Purpose and Scope
The per-unit (pu) system is one of the most widely used normalisation frameworks in power system analysis. It appears in load-flow studies, fault calculations, EMT simulations, and equipment ratings. Despite its widespread use, confusion often arises around why the per-unit system works so well and when it provides real engineering value.
This article focuses on the conceptual foundation of the per-unit system. It intentionally avoids detailed base calculations, impedance derivations, or base-changing procedures, which are addressed in subsequent articles in this series.
What the Per-Unit System Is
The per-unit system expresses electrical quantities as ratios relative to chosen reference values, known as base quantities. Any quantity expressed in per-unit is dimensionless and represents how large that quantity is relative to its base.
In general form:
The key idea is not the arithmetic itself, but the normalisation it provides across different voltage levels and equipment ratings.
Why Normalisation Is Necessary in Power Systems
Power systems inherently span multiple voltage levels, power ratings, and equipment types. A single study may involve generators, transformers, transmission lines, and distribution feeders operating at very different scales.
Working directly in physical units quickly leads to wide numerical ranges and poor comparability. Per-unit normalisation compresses these ranges and places diverse components on a common numerical scale.
This directly improves numerical stability, interpretability of results, and error detection during modelling.
Why Per-Unit Values Tend to Be Similar Across Voltage Levels
A key property of the per-unit system is that impedances of similar equipment often fall within a narrow numerical range when expressed in per-unit, regardless of voltage level.
This reflects the fact that electrical equipment is designed to operate with similar electromagnetic loading and insulation stress when scaled appropriately. When base quantities are chosen consistently, these similarities appear naturally in per-unit values.
Relationship Between Per-Unit and Physical Laws
The per-unit system does not alter physical laws. Ohm's law, Kirchhoff's laws, and power relationships remain valid under per-unit transformation.
For example:
becomes:
provided the base quantities are consistent.
Practical Benefits in Engineering Studies
From an engineering perspective, the per-unit system offers several advantages. Results become easier to compare across network sections, typical operating ranges become intuitive, and modelling errors are easier to identify.
In EMT and RMS simulations, per-unit representation also reduces sensitivity to absolute scaling, helping engineers focus on physical behaviour rather than unit conversion.
What the Per-Unit System Does Not Do
The per-unit system does not eliminate the need to understand physical units or replace engineering judgement. Misaligned base quantities remain a common source of errors, particularly when combining models from different sources.
Role of Base Quantities
All per-unit calculations depend on the selection of base quantities, typically base apparent power and base voltage. From these, other base quantities are derived.
The implications of base selection are addressed in Per-Unit System: Base Quantities and Their Relationships. The Base Quantity Calculator can help derive consistent base quantities from selected inputs, reducing manual calculation errors.
Position Within the Article Series
This article establishes why the per-unit system is used and why it works. The following articles focus on implementation:
- This article — foundational theory
- Per-Unit System: Base Quantities and Their Relationships — deriving base values
- Per-Unit Impedance from Physical Parameters — converting R, L, C to per-unit
- Per-Unit System: Changing Base Across Transformers and Networks — changing between bases
- Per-Unit System: Common Mistakes in EMT and RMS Studies — common mistakes and how to avoid them
Reflective Questions
- Were per-unit bases explicitly documented in your recent studies?
- Have you encountered inconsistent per-unit values for similar equipment?
- Would automated base-consistency checks reduce errors in your workflow?