What Should Engineers Include in HVAC Demand Response Design?

Demand response HVAC design enables a building to automatically shed electrical load on utility signal through pre-cooling, staged load curtailment, and automated control sequences. Good designs use the building's thermal mass to coast through events, protect occupant comfort with guardrails, and get tested before they're needed. The grid's hardest hours are hot summer afternoons — exactly when commercial buildings are at peak cooling load. Demand response turns the building stock into a grid resource: thousands of buildings each trimming load for a few hours, avoiding peaker plants and grid emergencies. For the owner, it's utility payments plus lower demand charges for control sequences that cost little to implement.

The strategy starts with the building's thermal character: heavy buildings with real thermal mass can pre-cool and coast; light buildings need more active curtailment. I model or estimate the shed potential of each strategy — pre-cooling depth and duration, setpoint relaxation, ventilation reduction, equipment staging — and sequence them into event tiers (moderate, high, critical) matched to the utility's event levels. The controls implementation is the critical piece: the utility signal interface (OpenADR or utility-specific), the automatic sequence execution, the comfort guardrails that bound the response, and the return-to-normal sequence after the event (with care to avoid a rebound peak when every building's equipment restarts simultaneously). Then testing — simulated events that prove the building actually sheds what the design claims.

Demand response is a promise to the grid. These rules make it a promise the building can keep.

• Automate the response: utility signal to BAS to curtailment — no human in the loop during the event
• Bank comfort first: pre-cooling and thermal mass strategies before setpoint sacrifices
• Guardrail the comfort: maximum temperature drift and event duration enforced by the sequence
• Stage the shed: tiered strategies matched to event severity, life-safety and critical loads never shed
• Test with simulated events: prove the shed before the utility is counting on it — and manage the rebound