Physical model
Heat loads, airflow, chiller curves and tariffs are represented explicitly and calibrated on data from the facility.
Technology
A physical model, hard constraints and a stated reason for each decision.
Aphelium models the facility from the laws of thermodynamics and from operating data, then optimizes over that model.
Architecture
From sensors to measured savings.
Sensors
BMS, power meters, temperature sensors
Model
Heat loads, airflow, chiller curves, tariffs
Optimizer
Lowest energy inside hard limits
Recommendation or control
Applied by operators, or automatically
Measurement
Verified against the agreed baseline
Approach
Three components.
Optimization
The optimizer seeks the lowest energy consumption that satisfies all temperature, humidity and redundancy limits. A limit is a constraint, never a cost to be traded.
Explainable decisions
Each setpoint change states the expected kWh saving and the reason. Decisions can be traced and rolled back.
Phases
Four steps, from read-only to control.
Week 1
Connect
Read-only link to the BMS, power meters and temperature sensors. Any vendor, old or new.
Weeks 2–4
Model
A physical model of the facility: heat loads, airflow, chiller curves, tariffs.
Month 2
Recommend
Setpoint changes, each with the expected kWh saving and the reason. Your team applies them.
Month 3+
Control
Automatic control inside hard limits. Operators can override at any time.
Scope
What Aphelium reads and what it changes.
What Aphelium reads
- BMS points and alarms
- Power meters
- Temperature and humidity sensors
- IT load
- Electricity tariffs
What Aphelium changes
- Chilled water setpoint
- CRAH fan speed
- Chiller staging
- Supply air temperature
01Integration
Aphelium connects to existing BMS and SCADA systems and meters through standard industrial protocols, across vendors and equipment ages. No hardware replacement is required.
- BACnet
- Modbus
- OPC UA
- SNMP
02Roadmap
Grid flexibility services, stranded power recovery, joint optimization of on-site generation and storage with facility load, and facility-wide digital twins.