Monitoring systems are often commissioned too late in a project, after mechanical and electrical equipment has already been tested. This is a mistake. BMS, EPMS and DCIM are part of the operational infrastructure and must be proven with the same discipline as the equipment they monitor.
Start with the point list
Every monitored point should have a defined source, description, engineering unit, scaling, normal state, alarm state and destination platform. The point list should match the final equipment and not remain based on early design schedules.
Point-to-point testing
Point-to-point testing verifies that the physical or protocol signal shown on the monitoring platform corresponds to the actual equipment state. Digital points should be toggled where safe, and analog values should be compared with local readings or calibrated reference values.
Scaling and engineering units
A communication link can be healthy while a value is wrong because CT ratios, PT ratios, register scaling or engineering units are incorrect. Verify kW, current, temperatures, pressures, tank levels and percentages against the source device.
Alarm verification
Test alarm threshold, delay, priority, text, acknowledgement and escalation. A correct point with the wrong alarm priority is still a commissioning defect.
Also verify restoration behavior so alarms clear only when the condition has genuinely returned to normal.
Control sequences
Where the BMS issues control commands, functional testing should verify start/stop logic, lead/lag sequences, VFD commands, valve control, setpoints, interlocks and failure fallback.
Controls should be tested through realistic operating ranges rather than only by forcing software values.
Timestamps and event sequence
Ensure clocks are synchronized across BMS, EPMS, meters, relays, UPS controllers and other relevant devices. Without consistent time, incident investigation becomes unreliable.
Integration testing
Verify data passing between BMS, EPMS and DCIM, including protocol gateways, naming, units, alarms and quality flags. The receiving platform should clearly show when data is stale or communication is lost rather than continuing to display an old value as if it were current.
Network and server failure scenarios
- Loss of one monitoring server.
- Failure of a protocol gateway.
- Loss of network connectivity to a controller.
- Stale data or bad-quality value.
- Database or historian interruption.
- Failure of email/SMS notification service.
- Loss of time synchronization.
Dashboard validation
Dashboards should reflect the real system configuration and provide useful context. Verify equipment names, room names, A/B path identity, capacity values and alarm counts. Attractive graphics are not useful if they represent the wrong system.
Integrated Systems Testing
During utility-failure, cooling-failure and other integrated scenarios, verify that the monitoring platforms capture the correct sequence, generate appropriate alarms and provide operators with enough information to understand the event.
Handover requirements
Handover should include final point lists, architecture diagrams, network addressing, protocol maps, backups, user accounts, alarm philosophy, trend configuration, maintenance procedures and training. Monitoring systems should be supportable without depending on undocumented knowledge held by one integrator.
Key takeaway
BMS, EPMS and DCIM commissioning is not a graphical-interface review. It is a technical proof that every important point is accurate, every alarm is meaningful, every control sequence behaves correctly, integrations fail safely and operators receive trustworthy information during real events.
References and Further Reading
- ANSI/ASHRAE Standard 135, BACnet.
- ISO/IEC TS 22237-7:2018, data center management and operational processes.
- Applicable BMS, EPMS and DCIM manufacturer commissioning documentation.
