Applying predictive maintenance to power quality

Abstract

The purpose of this paper is to explain how predictive maintenance can be used to prevent and therefore avoid power problems in electrical distribution systems. Predictive maintenance techniques have long been employed to prevent loss of operation in mechanical equipment such as engines, and in electromechanical equipment, such as motors. By analyzing the frequency spectrum of vibration in bearings, for example, and comparing how the frequency spectrum changes over time, it is possible to provide early warning of impending failures. The motor can be scheduled for maintenance and the weak bearing replaced. Similar early warning of incipient problems in an electrical distribution system can now also be achieved. By using monitoring instruments that capture power information in great breadth and depth and by using powerful software tools that analyze trends in power quality, harmonics and other power information, deteriorating conditions can be identified and action taken. Predictive maintenance techniques can be applied enterprise-wide to any operation or facility. The greatest benefits of predictive maintenance are realized when multiple monitoring instruments are placed at various critical locations on the electrical infrastructure to continuously monitor power. This paper also discusses how predictive maintenance can also be accomplished with a single portable instrument that is moved around critical locations to monitor for a limited time. A significant technological advance that has made predictive maintenance for electrical distribution systems possible is the invention of full disclosure power monitoring instruments. These monitors comprehensively capture detailed information on power disturbances, harmonics, flicker and power consumption. They use digital signal processing, high-speed sampling and personal computers to capture and store all aspects of power in a database on an internal hard disk. They differ from conventional monitors in two major respects. They do not require thresholds or set-points to isolate power disturbances, and they capture all aspects of power in great breadth and depth. When thresholds are used any underlying information that indicates deteriorating conditions is filtered out and not captured. Full disclosure monitors however capture all power information including information on conditions that would normally be below the thresholds. By tracking these conditions over time degeneration in the power system that would lead to a marked loss of efficiency or lead to failure is made apparent. The availability of high-speed and affordable personal computers has led to the development of software tools that can rapidly analyze the electrical characteristics and compare multiple databases, either from a single instrument or from many instruments. This paper also discusses the use of a power quality index that takes into account all of the various elements of power disturbances such as impulses, transients, sags and swells. By trending this power quality index over time it becomes possible to provide advance warning of deteriorating electrical components that would lead to system failure. This paper also presents scenarios showing how these tools can be used to allow electrical engineers to become proactive to achieve the goal of system reliability.