When a routine start-up leads to an unexplained performance loss
A hot start should be a routine operation. The sequence is well known: the gas turbine starts and synchronizes, steam is produced progressively, and the steam is then directed towards the steam turbine, which also synchronizes. The sequence is governed by control logic that protects equipment as temperatures and pressures rise, with most steps executed automatically.
In this case, after a hot start, the combined-cycle plant began losing around 6 MWe. The site had a strong diagnostic lead, but not enough evidence to identify the exact valve involved or explain why it had remained open.
A quantified loss, but an incomplete explanation
The initial diagnostic pointed to deviations in pressure and temperature around the high-pressure steam turbine. Metroscope’s Diagnostics module identified a loss consistent with a drain leak in the high-pressure steam line and quantified its impact at up to 6 MWe.
This gave the team valuable clues: a measurable production loss that justified an investigation and a relatively small area in which to look. Yet important questions remained: Which valve was involved? And why had the issue appeared after a start-up?
Drain valves are particularly difficult to monitor in this context. They are actively controlled during start-up, and their expected position depends on the stage of the sequence. In this case, the steam turbine drain valves should open after gas turbine synchronization and close after steam turbine synchronization. A valve that remains open beyond that point can create performance losses, thermal stress and, in some cases, an increased risk of trips.
From the quasi-steady operating data alone, the diagnostic could not determine which valve had behaved abnormally during the transient phase.
What the Transient Monitoring module revealed during start-up
To investigate that missing part of the story, the team used Metroscope's Transient Monitoring module—its machine-learning model had been trained on healthy start-ups.
Rather than comparing the plant against a thermodynamic reference at stable load, the model compared the progression of the start-up immediately preceding the identification of the performance loss with the behaviour expected from previous healthy sequences. It highlighted an anomaly during that start-up.
Finding the cause and recovering performance
Operators closed the valve identified through the investigation, and performance recovered immediately. The site also planned follow-up checks on the relevant actuators and control logic to understand why the valve had remained open.
The value of the investigation was not simply that an anomaly had been detected. The plant team had sufficient evidence to act: a quantified production loss, a fault hypothesis consistent with the process measurements, and a clear deviation from the expected start-up sequence.
What flexible operation changes for plant monitoring
As combined-cycle plants operate more flexibly, cycling, frequent start-ups and load variations make transient phases a larger and more critical part of their operating life. Some failures emerge during these rapidly changing conditions but leave only indirect symptoms once the plant has returned to stable load.
The Diagnostics module, which is physics-based, remains highly effective in quasi-steady conditions: it can quantify the performance impact of an issue and identify the most likely fault mechanism. However, measurements taken after the event do not always reveal the sequence that created the issue or identify the exact component involved. The Transient Monitoring module, meanwhile, can detect when a start-up or load change deviates from healthy behaviour and highlight the measurements driving that deviation, without necessarily quantifying its subsequent operational impact.
The two approaches therefore contribute different evidence:
- The physics-based Diagnostics module quantifies the consequences and identifies the most likely failure mode.
- The data-driven Transient Monitoring module traces the issue back to the operating sequence and highlights the measurements associated with its emergence.
Used together, they connect what happened during a transient with what becomes measurable at stable load. This gives plant teams visibility across the full operating cycle and turns an incomplete diagnostic hypothesis into a targeted, actionable investigation.
For HRSGs under flexible operation, that can mean finding a 6-MWe loss before it becomes just another unexplained performance gap.
---
Looking to connect transient events with performance losses at your plant? Get in touch with our team.
