CHP Engineer Series · Remote Monitoring
Remote Monitoring for CHP Plants
A modern block heat & power plant (BHKW/CHP) reports every minute — so a fault is caught early and maintenance is planned without an on-site visit. Here is exactly how it works, which alarms matter, and where the thresholds sit.
01How is a CHP remotely monitored?
Modern CHPs communicate over Modbus/TCP or an HTTP API with a central monitoring system. Operating parameters are written to a database every minute — that continuous stream is what makes early fault detection and predictive maintenance planning possible.
SOURCE
Plant controller
Engine, generator, grid-handover point and heat side expose live parameters.
TRANSPORT
Modbus/TCP · HTTP API
Values pushed to the operator portal at a one-minute cadence.
DETECTION
Portal & database
Trends stored, thresholds evaluated, alarms raised before a plant trips.
02Remote alarm: what the plant calls you about
Modern CHP controls send fault notifications by SMS, email or push notification. Fast response minimises downtime and prevents consequential damage. These are the alarms that matter most:
Oil pressureSAFETY
Loss of lubrication pressure trips the engine — one of the fastest hard-stop protections.
OvertemperatureTHERMAL
Cooling-water and exhaust temperatures rising past limits signal cooling or load faults.
Grid protectionELECTRICAL
Voltage or frequency excursion at the grid-handover point — see the relay thresholds below.
NOx limitEMISSIONS
Exceeding the NOx limit risks non-compliance; the control flags it before it becomes a reportable event.
03Grid-protection relay — the numbers
The relay watches the handover point between the CHP and the public network. On a grid failure it disconnects within 200 ms, which prevents islanding and protects the plant from grid disturbances.
| Quantity | Upper trip | Lower trip |
|---|
| Voltage (U) | U>> 110 % | U< 85 % |
| Frequency (f) | f> 50.2 Hz | f< 49.8 Hz |
| Disconnect time | ≤ 200 ms on grid failure |
04Temperature watch on the heat side
Temperature monitoring on a heat generator such as a Jenbacher 500 kWel CHP tracks cooling-water and exhaust temperatures continuously. Because the plant produces heat and power simultaneously, a drift on the thermal side is often the earliest visible sign of a developing fault — long before an electrical protection would trip.
05Statistical alarm levels — gas-pressure example
Fixed thresholds miss slow drift. A better approach sets alarm levels on the statistical deviation from the plant's own baseline (standard deviations, σ). On real gas-pressure data we see Z-values between 3.1 and 6.8 — so graded levels catch a problem while it is still developing:
Yellow warning
+1.5 σ
Early drift — log & watch the trend.
Orange warning
+2.5 σ
Clear deviation — plan an intervention.
Red alarm
+3.5 σ
Act now — notify and protect availability.
A monitoring system that continuously measures, detects statistical deviations, sends automated notifications and stores history for trend analysis turns raw sensor data into plant availability.
How Stromfee monitors CHP: diagnostics written as code
We design CHP diagnostics computationally — inspired by LEAP 71's Computational Engineering Models. Thirty years of energy-technology expertise, encoded into causal chains and live sensors, to operate large-scale CHP plants in an extremely flexible electricity market.
→ View the live score & active alarms
06FAQ
How is a CHP remotely monitored?
Modern CHPs communicate via Modbus/TCP or an HTTP API with central monitoring systems; operating parameters are written to databases every minute — remote monitoring enables early fault detection and maintenance planning without on-site visits.
How does the CHP remote alarm work?
Modern CHP controls send fault notifications via SMS, email or push notification. Typical alarms: oil pressure, overtemperature, grid protection and NOx limit — fast response minimises downtime and prevents consequential damage.
What does the grid-protection relay do?
It monitors voltage (U>> 110 %, U< 85 %) and frequency (f> 50.2 Hz, f< 49.8 Hz); on grid failure it disconnects within 200 ms — this prevents islanding and protects the plant from grid disturbances.
Why use statistical alarm levels instead of fixed limits?
Fixed limits only trip at the end. Graded levels on the standard deviation (yellow +1.5 σ, orange +2.5 σ, red +3.5 σ) detect drift while the plant is still running normally, so you can plan the intervention instead of reacting to a trip.