China
Africa
Explore chapters and articles related to this topic
A Brief History of Rotor Systems and Recent Trends
Published in Rajiv Tiwari, Rotor Systems: Analysis and Identification, 2017
Another area in which a lot of development took place is on the assessment of turbomachinery condition monitoring and failure prognosis technology (Collacott, 1977; Lipovszky et al., 1990; Wowk, 1991; Mitchell, 1993; Davies, 1998; Adams, 2001; Rao, 2007; Lees, 2016). High-performance turbomachines are now extremely important elements of the industry worldwide. The electric power, petrochemical, mining, marine, and aircraft industries are prime examples for which turbomachinery is crucial to business success (Figure 1.10). Condition monitoring involves the continuous or periodic assessment of the condition of a plant or a machinery component during its operation. Basically, condition monitoring is the process of monitoring some parameters from the machinery, such that a significant change in a parameter can give information about the health of the machinery. The acoustics and vibration signals from machines can contain vital information about the internal process and can provide valuable information about a machine's running condition. Acoustic signals are measured in a region of proximity to the external surface of the machine, whereas vibration signals are measured on the surface of the machine. Most acoustic and vibration analysis instruments utilize a fast Fourier transform (FFT), which is a special case of the generalized discrete Fourier transform (DFT). Spin-off of it is the full spectrum that displays both the forward and backward whirl amplitude with frequency (refer to Chapter 16).
A Case Study on the Smart Streetlighting Solution Based on 6LoWPAN
Published in Mohammad Ayoub Khan, Internet of Things, 2022
Manoj Kumar, Prashant Pandey, Salil Jain
Another important feature of a smart streetlight system is the condition monitoring and fault reporting capability. Condition monitoring is a feature which can warn about an impending failure and thus facilitating preventive maintenance of the system. For instance, if the system observes that an LED module has started to draw more than usual current while operating at some illumination level, or the temperature of an enclosure is too high compared to other lights with same configuration, it is possibly an indication of impending failure. If a failure does occur, the smart light should inform the maintenance team through the CMS, so that the failure can be fixed even before users of the system make a complaint.
Reliability Analysis Using Condition Monitoring Approach in Thermal Power Plants
Published in Harish Garg, Mangey Ram, Reliability Management and Engineering, 2020
Hanumant Jagtap, Anand Bewoor, Ravinder Kumar, Mohammad H. Ahmadi, Dipen Kumar Rajak
In the case of rotating machinery, the performance parameters of machines, such as vibration, acoustic emission, wear debris in oil, thermography, and temperature, are useful indicators of the condition of the machinery (Han & Song, 2003). Successful implementation of condition monitoring programs allows the machine to operate without failures (B. R. Kumar, Ramana, & Rao, 2009). In recent years, various effective condition monitoring techniques have been developed, which includes vibration analysis, acoustic emission monitoring, wear debris analysis, thermography, temperature analysis, ultrasonic monitoring, testing, visual inspection, motor condition monitoring, and motor current signature analysis (Bagavathiappan et al., 2013). The ISO standard provides general guidelines for the selection of condition monitoring programs, appropriate measurement methods, and monitoring parameters. Generally, these describe the acceptable limits for evaluating the performance parameters of various systems, machines, or components such as rolling element bearings, shafts, pumps, fans, steam turbines, gears, centrifugal compressors, induction motors, screw compressors, large generators, and steam turbine generator sets. The selection of a maintenance strategy has a significant influence on the operational cost and the operational availability of the system. The maintenance task has been classified into three types, viz. corrective maintenance, preventive maintenance, and predictive (condition-based) maintenance. The condition-based maintenance strategy recommends that the maintenance decision should be based on the information collected through implemented condition monitoring techniques. The use of such a condition monitoring based maintenance (CMBM) strategy has not only been widely recommended by researchers but also been adopted by industries at large.
Enhancing resilience in marine propulsion systems by adopting machine learning technology for predicting failures and prioritising maintenance activities
Published in Journal of Marine Engineering & Technology, 2023
Mohsen Elmdoost-gashti, Mahmood Shafiee, Ali Bozorgi-Amiri
In general, maintenance strategies can be classified into three types: corrective maintenance, preventive maintenance, and CBM (Coraddu et al. 2016). Under the corrective maintenance strategy, the equipment or system is allowed to run until it fails and then it is repaired or replaced. Preventive maintenance (PM) is a strategy that involves routine repairs according to a defined time interval or usage level of the asset. CBM is a maintenance strategy that has attracted the attention of many researchers in recent years (Asuquo et al. 2021; Kimera and Nangolo 2022). This strategy stipulates that maintenance should only be performed when certain indicators show signs of performance degradation or impending failure. For this purpose, it incorporates all diagnosis, process and performance data, maintenance histories, operator logs and design information to make timely maintenance decisions. CBM provides the ability to increase the equipment reliability and improve the efficacy of maintenance operations based on the data gathered from condition monitoring systems. Condition monitoring systems include various tools that are used to record and evaluate different parameters such as vibration, acoustic, temperature, flow signal and oil colour. The data is then processed to determine the health status of the equipment and predict the remaining useful life (RUL) (Pascual 2015). An optimal maintenance plan is then prepared based on the predicted health condition of the equipment so that a preventive repair/replacement can be performed when there is a high risk of failure (Liao et al. 2006; Vachtsevanos et al. 2006).
Wavelet-based features for prognosis of degradation in rolling element bearing with non-linear autoregressive neural network
Published in Australian Journal of Mechanical Engineering, 2021
Improvements in production processes are always focused in industries to enhance production capability and reduction in losses. On another way, elimination of unscheduled downtime and unexpected breakdown can be possible with proper planning of maintenance programme. A different approach has been suggested in the literature to advance the computer-based maintenance system. This advancement, development of the model, is not substantial to judge the component behaviour before failure. Therefore, it needs to develop such a methodology to sense behaviour of a component in advance before it is going to fail. For this development, the collected machine condition information will help to arrange the sequence of activity of component. The machine condition identified using the sensor information is known as condition monitoring. The machine condition information can be helpful to avoid the failure of the component. The remaining life of the component is being estimated using this information to safeguard the work of component for maximum life.
A generic framework for multisensor degradation modeling based on supervised classification and failure surface
Published in IISE Transactions, 2019
Changyue Song, Kaibo Liu, Xi Zhang
With the rapid development of sensor technology, condition monitoring has been widely adopted in attempts to limit or even prevent unexpected failures and reduce the maintenance cost of critical units such as machines, automotive batteries, and aircraft engines. In condition monitoring, the degradation modeling and analysis of the signals collected by sensors play a critical role in estimating the degradation status and predicting the Remaining Useful Life (RUL) of units (Nelson, 1990; Meeker and Escobar, 1998). Currently, most of the literature on degradation modeling focuses on analyzing a single sensor signal (Si et al. 2011; Ye and Xie, 2015). However, as discussed in Brotherton et al. (2002) and Jardine et al. (2006), a single sensor signal is often insufficient to fully characterize the degradation status of the unit, as one sensor only collects measurements with respect to one characteristic of the degradation process. In order to gather information from different characteristics and predict the RUL more accurately, it has become common practice to deploy multiple sensors to monitor one unit simultaneously. This creates a pressing need for multisensor degradation modeling.