Reddy GeneratorsREDDY GENERATORSThe dependable power professionals

Power Management System (PMS) Architecture Guide

What a power management system does: generator dispatch, load sharing, load shedding, restoration and reporting, and how it sits among protection, device controllers and SCADA.

In short. A PMS keeps generation matched to demand. It decides how many sources run, shares load, sheds load when capacity is short and restores it in order. It sits above device controllers and below SCADA, and it must degrade gracefully when it fails.

Core functions

Inputs and outputs

InputsOutputs
Generator status, kW, kVAr, availabilityStart and stop commands
Breaker positionsBreaker commands (within permissives)
Bus frequency and voltageSpeed and voltage set points
Load priorities and feedersShed and restore commands
Utility status and limitsMode and alarm messages

Spinning reserve

Dispatch should keep enough online capacity to absorb the loss of the largest running generator, or the largest step, without a blackout. Reserve is a design decision tied to the redundancy level. See multi-DG capacity.

Fast load shedding

When a generator trips, frequency can fall within a fraction of a second. A PMS can shed pre-selected loads based on the event, before frequency falls, which is faster than waiting for under-frequency relays. Pair with relays as backup.

Where it fits

The PMS must not remove protection from device controllers or relays. It should have defined behaviour on loss, for example generators continue sharing load by their own controllers. See control hierarchy.

Illustrative example

A plant with four generators and a 3 MW load runs two sets at light load and three at higher load. When one running set trips, the PMS sheds the lowest-priority feeders in under 100 ms, starts the standby set and restores the feeders in stages after it synchronises. Frequency stays within limits and no breaker trips on under-frequency.

What a power management system does

A PMS coordinates several generators, the utility and loads. It decides how many sets run, how load is shared and what to shed when capacity falls short.

FunctionPurpose
Load-dependent start/stopStart another set when demand rises, stop one when it falls
Active and reactive load sharingEqual percentage loading
Frequency and voltage controlHold bus within limits
Load sheddingProtect the running sets in a shortage
Blackout recoveryRestart in an orderly sequence
Standby selection and run-hour balancingEven wear, ready spare

Start/stop thresholds (illustrative)

Start the next set when loading exceeds 80 percent for 10 seconds. Stop a set when the remaining sets would be below 65 percent after it leaves, with a minimum run time of 10 to 15 minutes to avoid cycling. The gap between thresholds (hysteresis) stops hunting.

Architecture

  1. Distributed: each generator controller holds the logic and communicates by a dedicated bus. No single point of failure.
  2. Centralised: a PLC or dedicated controller coordinates. Easier to modify, needs redundancy.

Design checks

See multi-DG sequencing and load shedding automation.

The master engineering question

At any moment, what is the largest single event the plant can absorb without losing priority loads, and is the PMS keeping that margin?

What to send us

Send generator data, load priority list and one-line diagram. Send your application details.

Questions engineers ask

Is PMS the same as EMS?

No. PMS is real-time plant control. EMS optimises energy and cost over longer periods. See EMS.

Do small plants need a PMS?

Two or three sets with simple needs may manage with generator controllers.

What happens if the PMS fails?

Generators should keep sharing load and protection should remain active.

Can it be added later?

Often yes, if device controllers and breakers have the required interfaces.

This guide gives general principles. PMS design must be project-specific.

CallWhatsAppEnquire
STAGING PREVIEW ยท NOT INDEXED