Reliability Is the Product
In power generation, reliability isn’t a maintenance department objective — it’s the core business deliverable. A generating unit that can’t run when dispatched loses revenue, incurs replacement power costs, and may face contractual penalties. Forced outage rates directly impact the plant’s financial performance. Every percentage point of forced outage rate represents millions of dollars in lost revenue for a large generating unit.
The North American Electric Reliability Corporation (NERC) tracks generating unit performance through metrics like Equivalent Forced Outage Rate (EFORd), which measures the probability that a unit will not be available when needed due to forced outages and deratings. Top-quartile performance for coal and gas-fired units targets EFORd below 3-4%. Achieving this requires disciplined maintenance, comprehensive condition monitoring, and effective outage planning.
Gas Turbine Reliability
Gas turbines — both heavy-frame industrial units and aeroderivative designs — are the primary prime movers in modern power generation. They operate at extreme temperatures (firing temperatures exceeding 1,300°C in modern designs) and high rotational speeds, creating demanding conditions for hot gas path components, bearings, and auxiliary systems.
Hot Gas Path Components
Combustion liners, transition pieces, first-stage nozzles, and first-stage buckets (blades) experience the most severe thermal and mechanical stress in the turbine. Their condition directly determines turbine output capability and efficiency.
OEMs define inspection intervals in terms of equivalent operating hours (EOH) or equivalent starts — both adjusted for operating factors like fuel type, load cycling, and peak firing temperature. A typical heavy-frame gas turbine hot gas path inspection interval is 24,000-32,000 EOH. Combustion inspections (limited to the combustion section) occur more frequently — every 8,000-12,000 EOH.
Between scheduled inspections, monitor combustion dynamics (pressure pulsations in the combustor) to detect combustion instability that accelerates liner and transition piece degradation. Exhaust temperature spread — the variation in exhaust thermocouple readings across the turbine exhaust plane — indicates combustion imbalance. A spreading exhaust temperature pattern suggests a deteriorating fuel nozzle or combustion liner.
Gas Turbine Vibration
Gas turbines require continuous vibration monitoring per API 670. Proximity probes at each bearing measure shaft vibration. Accelerometers on bearing housings measure casing vibration. Dual-channel monitoring at each bearing provides orbit plots that reveal shaft dynamic behavior.
Critical trip setpoints are typically set at 5-7 mils (125-175 microns) peak-to-peak shaft vibration for large frame units, though specific values depend on the OEM and unit design. Alert setpoints at 50-60% of trip values provide early warning for trending and investigation.
Inlet Air Filtration
Gas turbine compressor fouling from airborne contaminants reduces power output by 3-7% between washes. Online water washing restores some lost performance, but offline chemical washes (performed during planned outages) are needed to recover fully. The frequency of washing depends on the ambient air quality — coastal or industrial environments require more frequent washing than rural locations.
Invest in high-quality inlet air filtration. Multi-stage filter systems (prefilter, high-efficiency final filter, and possibly HEPA in harsh environments) extend the interval between compressor washes and protect hot gas path components from particle erosion.
Steam Turbine Reliability
Steam turbines in combined cycle and conventional steam plants have long overhaul intervals — typically 8-12 years for major inspections. Between overhauls, condition monitoring ensures the unit reaches the next planned outage without forced events.
Key Monitoring Parameters
- Vibration — Shaft and bearing vibration per API 670. Steam turbines are sensitive to thermal transients during startup and load changes. Differential expansion (the difference in thermal growth between the rotor and casing) must be managed to prevent rub events.
- Bearing condition — Journal bearing metal temperatures and oil drain temperatures trend bearing wear and lubrication condition. Thrust bearing monitoring is critical — axial displacement beyond limits causes blade tip contact with stationary seals or diaphragms.
- Valve testing — Control valves, stop valves, and intercept valves require periodic stroke testing to verify they’ll function during a trip event. Online valve testing (partial stroke) on a monthly or quarterly basis confirms valve freedom without interrupting generation.
- Condenser performance — Condenser backpressure directly affects turbine efficiency. Rising backpressure from tube fouling, air in-leakage, or cooling water flow restriction reduces output. Monitor condenser vacuum, terminal temperature difference, and cleanliness factor.
Generator Reliability
Generators are the revenue-producing asset — if the generator is out of service, the plant isn’t earning. Generator maintenance focuses on rotor and stator insulation, cooling systems, and hydrogen/seal oil systems (for hydrogen-cooled units).
Stator Winding Monitoring
Partial discharge (PD) monitoring per IEEE 1434 is the primary online insulation assessment tool for generator stator windings. Continuous PD monitoring systems detect insulation degradation trends over months and years. Step increases in PD activity indicate developing insulation problems that warrant investigation during the next planned outage.
Stator cooling water conductivity (for water-cooled stators) must be maintained below OEM limits to prevent electrical tracking. Monitor conductivity continuously with alarming for rising trends.
Rotor Monitoring
Generator rotor winding condition is assessed through flux probe monitoring (detecting shorted turns by measuring the magnetic flux pattern) and shaft voltage monitoring (detecting ground faults in the rotor winding). Shorted turns reduce generator reactive capability and create thermal asymmetry that can cause vibration.
Balance of Plant: The Systems That Support Generation
Prime movers and generators get the most attention, but balance-of-plant (BOP) systems cause a disproportionate share of forced outages. Feedwater pumps, circulating water pumps, cooling towers, fuel gas systems, and instrument air systems are all single points of failure in many plant configurations.
Feedwater and Boiler Feed Pumps
These are among the most critical rotating equipment in a steam plant. They operate at high pressure (boiler feed pumps often exceed 2,000 psig discharge) and are typically furnished to API 610 standards. Full condition monitoring — vibration, oil analysis, bearing temperature, seal system monitoring — is standard. Many plants install redundant pumps specifically because feedwater system reliability is non-negotiable.
Cooling Water Systems
Circulating water pumps, cooling tower fans, and the condenser tubes require consistent maintenance to support turbine efficiency. Biofouling control, chemistry management, and tube cleaning schedules directly affect plant heat rate and output capability. Neglecting the cooling water system is like neglecting the condenser — it silently erodes plant performance until a hot day reveals that you can’t make full load.
Fuel Systems
Natural gas fuel systems require gas pressure regulation, heating (to prevent hydrocarbon liquids from reaching the turbine), and filtration. Fuel gas filter differential pressure monitoring and regular strainer cleaning prevent fuel quality issues that damage gas turbine fuel nozzles and combustion hardware.
Outage Planning
Planned outages in power generation are high-stakes events. Every day of outage beyond the planned duration is a day of lost revenue. Effective outage planning is a reliability discipline in itself.
Condition monitoring data drives outage scope development. What work actually needs to be done? Vibration trends, oil analysis results, performance data, and borescope inspection findings from the period between outages determine which components need attention and which can continue in service. This condition-based outage scoping replaces the traditional approach of opening everything on a calendar interval — reducing outage duration, cost, and the risk of maintenance-induced failures from unnecessary disassembly.
Track outage duration, scope changes (work added after the outage began), and post-outage forced outage events (maintenance-induced failures). Use these metrics to improve outage planning for the next cycle. Plants that track and analyze outage performance data achieve shorter, more predictable outages over time.
Power generation reliability is a mature discipline with well-established practices, standards, and performance benchmarks. The fundamentals — comprehensive condition monitoring, rigorous maintenance execution, and data-driven outage planning — are well understood. The plants that execute these fundamentals consistently are the ones that achieve top-quartile availability year after year.