Time Domain Statistical Features Research Articles

An ensemble maximal feature subset selection for smartphone based human activity recognition

Smartphone-based Human Activity Recognition (HAR) uses spatiotemporal time series data collected from a smartphone’s in-built accelerometer, gyroscope, and magnetometer sensor. Real-time HAR datasets such as UCI-HAR and UCI-HAPT extract time and frequency domain statistical features for activity recognition from the collected time series signals. Various Machine Learning techniques have been proposed in the perspective of feature selection through ensemble, traditional, hybrid, and meta-heuristic-based techniques to improve the recognition performance. Though the feature selection techniques have shown promising results, the computational time and selection of valid significant features for better classification performance still need to be included. Moreover, these techniques are highly parameter-dependent and critical to threshold values used for experimentation. This research proposes a novel method of parameter-free ensemble feature selection by combining traditional feature selection algorithms with the concept of association rule mining to obtain the Maximal Feature Subset (MFS), which is used for training and testing the classification model for HAR. Mathematical association rule mining is framed rather than the traditional one to select the associated frequent feature(s) from the feature set evolved by traditional techniques. Experimentation has been carried out on UCI-HAR and UCI-HAPT, and the proposed method has increased the classification efficiency by selecting significant features and reduces computation time significantly.

A Method of Diagnosing Analog Circuit Soft Faults Using Boruta Features and LightGBM

Modern electronic power systems rely heavily on analog circuits. The accurate detection of analog circuit faults, especially soft faults, is of great significance for the maintenance and inspection of electronic systems. This paper proposes the application of the Boruta feature selection method to the field of the soft fault diagnosis of analog circuits to screen out low-dimensional and efficient feature components from the high-dimensional time-domain statistical features and frequency-domain features of circuit responses. Then, the feature components are used as the input to train the LightGBM classification model, and the Bayesian optimization method is introduced to optimize the model’s hyperparameters. Finally, the trained fault diagnosis model is verified in two typical experimental circuits, and satisfactory accuracy is obtained.

Fault diagnosis of injection molding machine non-return valve based on data-driven model

Non-return valve (NRV) is one of the key components in determining the consistency of the quality of the products molded by injection molding machine. Wear on the NRV affects the quality of the molded product. Nevertheless, detecting wear on the NRV can be challenging and disassembly of the machine is the only diagnostic method, which can have a negative impact on productivity. In this paper, a data-driven fault diagnosis method is proposed, which uses Stacked Auto Encoder (SAE) to analyze the pressure, torque, and displacement signals of the injection molding machine and combined with XGBoost (Extreme Gradient Boosting) to diagnose the faults of the NRV. The experimental results indicate that the SAE-XGBoost method accurately predicts NRV failures. Compared to using only XGBoost for prediction, the accuracy has improved from 97.5% to 99.6%. Eventually, the SAE-XGBoost model is integrated into the control program of the injection molding machine in the form of functional modules. Throughout the production process, the model adeptly monitors and identifies the production profile, promptly dispatching warning messages to users when diagnosing NRV wear. This facilitates intelligent diagnosis of the service status of injection molding machine components, which will have a positive influence on improving the production efficiency and intelligence of injection molding machines. The results of this study represent a synergistic application of artificial intelligence and time-domain statistical features in the realm of fault diagnosis for injection molding machines. This has the potential to significantly broaden the scope of AI utilization within the domain of injection molding processes, thereby advancing the intelligent technology associated with injection molding machines.

Inovativna primena veštačkih neuronskih mreža za efikasnu lokalizaciju prslina rotacionog vratila

Rotational shafts are pivotal components in industrial settings and are responsible for transmitting torque and rotational motion. Despite their significance, these shafts are susceptible to faults, particularly cracks, which can adversely affect the system's performance and safety. Hence, efficient crack detection and diagnosis ensure safety, reliability, and costeffectiveness. This research aims to develop an Artificial Neural Network (ANN) model that can effectively identify cracks occurring at different depths and locations in rotating shafts, which operate at varying rotational speeds. Vibration signals were obtained and subjected to preprocessing using a bandpass filter to isolate the shaft signals from other components. Subsequently, time-domain statistical features were extracted from the filtered signals. An optimal feature selection methodology was employed to rank the extracted features, and the highest-ranking features were chosen for training the ANN model. The findings of this research indicate that the developed model achieved a classification accuracy of 94.4%.

Deep Learning-Based Positioning Error Fault Diagnosis of Wire Bonding Equipment and an Empirical Study for IC Packaging

Little research has been done for artificial intelligence applications of semiconductor backend. This study aims to develop a deep learning based fault diagnosis framework as prognostics and health management (PHM) solutions with a comprehensive analytics process. A linear encoder sensor is employed to measure position, while data preparation is conducted to convert position information into a signal. Then, signal processing is used to extract the features, in which high pass filter is applied to normalize the signal and emphasize the features. High-dimensional features including time domain statistical features, time-frequency domain features obtained by continuous wavelet transform (CWT), and frequency domain features converted by fast Fourier transform (FFT) are extracted to prepare the dataset. An ensemble neural network model that integrates deep neural network (DNN) and convolutional neural network (CNN) is employed to recognize the pattern and diagnose the positioning errors. The accuracy and false alarm rate are considered to support maintenance decisions. An empirical study was conducted in a world leading IC assembly and testing company for validation. The results have shown that the proposed approach can effectively detect inappropriate installation of the bond head in wire bonding equipment for predictive maintenance and cost reduction.

Proactive Fault Diagnosis of a Radiator: A Combination of Gaussian Mixture Model and LSTM Autoencoder.

Radiator reliability is crucial in environments characterized by high temperatures and friction, where prompt interventions are often required to prevent system failures. This study introduces a proactive approach to radiator fault diagnosis, leveraging the integration of the Gaussian Mixture Model and Long-Short Term Memory autoencoders. Vibration signals from radiators were systematically collected through randomized durability vibration bench tests, resulting in four operating states-two normal, one unknown, and one faulty. Time-domain statistical features of these signals were extracted and subjected to Principal Component Analysis to facilitate efficient data interpretation. Subsequently, this study discusses the comparative effectiveness of the Gaussian Mixture Model and Long Short-Term Memory in fault detection. Gaussian Mixture Models are deployed for initial fault classification, leveraging their clustering capabilities, while Long-Short Term Memory autoencoders excel in capturing time-dependent sequences, facilitating advanced anomaly detection for previously unencountered faults. This alignment offers a potent and adaptable solution for radiator fault diagnosis, particularly in challenging high-temperature or high-friction environments. Consequently, the proposed methodology not only provides a robust framework for early-stage fault diagnosis but also effectively balances diagnostic capabilities during operation. Additionally, this study presents the foundation for advancing reliability life assessment in accelerated life testing, achieved through dynamic threshold adjustments using both the absolute log-likelihood distribution of the Gaussian Mixture Model and the reconstruction error distribution of the Long-Short Term Memory autoencoder model.

Preventing Forklift Front-End Failures: Predicting the Weight Centers of Heavy Objects, Remaining Useful Life Prediction under Abnormal Conditions, and Failure Diagnosis Based on Alarm Rules.

This paper addresses the critical challenge of preventing front-end failures in forklifts by addressing the center of gravity, accurate prediction of the remaining useful life (RUL), and efficient fault diagnosis through alarm rules. The study's significance lies in offering a comprehensive approach to enhancing forklift operational reliability. To achieve this goal, acceleration signals from the forklift's front-end were collected and processed. Time-domain statistical features were extracted from one-second windows, subsequently refined through an exponentially weighted moving average to mitigate noise. Data augmentation techniques, including AWGN and LSTM autoencoders, were employed. Based on the augmented data, random forest and lightGBM models were used to develop classification models for the weight centers of heavy objects carried by a forklift. Additionally, contextual diagnosis was performed by applying exponentially weighted moving averages to the classification probabilities of the machine learning models. The results indicated that the random forest achieved an accuracy of 0.9563, while lightGBM achieved an accuracy of 0.9566. The acceleration data were collected through experiments to predict forklift failure and RUL, particularly due to repeated forklift use when the centers of heavy objects carried by the forklift were skewed to the right. Time-domain statistical features of the acceleration signals were extracted and used as variables by applying a 20 s window. Subsequently, logistic regression and random forest models were employed to classify the failure stages of the forklifts. The F1 scores (macro) obtained were 0.9790 and 0.9220 for logistic regression and random forest, respectively. Moreover, random forest probabilities for each stage were combined and averaged to generate a degradation curve and determine the failure threshold. The coefficient of the exponential function was calculated using the least squares method on the degradation curve, and an RUL prediction model was developed to predict the failure point. Furthermore, the SHAP algorithm was utilized to identify significant features for classifying the stages. Fault diagnosis using alarm rules was conducted by establishing a threshold derived from the significant features within the normal stage.

Time domain differential protection scheme applied to power transformers

This paper presents a new differential protection scheme for power transformers based on Intrinsic Time-Scale Decomposition (ITD) and Ensemble of Bagged Tree (EBT) as a classifier. The ITD decomposes the input nonstationary signal into Proper Rotation Components (PRCs) and a monotonic trend signal. In the proposed scheme, statistical time-domain features extracted from the ITD output are fed to the EBT classifier, which classifies different events, including internal fault, internal fault combined with inrush current, different inrush currents (energizing transformer, sympathetic, and recovery inrush current), and external fault. The combination of ITD and EBT tools requires only a limited sample of post-fault data to protect the transformer. The proposed scheme is tested under various operating conditions to confirm its accuracy, and the effect of current transformer (CT) saturation on the scheme's performance is evaluated. The scheme is immune to noisy conditions and does not require transformer parameters. The power systems were simulated using PSCAD/EMTDC and MATLAB software.

Fault Prognosis of Turbofan Engines

In the era of industrial big data, prognostics and health management is essential to improve the prediction of future failures to minimize inventory, maintenance, and human costs. Used for the 2021 PHM Data Challenge, the new Commercial Modular Aero-Propulsion System Simulation dataset from NASA is an open-source benchmark containing simulated turbofan engine units flown under realistic flight conditions. Deep learning approaches implemented previously for this application attempt to predict the remaining useful life of the engine units, but have not utilized labeled failure mode information, impeding practical usage and explainability. To address these limitations, a new prognostics approach is formulated with a customized loss function to simultaneously predict the current health state, the eventual failing component(s), and the remaining useful life. The proposed method incorporates principal component analysis to orthogonalize statistical time-domain features, which are inputs into supervised regressors such as random forests, extreme random forests, XGBoost, and artificial neural networks. The highest performing algorithm, ANN–Flux with PCA augmentation, achieves AUROC and AUPR scores exceeding 0.94 for each classification on average. In addition to predicting eventual failures with high accuracy, ANN–Flux achieves comparable remaining useful life RMSE for the same test split of the dataset when benchmarked against past work, with significantly less computational cost.

Detecting insufficient lubrication in rolling bearings, using a low cost MEMS microphone to measure vibrations

This paper proposes a method for detecting insufficient lubrication in rolling bearings, by utilising a low cost MEMS ultrasonic microphone. We run a set of grease lubricated bearings under load, alternating between good and insufficient lubrication conditions by adjusting speed and temperature, over an 18 day experiment, followed by a short validation experiment on a different set of bearings on a secondary test rig. Vibration signals are recorded from an ultrasonic microphone, using a purpose built prototype sensor, then band-pass and high-pass filtered and subsequently analysed using statistical time-domain features. The viscosity ratio Kappa of the lubricant is used as a metric for quality of oil-film formation, and is calculated throughout the experiment from rotating speed and temperature. The results are compared with the predictions from a linear regression model and a simple Neural Network trained on ensembles of three and ten selected features.The three Hjorth’s Parameters activity, mobility and complexity calculated on high-pass filtered microphone signals from the experiments show a good correlation with the viscosity ratio Kappa. When combining these, using a Linear Regression model or a simple two layer Neural Network model, we are able to predict the viscosity ratio with good precision. The proposed method may be used to assess the quality of oil film formation in bearings during operation under varying speed and temperature conditions.

Statistical Machine Learning Strategy and Data Fusion for Detecting Incipient ITSC Faults in IM

The new technological developments have allowed the evolution of the industrial process to this new concept called Industry 4.0, which integrates power machines, robotics, smart sensors, communication systems, and the Internet of Things to have more reliable automation systems. However, electrical rotating machines like the Induction Motor (IM) are still widely used in several industrial applications because of their robust elements, high efficiency, and versatility in industrial applications. Nevertheless, the occurrence of faults in IMs is inherent to their operating conditions; hence, Inter-turn short-circuit (ITSC) is one of the most common failures that affect IMs, and its appearance is due to electrical stresses leading to the degradation of the stator winding insulation. In this regard, this work proposes a diagnosis methodology capable of performing the assessment and automatic detection of incipient electric faults like ITSC in IMs; the proposed method is supported through the processing of different physical magnitudes such as vibration, stator currents and magnetic stray-flux and their fusion of information. Certainly, the novelty and contribution include the characterization of different physical magnitudes by estimating a set of statistical time domain features, as well as their fusion following a feature-level fusion approach and their reduction through the Linear discriminant Analysis technique. Furthermore, the fusion and reduction of information from different physical magnitudes lead to performing automatic fault detection and identification by a simple Neural-Network (NN) structure since all considered conditions can be represented in a 2D plane. The proposed method is evaluated under a complete set of experimental data, and the obtained results demonstrate that the fusion of information from different sources (physical magnitudes) can lead to achieving a global classification ratio of up to 99.4% during the detection of ITSC in IMs and an improvement higher than 30% in comparison with classical approaches that consider the analysis of a unique physical magnitude. Additionally, the results make this proposal feasible to be incorporated as a part of condition-based maintenance programs in the industry.

Investigation of optimal feature for milling chatter identification using supervised machine learning techniques

In order to overcome the complex mathematical models and tedious analytical skills and improve the machining performance, various effective methods have been developed using time domain, frequency domain and time and frequency domain-based features. However, the selection of these features can be difficult, and results can be alleviated if it is wrongly selected. This study proposes a methodology that helps identify the optimal feature from eight time-domain statistical features using supervised machine learning algorithms. In this work, 44 milling experiments have been performed and labelled as chatter, transient, and stable states by observing the tool-machining condition. Subsequently, the eight time-domain-based features, i.e. mean, variance, peak-to-peak, root mean square, crest factor, form factor, kurtosis and skewness, have been calculated. After that, four machine learning techniques, i.e., random forest, gradient boosting, support vector machine and logistic regression, were utilized, and their accuracy score was 92.86 %, 96.8 %, 94.8 % and 91 %, respectively. After that, their feature score was evaluated to investigate the effectiveness of all 8 time domain-based features. Feature scores for mean, variance, peak-to-peak, root mean square, crest factor, form factor, kurtosis and skewness are 0.28, 0.25, 0.11, 0.095, 0.085, 0.08, 0.065 and 0.035, respectively. The outcome of this research is that peak-to-peak time domain-based features, along with gradient boosting, can extract chatter features in the presence of an extraneous noisy signal.

Real-Time Monitoring and Control of the Breakthrough Stage in Ultrafast Laser Drilling Based on Sequential Three-Way Decision

Real-time monitoring and control of breakthrough stage are essential issues in ultrafast laser drilling (ULD) process. This paper proposed a novel intelligent methodology to address the decision bias problem in monitoring and control of breakthrough stage. Eight time-domain statistical features were extracted after analyzing the correlation between the optical emission signal and the drilling process and data processing. To build the identification model of breakthrough stage, support vector machine (SVM) method was adopted and the identification accuracy can up to 98.42%. Even so, there is a large deviation between the prediction breakthrough time by the identification model and the actual breakthrough time, which can up to 3.81s. Considering the decision bias, the comprehensive Gaussian weight sequential three-way decision (CGW-S3WD) method was proposed for the first time. Compared to the actual breakthrough time, the mean decision time is only delayed by 0.375s based on the combination of the identification model and CGW-S3WD method under simulated conditions. Finally, the method was applied to monitoring and control of breakthrough stage in the actual ULD process. Experimental results demonstrate that the control delay time is only 0.354s and the high-quality holes and protection of the back wall can be achieved simultaneously. The research confirms that the method can reduce the decision bias and provides a new perspective for accurate monitoring of complex industrial scenarios.

Ranked feature-based data-driven bearing fault diagnosis using support vector machine and artificial neural network

This article analyzes a data-driven fault diagnosis method for rolling element bearing under different operating conditions. The work is aimed to incorporate the filter-type feature selection algorithms, namely, neighborhood component analysis, minimum redundancy maximum relevance, chi-square, and ReliefF algorithms for the feature ranking of the extracted time-domain statistical features from the collected vibration data of the five different bearing defect conditions (normal, outer race defect, ball defect, inner race defect, and combined inner and ball defect). The vibration data have been recorded and an envelope analysis technique for pre-processing the signal is used to identify the fundamental defect frequencies of all the bearings. The artificial neural network and the support vector machine are used to predict the comparative performance of the feature selection algorithms. The results demonstrate that the neighborhood component analysis algorithm achieves the highest accuracies for the bearing fault detection with both the support vector machine and artificial neural network among all the feature selection methods. Therefore, the effective feature ranking before the fault classification by the machine learning algorithms can lead to an efficient and reliable bearing fault diagnosis.

An RF Fingerprint Extraction Method based on Time-frequency Domain Feature Fusion

Identification and authentication of communication radiation sources using RF fingerprint feature extraction technology is widely used in civilian and military applications and has become a research hotspot. In this paper, an RF fingerprint extraction method based on time-frequency domain feature fusion is proposed. The frequency domain features of WVD and the time domain statistical features (variance, skewness, kurtosis) of the signal are first extracted, and then fused and downscaled using multivariate discriminant analysis to obtain new time-frequency fusion features, which are finally input to the support vector machine for classification and recognition. This method is used to analyze signals of walkie-talkies and signals of cell phones with and without noise, and the experimental results show that this method effectively improves the recognition rate and has good noise immunity compared with using frequency domain features or time domain features alone, and the accuracy is improved by about 5% under the condition of low SNR. When the SNR is -10dB, the recognition accuracy of walkie-talkies can reach 81.6% and that of cell phones can reach 86.1%.

Machine learning augmented multi-sensor data fusion to detect aero engine fan rotor blade flutter

Abstract Flutter-induced fatigue failure investigation of the fan blades of aero-engines necessitates extensive testing. During engine ground testing, strain gauges on rotor fan blades and casing vibration sensors were employed to investigate structural dynamic aspects. The correlation between strain sensor signals and fan casing vibration signals allowed the diagnosis of fluttering fan blades. For automated flutter detection during engine development testing, a machine learning-augmented information fusion methodology was developed. The method analyses casing vibration signals by extracting time-domain statistical features, intrinsic mode function characteristics through empirical mode decomposition, and recurrence quantification features. Feature vectors obtained from a relatively large set of engine tests were subjected to dimension reduction by applying machine learning techniques to rank them. Reduced feature vector space was labelled as “flutter” or “normal” based on the correlation of rotor strain gauge signals. In addition, the labelled feature vectors were employed to train classifier models using supervised learning-based algorithms such as Support Vector Machines, Linear Discriminant Analysis, K-means Clustering, and Artificial Neural Networks. Using only vibration signals from the casing, the trained and validated classifiers were able to detect flutter in fan baldes with a 99% probability during subsequent testing.

A novel adaptive kernel-guided multi-condition abnormal data detection method

Data acquisition devices and environmental factors may lead to abnormal data and misjudgment of mechanical health status in mechanical work. A new method for abnormal data detection in mechanical equipment health monitoring is proposed in this paper. Firstly, an adaptive kernel bandwidth local outlier factor (AKLOF) detection method is proposed. Furthermore, multiple time-domain statistical features are obtained from sliding window segmentation data, and the anomaly degree of each data segment is calculated. Then, the relationship between the sliding window length and the maximum AKLOF function is studied, and the optimal sliding window length is obtained. Subsequently, the AKLOF value and threshold are calculated according to the optimal sliding window length to detect the final abnormal data. Four groups of experiments result show that the method can accurately identify all the missing data, and can also effectively detect other types of abnormal data, which provide guarantee for equipment condition monitoring.

Effective electroencephalogram based epileptic seizure detection using support vector machine and statistical moment’s features

Epilepsy is one of the widespread disorders. It is a noncommunicable disease that affects the human nerve system. Seizures are abnormal patterns of behavior in the electricity of the brain which produce symptoms like losing consciousness, attention or convulsions in the whole body. This paper demonstrates an effective electroencephalogram (EEG) based seizure detection method using discrete wavelet transformation (DWT) for signal decomposition to extract features. An automatic channel selection method was proposed by the researcher to select the best channel from 23 channels based on maximum variance value. The records were segmented into a nonoverlapping segment with long 1-S. The support vector machine (SVM) model was used to automatically detect segments that contain seizures, using both frequency and time domain statistical moment features. The experimental result was obtained from 24 patients in CHB-MIT database. The average accuracy is 94.1, sensitivity is 93.5, specificity is 94.6 and the false positive rate average is 0.054.

Comparison of LSTM And SVM For Classification of Eye Movements in EOG Signals

People with disabilities suffer from inability to communicate with their surroundings, so Human-Computer Interaction (HCI) technologies are used to have a means of communication for people with disabilities with their surroundings. HCI is an emerging technology in the disciplines of Artificial Intelligence and Biomedical Engineering. To power an external device, HCI technology uses several basic signals such as ECG, EMG, and EEG. Electrooculography (EOG) is a technique for measuring the potential difference between the cornea and the retina located between the front and back of the human eye, and the main application of EOG is to determine the directions of different eye movements. This study aims to assess eye movement for communication by persons with disabilities using electrocardiogram (EOG) data. In this study, the Supporting Vector Machine (SVM) and Long- Short term memory (LSTM) classification techniques was used and two types of features (statistical and time domain features) were used. Classification accuracy was 90.7% and 93.9% when using SVM with statistical domain and time domain features, respectively ,whereas Classification accuracy was 90.1% when using LSTM .

Fault Diagnosis of Wind Turbine Based on Convolution Neural Network Algorithm.

Relying on expert diagnosis, it solves the problem of fan failure efficiency and meets the needs of automatic inspection and intelligent operation monitoring of fans. In order to make up for the deficiency of intelligent diagnosis of bearing fault based on vibration signal detection, signal transformation, and convolution neural network identification and improve the ability of intelligent diagnosis, this study designs a deep convolution neural network model and diagnosis algorithm with three pairs of convolution pooling layers and two full connection layers. The experimental verification of the proposed method is carried out based on the public data set, and the effects of three different signal transformation methods based on vibration signal through vibration gray map, short-time Fourier transform time-frequency map, and continuous wavelet transform time-frequency map on the accuracy of diagnosis model are compared and analyzed. A very accurate guarantee is received, close to 100%. The final experimental results demonstrate the effectiveness of the information on the accuracy of diagnostic testing and provide new ideas for the verification and testing of wind turbine wind energy. Compared with other machine learning algorithms, the real-time recognition of machine learning based on time-domain statistical features is lower than that of convolutional neural network methods. The effect of the scale of the trained model on the accuracy of the algorithm is discussed. A sample ratio of 50% and 70% was found to be appropriate.