Today, many companies choose to operate their power system equipment until it malfunctions. They believe the system being considered for maintenance can not be removed from service, even for a short period of time. Also, many think that since no prior breakdowns have occurred, the potential for future operational interruption is somehow mitigated. Such an approach assumes that when a failure occurs, the resulting outage and repair costs will be less than the investment required for a preventive maintenance program. Ultimately, breakdown maintenance will be cost effective only if no catastrophic failures occur. Of course, such an approach leaves the system open to major catastrophes because no precautions are taken to prevent them. Preventive maintenance (PM) is currently the most widely accepted approach to maintaining electrical equipment. PM is a calendar based program in which very comprehensive test routines are applied to off-line equipment. There are two very significant differences between the way the data is used in corrective maintenance (CM) and PM programs. In a preventive maintenance program:

• Data is collected during both on-line and off-line times. Off-line times are intentionally scheduled for the implementation of preventive maintenance procedures.
• When it is discovered that equipment needs repair, that equipment will be scheduled for outages to implement those repairs.

However, the principal problem with preventive maintenance programs is financial in nature. PM returns cannot be measured unless a catastrophic outage occurs. When such an outage occurs, the value of a maintenance program can be compared directly to the cost of the outage.

Maintenance has evolved over the years from simply reacting to machinery breakdowns (CM), to performing time-based PM, to today's emphasis on the ability to detect early forms of degradation in predictive maintenance (PdM) practices. Condition based maintenance (CBM) has been defined as maintenance actions based on actual condition obtained from in-situ, non-invasive tests, operating and condition measurement. The CBM approach is characterized by understanding the stressor levels intended during the machinery design process, measuring suitable parameters to quantify the existing stressor levels, and correcting operating environments to make these levels compatible with economic production versus equipment lifetimes.

U.S. Navy Applications for CBM
One of the Navy's ongoing needs is to reduce life-cycle costs as they relate to manning and equipment breakdown and repair in order to enhance mission readiness. Working with the Navy, the EMPF has been defining a CBM system based on fiber optic sensor technology for next generation shipboard electrical power plants. The overall objective of this project is to minimize these costs from an overall life cycle cost perspective by implementing CBM to monitor conditions of a power conversion module. A more specific goal is to develop methods for the diagnosis of electronic components for electric distribution equipment.

The EMPF is currently working with PEO carriers, PEO ships, ONR, Airak, and NSWC Philadelphia on the evaluation of fiber optic sensor networks to monitor power levels and to provide input to be utilized by a CBM system. This effort will involve the production and testing of a prototype system to demonstrate the feasibility of fiber optic technology for use in the monitoring of a ship-board power conversion unit. The utilization of fiber optics presents a number of significant possibilities in the development of a smart power level monitoring and feedback system.

CBM consists of PdM and real-time monitoring. PdM takes advantage of proven cause and effect relationships to predict the need for corrective action. PdM is comprised of methods which attempt to predict or diagnose problems in a piece of equipment based on trending of test results. PdM primarily uses non-intrusive testing techniques to measure and compute equipment performance trends. Real-time monitoring uses current performance data to assess machinery condition. CBM replaces arbitrarily timed maintenance with maintenance that is scheduled only when warranted by the equipment condition. Continued analysis of equipment condition data allows for the planning and scheduling of maintenance activities or repairs prior to functional failure. PdM utilizes CBM systems to detect fault sources well in advance of failure, making maintenance proactive rather than reactive. The concept is based on the belief that if equipment can be evaluated and yet still remain in service, the overall cost of maintenance will go down.

CBM adds two enormously important dimensions to classical PdM. First, CBM deals with the entire system as an entity. This holistic approach to maintenance represents a major shift from the piecemeal technologies of the past. While CBM can be implemented in single steps, its greatest potential is realized when it is applied consistently and evenly across the entire range of system maintenance concepts. The second added dimension is the concept of ignoring or extending maintenance intervals.

CBM systems, on the other hand, are a combination of visual and manual inspections, on-line monitoring of the mechanical condition (i.e. through built-in sensors), system efficiency data and many other indicators. When all this information is entered into the CBM system it is possible to accurately determine the actual overall system status, condition and the maintenance need.

Combining CBM with a traditional PM schedule (i.e. a time-driven schedule which is the most economical approach for certain wear components) results in a very cost-efficient total maintenance system that predicts the maintenance need. It minimizes unplanned down-time while maximizing safety and operational availability. Such a system also has a positive effect on logistical support by enabling "just-in-time" systems and keeping parts in stock at a practical and cost-efficient level. With CBM, a very important factor is the prospect of saving as well as earning money. Three important areas can be identified:

• Minimizing inspection overhauls
• Replacing parts at the optimal time
• Being able to tune the maintenance with other business factors

CBM Examples
Some applicable systems that would be covered by CBM include turbine engines, gear trains, hydraulic actuators and pumps, electric motors, valves, air conditioning systems, diesel engines, generators and electronics. Some application examples for CBM include automated throttle controls for ship systems and electro-hydrostatic actuators for aircraft and spacecraft. Airak, a manufacturer of fiber optic sensor products and an EMPF partner on a power conversion R&D project, has developed a fiber optic current sensor (Figure 4-1) to monitor electric current on Navy ships. This is important for the shipboard powerplant because by combining CBM and optical monitoring, the Navy will be able to reduce weight and manpower needed to service bulkier components (such as current transformers), while simultaneously improving safety and EMI immunity over the existing system.

The operational demands and high procurement costs of systems in these applications require a high degree of "uptime" and high reliability. Unexpected failures are costly to correct at best. At worst, such failures are catastrophic. In spite of this general need, highly effective, state-of-the-art diagnostic and prognostic systems have not been implemented. Safety-critical and mission-critical systems often employ special purpose, ad hoc, and redundant systems to provide marginal to limited protection. Automation systems are being applied to more complex and safety-critical systems and there is a commensurate need for increased safety and reliable operation. The driving force for some of the programs being sponsored by the Office of Naval Research is to facilitate a growth environment for CBM technology and applications that will return multifold benefits such as:

• Reduced maintenance costs through minimizing requirements for time directed maintenance as well as improved maintenance planning and logistics support
• Reduced logistics footprint through decreased sparing and transportation requirements
• Improved operational flexibility achieved by accurate prognostics that will enable operators to make tactical, mission specific decisions with full knowledge of the remaining useful life of vital equipment

Constant efforts to develop engine components with longer maintenance intervals are now paying off. An example is the introduction of engine-mounted computers, which has made cabling much more efficient with the possibility to include safety checks. Other vital parts of an automation system are sensors. In addition to their increasing reliability, new types of sensors and sensor communicating systems are being developed. Development of communication technology now covers much more than just engines. Now that mobile phones, other handheld devices and satellite communication are becoming more price competitive, a ship can today be "connected" in the sense of having access to information through the internet and intranets. Some vendors have worked out a complete CBM concept based on these developments. In addition to utilizing and combining available systems like engine-mounted sensors, remote-monitoring and diagnostics systems, the concept also involves setting up a CBM organization.

Conclusion
The introduction of the CBM concept will affect new maintenance layouts. In general, inspections will be performed less frequently even though some inspections will always be needed. Looking at the long-term cost statistics, CBM offers a substantial cost advantage and will be substantially less expensive as it conforms more closely to the equipment owner's operational availability requirements.

Factors such as rising costs, reduced budgets, competitive market structures, complex equipment and employee attrition have all added to the complexity and difficulty of successfully completing the electrical maintenance mission. Implementation of CBM has been delayed in many companies by the perception that it is expensive, difficult, or both. After reviewing recent progress being made in CBM systems, it can be demonstrated that they are not difficult to implement, nor are they initially expensive. The philosophy of CBM systems is ultimately driven by how to find the perfect balance between preventive and predictive maintenance.

Reference
1) Presenation, “An Optically-Interconnected Heat-Pipe Cooled HVIGBT MW Inverter Building Block for DER Application”, P. Duncan, Airak, Inc., Oct. 26, 2003.