See where performance is lost
Locate affected systems and see how individual faults contribute to the overall performance gap.
Act on performance losses with confidence
Metroscope combines thermodynamic Digital Twins with constrained AI to quantify power losses, identify causes, and help teams coordinate resolution actions.
- 10,000
- equations solved for each diagnosis
- 80
- specific failure modes modeled per plant
- 9/10
- fault suggestions validated after review
The operational challenge
Performance losses are hard to size and harder to explain
A clear diagnosis gives performance, operations, and maintenance the same basis for action.
How much power are you losing?
Changing load and operating conditions make the performance gap hard to quantify.
Where is the loss coming from?
Concurrent faults can affect the same measurements and obscure each other.
Where should your team act first?
Teams need to see which faults have the greatest impact before planning action.
Why Metroscope
Connect plant priorities with the evidence to act
Decide what needs immediate investigation
Review events by priority and status, with their detection history and associated losses.
Understand the fault and its impact
Locate the diagnosed fault in the process, compare measured and expected behavior, and review how symptoms developed to guide site checks.
Keep the investigation and follow-up connected
Track fault evolution, share site findings and action status, and assess performance recovery after maintenance.
How Diagnostics works
From plant measurements to a diagnosis
- Key principles
From plant data to a diagnosis
Diagnostics combines sensor data with a thermodynamic Digital Twin of your process. Differences between measured and expected behavior form symptoms, which the inferential engine uses to identify the most likely faults and quantify their impact on power output.
- Digital Twin
A thermodynamic model of your process
Our modeling team builds a numerical replica of your process. The Digital Twin simulates how the process behaves and how faults change sensor readings and power output.
- Inferential Engine
From symptoms to the most likely faults
The inferential engine uses a probabilistic approach to interpret the observed symptoms. It identifies the most likely combination of simultaneous faults, estimates each fault’s magnitude, and quantifies its impact on power output.
Testimony and use case
Diagnostics on operating plants
“Metroscope clearly identified a moisture separator reheater (MSR) tube rupture and quantified its impact, helping the Axpo team act quickly.”
Daniel Schupp
Thermal Performance Engineer, Axpo
Diagnosing a passing steam turbine bypass valve
Combined sensor deviations pointed to a passing HP bypass valve, with an estimated impact of 1 MW.
Delivery and support
Expert support from setup to daily use
A plant-specific model, practical training, and support for the investigations that follow.
Build and validate
Metroscope builds the diagnostic model for your plant and validates it with operating context.
Put it in your team’s hands
Your team is trained to read diagnostic findings and use the model in daily work.
Investigate over time
Metroscope helps interpret results as your team reviews findings and plans its response.
Your team
Provides operating context, checks findings on site, and decides on maintenance actions.
Metroscope
Builds and validates the model, trains users, and supports interpretation of the results.
Extend the investigation with complementary modules
Anomaly Detection
Which equipment needs attention first?
Monitor system reliability by detecting and investigating abnormal equipment behavior.
See moduleTransient Monitoring
Did the operating sequence behave as expected?
Automatically detect emerging issues during transient phases and support flexible operations.
See moduleCapacity Forecast
What can the plant deliver?
Forecast available power and associated heat rate based on plant and weather conditions.
See moduleBring greater clarity to your next maintenance decision