1.1 Background
The need for an effective condition monitoring (CM) and machinery maintenance program exists wherever complex, expensive machinery is used to deliver critical business functions. For example, manufacturing companies in today's global marketplace use their best endeavours to cut costs and improve product quality to maintain their competitiveness. Rotating machinery is a central part of the manufacturing procedure, and its health and availability have direct effects on production schedules, production quality, and production costs. Unforeseen machine failures may lead to unexpected machine downtime, accidents, and injuries. Recently it has been stated that machine downtime costs UK manufacturers £180bn per year (Ford 2017; Hauschild 2017). Moreover, Mobley stated that based on the specific industry, maintenance costs can represent between 15% and 60% of the cost of goods produced. For instance, in food‐related production, average maintenance costs represent approximately 15% of the cost of goods produced, whereas for iron and steel and other heavy industries, maintenance costs represent up to 60% of total production costs (Mobley 2002).
Components including motors, bearings, gearboxes, etc. are engaged to operate effectively to keep the rotating machine in a stable, healthy condition. For that reason, maintenance is performed by repairing, modifying, or replacing these components in order to ensure that machines remain in a healthy condition. Maintenance can be accomplished using two main approaches: corrective and preventive maintenance (Wang et al. 2007). Corrective maintenance is the most basic maintenance technique and is performed after machine failure, which is often very expensive particularly for large‐scale applications of rotating machines. Preventive maintenance can be applied to prevent a failure using either time‐based maintenance (TBM) or condition‐based maintenance (CBM), which can be localised CBM or remote CBM (Higgs et al. 2004; Ahmad and Kamaruddin 2012). TBM uses a calendar schedule that is set in advance to perform maintenance regardless of the health of the machine, which makes this approach expensive in some large and complex machines. In addition, TBM may not prevent machines from failing.
With regard to CBM, it has been reported that 99% of rotating equipment failures are preceded by nonspecific conditions indicating that such a failure is going to happen (Bloch and Geitner 2012). Hence, CBM is regarded as an efficient maintenance approach that can help avoid the unnecessary maintenance tasks of the TBM approach. Numerous studies have shown the economic advantages of CBM in several applications of rotating machines (e.g. McMillan and Ault 2007; Verma et al. 2013; Van Dam and Bond 2015; Kim et al. 2016). In CBM, decisions about maintenance are made based on the machine's current health, which can be identified through the CM system. Once a fault occurs, an accurate CM technique allows early detection of faults and correct identification of the type of faults. Thus, the more accurate and sensitive the CM system, the more correct the maintenance decision that is made, and the more time available to plan and perform maintenance before machine breakdowns.
Condition monitoring of rotating machine components can minimise the risk of failure by identifying machine health via early fault detection. The main aim of condition monitoring is to avoid catastrophic machine failures that may cause secondary damage, machine downtime, potentially safety incidents, lost production, and higher costs associated with repairs. The CM techniques in rotating machinery encompass the practice of monitoring measurable data (e.g. vibration, acoustic, etc.), which can be used individually or in combination to identify changes in machine condition. This allows the CBM program to be arranged, or other actions to be taken to prevent machine breakdowns (Jardine et al. 2006). Based on the types of sensor data acquired from rotating machines, CM techniques can be grouped into the following: vibration monitoring, acoustic emission (AE) monitoring, electric current monitoring, temperature monitoring, chemical monitoring, and laser monitoring. Of these techniques, vibration‐based condition monitoring has been widely studied and has become a well‐accepted technique for planned maintenance management (Lacey 2008; Randall 2011). In the real world, different fault conditions generate different patterns of vibration spectrums. Thus, vibration analysis in principle allows us to examine the inner parts of the machine and analyse the health of the operating machine without physically opening it (Nandi et al. 2013). Moreover, various characteristic features can be observed from vibration signals that make this one of the best selections for machine CM.
This chapter describes maintenance approaches for rotating machines failures and applications of machine condition monitoring (MCM). It also provides a description of various CM techniques used for rotating machines.
