Real-time Abnormality Research Articles

Anomaly detection and diagnosis for wind turbines using long short-term memory-based stacked denoising autoencoders and XGBoost

An anomaly detection and diagnosis method for wind turbines using long short-term memory-based stacked denoising autoencoders (LSTM-SDAE) and extreme gradient boosting (XGBoost) is proposed in this paper. First, an abnormal data recognition algorithm based on the local outlier factor and adaptive K-means was developed to implement data preprocessing and noise extraction. The LSTM-SDAE model was then established to obtain the nonlinear temporal relationship among multivariate variables in normal behavior modes. The Mahalanobis distance was calculated based on reconstruction errors and the threshold for anomaly detection was set with a 99.7% confidence interval for the distribution curve fitted by kernel density estimation. An alarm mechanism based on the sliding window technique was set up to detect abnormalities in real time. Finally, contribution analysis was conducted to extract the parameter features under different abnormal modes, and the XGBoost was trained by extended data from wind turbines of the same type in the same wind farm to realize anomaly location and diagnosis. To verify the proposed method, real SCADA data from a wind farm located in northeastern China were applied. The results show the capability of the proposed method in anomaly detection and diagnosis for wind turbines.

Development of an On-Line Defect Detection System for EDM Process

In the electrical discharge machining process, preliminary research has been able to effectively estimate machining accuracy in response to its long machining history and high discharge frequency characteristics. However, when processing abnormalities occur, it is difficult to identify them since the electrical discharge process contains multiple processing parameters, which increases the cost of repair or loss afterwards. Therefore, the question concerning how to monitor the abnormality of the discharge process in real time represents the main purpose of this research. This research develops an EDM process abnormal diagnosis system. First, the data are stored in a circular array to speed up the processing time, and the coefficient of variation feature is added, which has effectively extracted the abnormal characteristics. In terms of diagnostic methods, the composite voting model established by neural networks, random forests, and XGB-RF (extreme gradient boosting applying RF) can provide robust diagnostic results. Finally, through the Node-RED webpage and MQTT agreement, it can provide the ability to monitor machine abnormalities in real time. Through refinement and optimization of the previous research results, this study took the electrical discharge machining diamond grinding wheel as an example, and developed a warning that can be issued within 3 min when abnormalities (abnormal patterns such as polycrystalline diamond high protrusions) occur, with an accuracy of 93% and a false positive rate. The abnormal diagnosis ability is less than 0.2%. Therefore, the online abnormality monitoring system developed by this research institute will be able to provide online abnormality diagnosis for electrical discharge machining.

Artificial Intelligence in Medicine: Real Time Electronic Stethoscope for Heart Diseases Detection

Diseases of the cardiovascular system are one of the major causes of death worldwide. These diseases could be quickly detected by changes in the sound created by the action of the heart. This dynamic auscultations need extensive professional knowledge and emphasis on listening skills. There is also an unmet requirement for a compact cardiac condition early warning device. In this paper, we propose a prototype of a digital stethoscopic system for the diagnosis of cardiac abnormalities in real time using machine learning methods. This system consists of three subsystems that interact with each other (1) a portable digital subsystem of an electronic stethoscope, (2) a decision-making subsystem, and (3) a subsystem for displaying and visualizing the results in an understandable form. The electronic stethoscope captures the patient’s phonocardiographic sounds, filters and digitizes them, and then sends the resulting phonocardiographic sounds to the decision-making system. The decision-making system classifies sounds into normal and abnormal using machine learning techniques, and as a result identifies abnormal heart sounds. The display and visualization subsystem demonstrates the results obtained in an understandable way not only for medical staff, but also for patients and recommends further actions to patients. As a result of the study, we obtained an electronic stethoscope that can diagnose cardiac abnormalities with an accuracy of more than 90%. More accurately, the proposed stethoscope can identify normal heart sounds with 93.5% accuracy, abnormal heart sounds with 93.25% accuracy. Moreover, speed is the key benefit of the proposed stethoscope as 15 s is adequate for examination.

Use of Plasma Information in Machine-Learning-Based Fault Detection and Classification for Advanced Equipment Control

For advanced equipment control, two schemata of real-time fault detection were performed using machine learning algorithms in silicon etching in SF <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</sub> /O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /Ar plasma. Fault detection and classification is investigated with the plasma state information with optical emission spectroscopy (OES) data to find the root cause of the anomaly in the process parameters. Fault detection and control is also demonstrated to predict the shift of the process parameter along the amount of process gas flow rate injected into the chamber, considering a fault. Especially, plasma information (PI), such as electron temperature and electron density, was derived from OES data into equation-based corona model. These were utilized to evaluate which process parameter is the most significantly affecting on the performance of the established model through Shapley value in fault detection and control. By the combination of isolation forest algorithm for finding the plasma abnormalities in real time and Adaboost algorithm for classifying root causes of faults, the suggested algorithm could accurately detect the root cause. DeepSHAP algorithm helped not only the prediction of gas flow rate, but PI was identified as critical parameter, interpreting the model through Shapley value. We propose a new multi-function integrated algorithm by the ensemble algorithms.

&lt;b&gt;Fast Calculation Method of Abnormality Degree for Real Time Abnormality Detection in Vehicle Equipment&lt;/b&gt;

Vibration monitoring is effective for early detection of equipment failure. In the vibration monitoring system proposed in this paper, abnormality detection is performed by applying the nearest neighbor method (NN) to the octave band analysis results of vibration. However, the NN requires a long calculation time and is not suitable for detecting abnormalities in real time. Therefore, applying the One Class Support Vector Machine (OCSVM) to abnormality detection was considered. In this paper, the OCSVM was applied to actual vibration data, and the calculation time was compared with those of the NN. The result shows that the calculation time is significantly reduced compared to the NN approach.

Efficient mixture control chart pattern recognition using adaptive RBF neural network

In this paper, a stochastic gradient method based adaptive version of the radial basis function neural network has proposed to map the pattern features of the control chart patterns in different categories to recognize their belonging class. Adaptiveness has given over the spreadness and centers of Gaussian basis function appeared in the hidden nodes of the radial basis function neural network. Along with normal abnormalities in patterns, the mixture of different abnormal patterns has also considered capturing the worst possible conditions of abnormalities in real time. The advantages of the proposed method have appeared as very high recognition accuracy, minimum error in learning and generalize performance with small training dataset in control chart pattern recognition. Achieved performance has compared with the state of art results available in the literature which has applied feature based recognition using Support vector machine and Genetic algorithm. The proposed method has enhanced the recognition generalization of control chart patterns with simplicity in design and high level of decision confidence. The performances have achieved through the simulation-based experiments over a huge number of patterns containing ten different types of pattern and on average, 99.99% accuracy has achieved.

Radionuclide-Labeled Peptides for Imaging and Treatment of CXCR4- Overexpressing Malignant Tumors.

Malignant tumors are a major cause of death. The lack of methods that provide an early diagnosis and adequate treatment of cancers is the main obstacle to precision medicine. The C-X-C chemokine receptor 4 (CXCR4) is overexpressed in various tumors and plays a key role in tumor pathogenesis. Therefore, CXCR4-targeted molecular imaging can quickly and accurately detect and quantify CXCR4 abnormalities in real time. The expression level and activation status of CXCR4 are very important for screening susceptible populations and providing an accurate diagnosis and optimal treatment. In view of the fact that radionuclide-labeled peptides have become widely used for the diagnosis and treatment of tumors, this manuscript reviews the potential of different radionuclide-labeled peptide inhibitors for the targeted imaging of CXCR4- positive tumors and targeted treatment. The article also discusses the specificity and in vivo distribution of radionuclide-labeled peptide inhibitors, and translation of these inhibitors to the clinic.

Can Molecular Imaging Enable Personalized Diagnostics? An Example Using Magnetomotive Photoacoustic Imaging

The advantages of photoacoustic (PA) imaging, including low cost, non-ionizing operation, and sub-mm spatial resolution at centimeters depth, make it a promising modality to probe nanoparticle-targeted abnormalities in real time at cellular and molecular levels. However, detecting rare cell types in a heterogeneous background with strong optical scattering and absorption remains a big challenge. For example, differentiating circulating tumor cells in vivo (typically fewer than 10cells/mL for an active tumor) among billions of erythrocytes in the blood is nearly impossible. In this paper, a newly developed technique, magnetomotive photoacoustic (mmPA) imaging, which can greatly increase the sensitivity and specificity of sensing targeted cells or molecular interactions, is reviewed. Its primary advantage is suppression of background signals through magnetic enrichment/manipulation with simultaneous PA detection of magnetic contrast agent targeted objects. Results from phantom and in vitro studies demonstrate the capability of mmPA imaging to differentiate regions targeted with magnetic nanoparticles from the background, and to trap and sensitively detect targeted cells at a concentration of a single cell per milliliter in a flow system mimicking a human peripheral artery. This technique provides an example of the ways in which molecular imaging can potentially enable robust molecular diagnosis and treatment, and accelerate the translation of molecular medicine into the clinic.

적외선 열화상 장비를 이용한 실시간 변압기 진단 및 평가

It was diagnosed customer transformers and other equipment by using non-contact infrared equipment. Examiners can identify not only abnormalities in real time in the field of the transformer immediately, as well as they were able to safely perform the scan. Thanks to successful transformer diagnosis, we can easily diagnose abnormalities of transformer itself which can be caused by deterioration of the oil used as overlaod or refrigerant and we can also diagnose abnomalities from low voltage bushing which can be result by external environmental factors and physical factors and abnomalities from its connections at the same time we found it is very useful at proactive diagnostics.

Molecular imaging using fluorescent lectins permits rapid endoscopic identification of dysplasia in Barrett's esophagus

Barrett's esophagus is an example of a pre-invasive state, for which current endoscopic surveillance methods to detect dysplasia are time consuming and inadequate. The prognosis of cancer arising in Barrett's esophagus is improved by early detection at the stage of mucosal carcinoma or high-grade dysplasia. Molecular imaging methods could revolutionize the detection of dysplasia, provided they permit a wide field of view and highlight abnormalities in real time. We show here that cell-surface glycans are altered in the progression from Barrett's esophagus to adenocarcinoma and lead to specific changes in lectin binding patterns. We chose wheat germ agglutinin as a candidate lectin with clinical potential. The binding of wheat germ agglutinin to human tissue was determined to be specific, and we validated this specific binding by successful endoscopic visualization of high-grade dysplastic lesions, which were not detectable by conventional endoscopy, with a high signal-to-background ratio of over 5.

New method of on-line quantification of regional wall motion with automated segmental motion analysis

We have recently developed an automated segmental motion analysis (A-SMA) system, based on an automatic “blood-tissue interface” detection technique, to provide real-time and on-line objective echocardiographic segmental wall motion analysis. To assess the feasibility of A-SMA in detecting regional left ventricular (LV) wall motion abnormalities, we performed 2-dimensional echocardiography with A-SMA in 13 healthy subjects, 22 patients with prior myocardial infarction (MI), and 9 with dilated cardiomyopathy (DCM). Midpapillary parasternal short-axis and apical 2- and 4-chamber views were obtained to clearly trace the blood-tissue interface. The LV cavity was then divided into 6 wedge-shaped segments by A-SMA. The area of each segment was calculated automatically throughout a cardiac cycle, and the area changes of each segment were displayed as bar graphs or time-area curves. The systolic fractional area change (FAC), peak ejection rate (PER), and filling rate (PFR) were also calculated with the use of A-SMA. In the control group, a uniform FAC was observed in real time among 6 segments in the short-axis view (60% ± 10% to 78% ± 9%), or among 5 segments in either the 2-chamber (59% ± 12% to 75% ± 16%) or 4-chamber view (58% ± 13% to 72% ± 12%). The variations of FAC, PER, and PFR were obviously decreased in infarct-related regions in the MI group and were globally decreased in the DCM group. We conclude that A-SMA is an objective and time-saving method for assessing regional wall motion abnormalities in real time. This method is a reliable new tool that provides on-line quantification of regional wall motion. (J Am Soc Echocardiogr 2001;14:892-901.)

A systematic approach to integrated fault diagnosis of flexible manufacturing systems

A flexible manufacturing system (FMS) is an application of modern manufacturing techniques. Like for other manufacturing equipment, the success of an FMS is very much dependent upon its trouble-free operation. It is crucial to monitor all the possible faults or abnormalities in real time and, when a fault is detected, react quickly in order to maintain the productivity of the FMS. Because of the complexity of FMSs, the functionally complete diagnosis of an FMS should be based on all the available information and various advanced diagnostic techniques so as to get a satisfactory result. This paper proposes a systematic approach to fault diagnosis of FMSs that integrates condition monitoring, fault diagnosis and maintenance planning. An intelligent integrated fault-diagnosis system is designed with a modular and reconfigurable structure. The implementation of the integrated diagnosis system is presented in detail. The system can monitor the major conditions and diagnose the major faults of an FMS, and give corresponding maintenance planning as well. The developed system has been applied to an existing FFS-1500-2 FMS in Zhengzhou Textile Machinery Plant and has achieved good results.