Proven at compressor stations
Deployed random-forest and gradient-boosting fuel-optimization models across 30+ North American compressor sites delivered 3–5% savings — then drifted within weeks. Adaptive DRL per component holds the gains and goes further.
~11%
energy & emissions savings with edge-deployed adaptive DRL
$535k
annual savings per typical station at 2026 Canadian carbon pricing
$778k
annual savings per typical station at 2030 carbon pricing
30+
North American sites of hands-on compressor optimization experience
Cost breakdown per typical station
2026 Canadian carbon pricing — savings vs baseline across optimization approaches.
| Scenario | Gas cost ($k) | Carbon cost ($k) | Total ($k) | Saving vs baseline |
|---|---|---|---|---|
| Baseline | 955.3 | 3,907.2 | 4,862.5 | — |
| DCS tuning (2%) | 936.2 | 3,829.1 | 4,765.3 | $97.3k |
| Traditional ML (4%) | 917.1 | 3,750.9 | 4,668.0 | $194.5k |
| OmniPath Edge DRL (11%) | 850.2 | 3,477.4 | 4,327.6 | $534.9k |
Per typical station, 2026 Canadian carbon pricing. Assumptions: 11 MW average duty, 8,400 h/yr; AECO-C fuel pricing; 0.37 t CO₂e/MWh; federal carbon price C$110/t (2026), C$170/t (2030).