Case studyPower & energyOptimize throughput

Heat rate cut 3.1% across a full season at a waste-to-energy plant.

Combustion advisor uses fuel-mix forecasting to hold steam quality at lower excess air.

Talk to an expertDownload 2-page PDF
ForecastingPrescriptive optimizationOperator workflow
MATURITY LADDERclosed-loop
01Monitor
100%
02Recommend
84%
03Approve
71%
04Automate
62%
At a glancetimeframes and caveats below
−3.1%
heat rate
12 months
Same fuel mix bands, weather-normalised
0
advisor-attributable trips
Same period
Verified in trip log
+1.9%
net MWh output
Same period
On comparable availability
14 weeks
pilot length
Single unit
Before season rollout
Challenge

What was happening before.

Heat rate had been flat for three seasons. Operators wouldn't tighten the air margin without a defensible signal that combustion would stay within stability limits.

Constraints
  • Boiler stability over efficiency, always.
  • Emissions remain within permit at every minute.
  • Recommendations must respect ramp limits on ID fan and grate speed.
Solution

What we built, on what data.

Data
  • › Historian tags: O₂, CO, NOx, steam flow/temp, grate speed, ID fan
  • › Fuel intake: moisture sample, calorific estimate
  • › Permit data: continuous emissions monitoring
  1. 01
    Forecast fuel calorific value
    Combined moisture, season, and intake mix into a 4-hour calorific forecast.
  2. 02
    Prescribe O₂ and grate settings
    Optimizer proposes setpoints that hold steam quality at lower excess air.
  3. 03
    Operator approval per shift
    Operator approves window; recommendations stay within learned safe envelope.
  4. 04
    Post-shift review
    Plant manager reviews compliance vs efficiency tradeoff with one chart.
Adoption

Who uses it, and where it fits in the day.

Used every shift by the control-room operator. Plant manager reviews weekly. The combustion engineer owns the model and signs off on retrains.

Modules in use
ForecastingPrescriptive optimizationOperator workflow
Results

Quantified, with timeframe and caveats.

  • R1Heat rate down 3.1% over a 12-month season.
  • R2Net MWh output up 1.9% on comparable availability.
  • R3Zero boiler trips attributable to the advisor in trip-log review.
MATURITY LADDERclosed-loop
01Monitor
100%
02Recommend
84%
03Approve
71%
04Automate
62%

Operational outcome (proof rule): Heat rate improved 3.1% over a 12-month season at a single waste-to-energy unit with no boiler trips attributable to recommendations.

Voice from the floor
"We're running the same boiler, with the same crew, just better tuned to what's actually in the bunker today. That's the whole game."
Combustion Engineer · Process & control · Returkraft Energy
Replication

Relevant for teams trying to…

Power & energy operators on boilers, turbines, or combustion units with variable fuel.

Solution: Optimize throughput →Industry: Power & energy →
Lessons
  • Forecast the fuel before you forecast the steam. The downstream models are only as good as the calorific estimate.
  • Tie every advisor recommendation to the permit metric in the same view, or the control room will not act on it.
  • Pilot through one season — short pilots miss the moisture-driven regimes.
Get started

Talk to an expert about a similar use case.

Tell us your unit and your KPI. We'll come back with what data you'd need and a realistic timeline based on Returkraft Energy's rollout.

Book a demoDownload 2-page PDF