Check sensor history against the target curve
Asset WP-07, deviation from 14 June, lasting 9 days
Energy
It only shows up in the monthly report.Open agent
Enlarge demo
Developed in Germany · Hosted in the EU or on-premises
Delivered together with
The readings flow into the control room, the maintenance jobs sit in the maintenance system, the yields come from billing. When a performance curve drops, someone has to line up the three data series over the same period and read through the fault reports alongside them. For twelve assets, that is a week's work every month.
12
assets
Technical operations is responsible for twelve generation assets.
18
measured variables per asset
Output, availability, temperatures and meter readings add up to eighteen series per asset.
216
time series per monthly report
Twelve assets times eighteen variables gives the number of curves someone would have to look at.
5 days
until the cause is found
That is how long it typically takes to match a drop in a performance curve to a maintenance entry.
The deviation is identified on the day it occurs rather than at month end, which is where the 15% higher efficiency comes from.
The 15% efficiency gain is this agent's headline figure. The number of assets, the measured variables and the days to the cause trace the everyday work of a technical operations team and are illustrative, not measured.
The agent puts the deviation, sensor history and maintenance entry side by side; any measure is carried out from the control room once the operations manager approves it.
Asset WP-07, deviation from 14 June, lasting 9 days
Fault report 4471 of 13 June, pitch control readjusted
anlagennotiz_wp07_juni.pdf, lost yield of 41,300 kWh identified
It describes the deviation and the possible interventions. Any measure is carried out from the control room once the operations manager approves it.
It names fault report 4471 as the entry that matches in time. The technical operations manager decides whether a technician goes out to the asset.
To the agent, a frozen data series looks like an unusual curve, not a device fault. Whether the value or the sensor is the problem is for the plant engineer on site to settle.
Connected systems
Control system
maintenance system
ERP system
Operations
Deployment model
Data processing
Responsible teams
How quickly is the Asset Performance Agent ready to use?
Where is the data processed?
Where does the figure “15% higher efficiency” come from?
All agents
40%
faster case handling
Grid operations, control centre
50%
less reporting effort
Sustainability, management
60%
less preparation time
Executive board, management
30%
fewer peak load hours
Network control centre, generation, dispatch
50%
less follow-up time
Project management, sales, executive management
40%
fewer delivery delays
Purchasing, dispatch, production