Condition monitoring

About: Condition monitoring is a research topic. Over the lifetime, 13911 publications have been published within this topic receiving 201649 citations.

Effect of SVM kernel functions on classification of vibration signals of a single point cutting tool

Abstract: The studies on tool condition monitoring along with digital signal processing can be used to prevent damages on cutting tools and workpieces when the tool conditions become faulty. These studies have become more relevant in today's context where the order realization dates are crunched and deadlines are to be met in order to catch up with the competition. Based on a continuous acquisition of signals with sensor systems it is possible to classify certain wear parameters by the extraction of features. Data mining approach is extensively used to probe into structural health of the tool and the process. This paper discusses condition monitoring of carbide tipped tool using Support Vector Machine and compares the classification efficiency between C-SVC and @n-SVC. It further analyses the results with other classifiers like Decision Tree and Naive Bayes and Bayes Net. The vibration signals are acquired for various tool conditions like tool-good condition, tip-breakage, etc. The effort is to bring out the better features-classifier combination.

Condition Monitoring of DC-Link Capacitors in Aerospace Drives

Abstract: This paper proposes a novel online condition monitoring system of dc-link metallized polypropylene capacitors in fault-tolerant drives for aerospace applications. The estimation technique makes use of voltage and current sensors which are already in place for protection and control purposes. The novel aspect of the proposed technique relates to monitoring capacitors in real time while the motor is operational. No external interferences, such as injected signals or special operation of the drive, are required. The condition monitoring system is independent of torque and speed and hence independent of a variation in the load. The condition monitoring system is validated using manual calculations, simulation, low-voltage experimentation, and high-voltage implementation on an aerospace drive.

Robust condition monitoring of rolling element bearings using de-noising and envelope analysis with signal decomposition techniques

Abstract: This paper proposes a robust condition monitoring methodology for rolling element bearings.It employs an empirical mode decomposition (EMD)-based de-noising technique.It also utilizes discrete wavelet packet transform (DWPT)-based envelope analysis.This paper explores the impact of sub-band signals decomposed by the DWPT technique.The proposed scheme outperforms other conventional methods in MPR values. This study presents a robust condition monitoring methodology for rolling element bearings that employs a novel empirical mode decomposition (EMD)-based method to eliminate high-level noise from an acoustic emission (AE) signal and a discrete wavelet packet transform (DWPT)-based envelope analysis technique to effectively search for symptoms of defective bearings. First, the proposed EMD-based de-noising scheme enhances the signal-to-noise ratio by using a Naive Bayes classifier that partitions intrinsic mode functions (IMFs) into noise-dominant and noise-free categories, employing a soft-thresholding-based noise reduction technique for the noise-dominant IMFs, finally obtaining a de-noised acoustic emission (AE) signal via the reconstruction process using both de-noised IMFs and noise-free IMFs. The de-noised AE signal is then decomposed into a set of uniformly spaced sub-bands using three-level DWPT, and the most informative sub-band is determined for early detection of bearing failures. The performance of the proposed condition monitoring scheme is compared with the performance of conventional methods in terms of mean-peak ratio (MPR), which is a metric used to evaluate the degree of defectiveness of the bearings. The experimental results show that the proposed method outperforms the conventional schemes by achieving up to 23.48% higher MPR values, even in a very noisy environment.

Multi-Target Defect Identification for Railway Track Line Based on Image Processing and Improved YOLOv3 Model

Abstract: The condition monitoring of railway track line is one of the essential tasks to ensure the safety of the railway transportation system. Railway track line is mainly composed of tracks, fasteners, sleepers, and so on. Given the requirements for rapid and accurate inspection, innovative and intelligent methods for multi-target defect identification of the railway track line using image processing and deep learning methods are proposed in this paper. Firstly, the track and fastener positioning method based on variance projection and wavelet transform is introduced. After that, a bag-of-visual-word (BOVW) model combined with spatial pyramid decomposition is proposed for railway track line multi-target defect detection with a detection accuracy of 96.26%. Secondly, an improved YOLOv3 model named TLMDDNet (Track Line Multi-target Defect Detection Network), integrating scale reduction and feature concatenation, is proposed to enhance detection accuracy and efficiency. Finally, to reduce model complexity and further improve the detection speed, with the help of dense connection structure, a lightweight design strategy for the TLMDDNet model named DC-TLMDDNet (Dense Connection Based TLMDDNet) is proposed, in which the DenseNet is applied to optimize feature extraction layers in the backbone network of TLMDDNet. The effectiveness of the proposed methods is demonstrated by the experimental results.

Modeling and experimental verification of on-line PD detection in MV covered-conductor overhead networks

Abstract: The falling trees on covered-conductor (CC) overhead distribution lines produce partial discharges (PDs). The measurements have been taken in the laboratory and PD signal characteristics under various circumstances have been described. In this paper, an on-line single-phase PD monitoring system using Rogowski coil is simulated in EMTP-ATP. The simulation results are compared with those obtained from the laboratory measurements. The proposed model can be used to estimate the length of the practical CC lines at which PDs due to falling trees can be detected and localized; thus, deciding the number and positioning of the sensors over a particular length of the CC lines. The different noise sources have been described that cause interference with low level PD signals, which is a major challenge for on-line/on-site condition monitoring. The design aspects of the wireless sensor for this specific application are also discussed. Automatic detection of falling trees will reduce visual inspection work after storms and it will improve reliability and safety of the CC distribution system.