivity features (PCC and PLV) were more suitable as the biomarkers for the objective diagnosing of tinnitus. And the results were helpful for clinicians in the initial diagnosis of tinnitus.

Radar HRRP Target Recognition Method Based on Multi-Input Convolutional Gated Recurrent Unit With Cascaded Feature Fusion

Over the past decades, radar high-resolution range profile (HRRP) has been one of the research highlights in the field of radar automatic target recognition (RATR) due to its advantages of easy acquisition, small amount of data, and rich target structure information. However, most of existing methods only consider its amplitude (time domain) characteristics, thereby neglecting the temporal dependence and multi-domain features inside the HRRP sequence. To this end, we propose an end-to-end multi-input convolutional gated recurrent unit neural network, called MIConvGRU, for RATR by both exploiting the multi-domain and temporal information to improve the recognition performance of HRRP target. Initially, the data-preprocessing module is employed to extract the multi-domain features of the target, including time domain, frequency domain, and time-frequency domain features, in order to further enhance the target representation. In addition, a cascaded multi-input GRU structure is designed to acquire the multi-domain temporal dependence feature of HRRP sequence from low to high level. Finally, these temporal features are adaptively fused by a parameter learnable strategy. The experimental results show that the proposed MIConvGRU can effectively learn the multi-domain temporal dependence correlation features in HRRP sequences, improving the target recognition performance.

Online Evaluation of Surface Hardness for Aluminum Alloy in LSP Using Modal Acoustic Emission

This article researches the online evaluation method of surface integrity for aluminum alloy in laser shock peening (LSP) based on multiple-sources modal acoustic emission (MAE). First, a new comprehensive index e.g., sub-surface microhardening rate (SHR) for 7075 aluminum alloy in LSP was constructed to characterize both the impact depth (<inline-formula> <tex-math notation="LaTeX">$150 \mu \text{m}$ </tex-math></inline-formula>) and its hardening rate. Then, a good linear and positive correlation between the SHR of 7075 Al alloy and the AE features in time domain and frequency domain was revealed based on the offline quantitative characterization of SHR and the antisymmetric zero (A0) mode separated from AE signal. In addition, the A0 mode-based Mel frequency cepstrum coefficient (AMFCC) was proposed to deeply extract the time-frequency features of MAE-A0 mode signal. Finally, the SHR classification model based on AMFCC and gradient boosting decision tree for 7075 Al alloy in LSP process was established and was carefully validated using the experimental data. The test results showed the highest mean accuracy of 96.24&#x0025; after the thorough comparison with other tree models and the traditional cepstrum methods. More importantly, deep analysis about the feature importance and their physical interpretation has been conducted from the perspective of non-linear effect inside the LSP material caused by dislocation multiplication as well as the increasing harmonic component of the MAE-A0 signal. This article can provide certain guidance for the development of LSP quality monitoring technology.

Automated Heart Valve Disorder Detection Based on PDF Modeling of Formant Variation Pattern in PCG Signal

Heart valve disorder (HVD) analysis from heart sound is being well known for a long period of time, and use of digital stethoscope gives opportunity to diagnose HVDs from phonocardiographic (PCG) signal. An automated HVD detection technique from PCG signal can play a key role as a first-hand diagnostic tool for the physicians. In this paper, in order to classify different HVDs, we propose to utilize the formant characteristic of the PCG signal, which is an acoustic property of the heart sound. PCG signals exhibit significant variations depending on different types of HVDs and thus conventional time frequency domain features or statistical features are extracted from PCG signal for disease classification. However, direct PCG signals are also used in sequential networks to classify HVDs. Similar to the formant peaks of voiced speech signal, the spectrum corresponding to the PCG signal exhibits distinguishable peaks, especially in the voiced part of the heart sound (lub-dub). Keeping this notable key point in consideration, Burg’s autoregressive model is used to find the parametric spectrum of the PCG signal. The first two formants of the PCG signal, that carry the most informative acoustic properties of the heart sound, are estimated from the Burg’s spectrum, and are used for feature extraction. The magnitude, frequency and phase of each formant are considered to evaluate these features. Instead of considering a long duration of PCG signal at a time, we consider the overlapping sub-frames, and extract formants from each sub-frame, which generates a temporal variation of the formants. Finally, we propose a PDF model fitting of the formant variation, and utilize the estimated model parameters along with some statistical features to classify the HVDs. Two famous publicly available PCG datasets are used to demonstrate the performance of the proposed method, that efficiently classify the binary/five classes of heart sounds. The results reveal that the proposed method has the overall accuracy values of 93.46% and 99.28% for the two datasets, which is better in comparison to other previously reported state-of-the-art techniques.

Iterative Outlier Removal Clustering Based Time-Frequency-Spatial Feature Selection for Binary EEG Motor Imagery Decoding

Electroencephalography (EEG)-based motor imagery (MI) brain-computer interface (BCI) is a bridge in the instruments of rehabilitation and motor assistance field to control external assist devices without external stimulation. Feature extraction (FE) is a key to improving the performance of EEG MI decoding. Existing FE methods usually extract single-domain features such as spatial, frequency features, or dual-domain features such as time–frequency features. However, the multidomains features which describe the intent more comprehensively are not simultaneously extracted. Therefore, inspired by short-time Fourier transform (STFT) and common spatial pattern (CSP) FE methods, a frequency-spatial-temporal multidomain (FSTMD) FE method is proposed. To solve the problem of the high dimensions of the FSTMD features, a feature selection (FS) method is designed. In the proposed FSTMD FE process, the STFT is employed to extract the time–frequency domain features first. Then, the time–frequency features of each channel are cascaded. After that, the cascaded features are divided according to time segments. At last, the covariance transform is performed for each time segment. Since the proposed features have a one-to-one correspondence with frequency, channel, and time, the extracted features have good interpretability. The proposed FS method selects the features in the frequency-spatial domain of each time segment based on mutual information and correlation metrics first and then selects the key time segment based on clustering with outlier removal. The performance of the proposed feature is verified in BCI Competition IV Dataset IIa and IIb. Compared to the CSP feature in the only spatial domain, the proposed feature has significantly lower deviation and slightly higher accuracy. The average recognition accuracy of the proposed MI decoding framework is 85.1% on the dataset IIb session 3, which is competitive compared with the state-of-the-art methods.

Fault Analysis of Wind Power Rolling Bearing Based on EMD Feature Extraction

In a wind turbine, the rolling bearing is the critical component. However, it has a high failure rate. Therefore, the failure analysis and fault diagnosis of wind power rolling bearings are very important to ensure the high reliability and safety of wind power equipment. In this study, the failure form and the corresponding reason for the failure are discussed firstly. Then, the natural frequency and the characteristic frequency are analyzed. The Empirical Mode Decomposition (EMD) algorithm is used to extract the characteristics of the vibration signal of the rolling bearing. Moreover, the eigenmode function is obtained and then filtered by the kurtosis criterion. Consequently, the relationship between the actual fault frequency spectrum and the theoretical fault frequency can be obtained. Then the fault analysis is performed. To enhance the accuracy of fault diagnosis, based on the previous feature extraction and the time-frequency domain feature extraction of the data after EMD decomposition processing, four different classifiers are added to diagnose and classify the fault status of rolling bearings and compare them with four different classifiers.