1.
Power System Loads
1.1 Load Characteristics
Every device connected to the grid is a load, and loads behave differently
depending on their nature:
Resistive Loads (heaters, incandescent lamps): draw current in
phase with voltage. They don’t create any reactive power, so they’re
the simplest to model.
Inductive Loads (motors, transformers): draw lagging current
because they need a magnetic field. They consume both active and
reactive power, which makes them affect system voltage levels
significantly.
Capacitive Loads (capacitor banks, some electronics): supply
reactive power back to the grid. They can support voltage but may
also cause overvoltage issues if not controlled.
Non-linear Loads (rectifiers, variable speed drives, computers):
draw current that is not sinusoidal, creating harmonics. Harmonics
can overheat transformers, trip relays, and waste energy.
1.2 Importance of Load Modelling
The power grid is designed and operated on mathematical models. If the
load is modeled incorrectly, simulations give misleading results.
For planning, inaccurate load models might suggest less or more
capacity than actually needed.
For operation, poor models can fail to predict instabilities during
faults.
For control, the wrong representation can result in under- or over-
compensation of reactive power.
2. Power System Stability
2.1 Voltage Stability
Voltage stability is about keeping bus voltages within acceptable limits.
When load increases, more reactive power is consumed. If the
system can’t supply it, voltages start to sag.
Induction motors, in particular, can worsen the collapse because
they draw more current at lower voltage, creating a downward
spiral.
Voltage collapse events (like in blackout case studies) often come
from poor understanding of load response.
2.2 Frequency Stability
System frequency depends on the balance between generation and load.
If load suddenly increases, frequency drops because generators slow
down.
Some loads (like motors) are frequency-sensitive and will reduce
demand when frequency falls, helping the system. Others, like
constant power loads, make the situation worse.
Frequency stability studies require accurate load response curves.
2.3 Small Disturbance Stability
These are small changes (tiny load variations, minor faults). The question
is whether the system damps out oscillations or keeps swinging.
If loads are modelled as constant, we might miss the damping
contribution of motors or electronic loads.
That’s why even small-load models matter in dynamic stability
simulations.
2.4 Transient Stability
Here we’re talking about big events—faults, sudden generator trips, line
disconnections.
Loads with dynamics (motors, composite models) can stall or trip,
changing system recovery.
If loads are wrongly modelled as static, simulations may show the
system as “stable” when in reality motors could stall and drag down
the grid.
3. Review of Past Work
3.1 Ranking of Components for Small Disturbance Stability
Researchers rank which parts of the system matter most for damping
oscillations.
Generators and exciters are usually first, but motor loads can also
significantly affect results.
Studies found that underestimating load dynamics can give a false
sense of stability.
3.2 Ranking for Voltage Stability
Voltage collapse events often trace back to large motor loads and
reactive-heavy industrial loads. In many studies, those loads ranked
higher than transmission elements in terms of risk.
3.3 Ranking for Transient Stability
Dynamic load models proved critical. With static-only models, faults
seemed survivable. With dynamic loads included, simulations showed
severe instability—much closer to what actually happens.
4. Power System Load Models and Load Modelling
4.1 Load Models
Static Exponential Load Model: Load ∝ (Voltage^α)
(Frequency^β). Good for steady-state.
Polynomial Load Model: Uses polynomial functions of voltage for
more accuracy.
Linear Load Model: Assumes proportionality, simplest but least
realistic.
Static Induction Motor Model: Adds torque-speed dynamics,
stalling behavior. Very useful for industrial systems.
Exponential Dynamic Load Model: Time-dependent, captures
load recovery.
Composite Load Model: Mixes static (lights, electronics) with
dynamic (motors) for real-world accuracy.
4.2 Load Modelling Methodology
Two main ways:
1. Component-Based: Build from bottom up (e.g., model 50%
lighting, 30% motors, 20% electronics). Good for design studies.
2. Measurement-Based: Use actual feeder or substation data. More
accurate for real-time applications.
4.3 Automatic Identification
With smart meters and PMUs, load models can be identified automatically.
Steps:
Collect voltage/current data.
Choose candidate models.
Use parameter estimation to fit curves.
Validate with test cases.
5. Importance of Load Management
5.1 System-Wide Impact
When demand is smooth, generation runs more efficiently, fewer reserves
are needed, and transmission losses drop.
5.2 Demand vs Customer Profile
A steel mill’s load profile looks nothing like a household’s. But combined,
they shape the system demand. Understanding both helps utilities plan.
5.3 Coordinated Load Management
When multiple customers cooperate, peak demand can be cut
significantly. For example, 10 industries shifting load together makes a
noticeable system difference.
5.4 Advantages
Cost savings (lower peak tariffs).
More reliable grid.
Delayed investment in new power plants.
Easier integration of renewables.
5.5 Industrial Example
A cement plant can reschedule grinding mills to night hours. It saves the
company money and helps the utility keep peak hours stable.
6. Market Aspects of Load Management
6.1 Market Activities
Load management creates opportunities like demand response programs
where customers are paid to reduce usage during peaks.
6.2 Contracts
Utilities offer contracts with incentives for reducing load on request.
Failure to comply might mean penalties.
6.3 Risk Treatment
Technical: Equipment may fail to shed load as required.
Financial: If incentives don’t cover costs, customers may drop out.
6.4 Utility-Customer Interaction
Trust is key. Customers must believe they’ll be fairly compensated, and
utilities need reliable compliance.
7. Methodology
7.1 Energy Estimation in Manufacturing
To plan load management in factories, measure each machine’s
consumption and operating hours. This gives baseline energy use.
7.2 Load Curve Development
Plot hourly usage to see peaks and valleys. This curve shows where to
shift or reduce demand.
