ECpE Department

Distribution System Planning

Dr. Zhaoyu Wang

1113 Coover Hall, Ames, IA
wzy@[Link]
Distribution System Planning

Acknowledgement:
The slides are developed based in part on Electric Power and Energy
Distribution Systems, Models, Methods and Applications, Subrahmanyan
S. Venkata, Anil Pahwa, IEEE Press & Wiley, 2022

ECpE Department
1. Overview
• Distribution system planning: Meeting future electricity needs
of customers.
• Complex and capital-intensive process due to changing loads
(population, technology, habits).
• Multiple objectives considered simultaneously, while meeting
technical constraints.
• Goal: Minimize total cost, ensure adequate capacity to supply
the load for the future with adequate reliability and
acceptable voltage quality.
• Plans address both short-term (1-2 years) and long-term (5-10
years) needs.
3

ECpE Department
1. Overview
• First stage: Load forecasting for future, identifying growth
areas.
• Second Stage: Determine location & capacity of substations;
number & size of transformers; and reinforce existing or build
new substations based on planning results.
• Last Stage: Feeder design to deliver power from substations
to customers.
• Feeder design includes both primary and secondary systems.

ECpE Department
1. Overview
• Decisions: Number of primary feeders, size and routing of
conductors.
• Secondary level decisions: Location, size of distribution
transformers.
• Most of the secondary construction in US has been overhead
so far.
• There is a trend toward making new secondary construction
underground.
• Underground system can be more expensive (5-10 times)
than overhead system, but underground secondary system is
preferred by some localities for greater reliability and
aesthetics. 5

ECpE Department
2. Traditional vs. Modern Approaches to Planning
Legacy distribution systems:
• Only loads on substation transformers and feeders are
measured.
• Little information beyond the feeder.
• Rely on customer billing data and total distribution
transformer capacity for load estimates.
• As part of the load research activity, utilities installed
recording devices at selected customer locations to record
loads at a predetermined interval (5 minutes, 15 minutes, 30
minutes, or 1 hour) to get an idea of the daily load profiles for
different classes of customers. These load profiles, especially
the peak demands, are also useful for planning purposes. 6

ECpE Department
2. Traditional vs. Modern Approaches to Planning
Legacy distribution systems:
• Diversity factor, maximum non-coincident demand, maximum
diversified demand
• Limited data led to ad hoc distribution planning, often
resulting in overdesigned systems.

ECpE Department
2. Traditional vs. Modern Approaches to Planning
• Modern systems: More metering capabilities thanks to
distribution automation and advanced metering
infrastructure. Increased automation provides accurate data,
eliminating some assumptions in planning.
• Traditional approach: Design to meet peak demand.

ECpE Department
2. Traditional vs. Modern Approaches to Planning
• Modern approach:
o Uses risk analysis.
o Requires data on hours spent at different load levels to
determine the risk of insufficient capacity to meet the
demand over a period of time.
o Specifically, the number of hours under peak conditions is
very important for risk assessment.
o System reliability, resiliency, aging assets, equipment
loading, regulatory environment considered in modern
planning.

ECpE Department
3. Long – Term Load Forecasting
• Load forecasting: Determining future load in an area, short-
term forecasts are needed (hours to days) for operation,
while long-term (1 year or more) forecasts are needed for
planning.
• Knowledge of future loads is crucial for designing distribution
system facilities and ensure the reliability.
• Planners need to know peak load and its timing.
• In addition to temporal forecasts, spatial forecasts are
important because they tell the locations of load growth
locations in addition to quantity of load growth.

10

ECpE Department
3. Long – Term Load Forecasting
• Load forecasts depend on local and national trends.
o At local level, population growth, new industrial or
commercial facilities planned.
o At national level, high efficiency appliances.
• Appliance manufacturers and consumer organizations
conduct regular surveys to keep track of consumer behaviors
to determine these trends.
• Another approach to forecasting is based on end-use
modeling in which loads of different end uses, such as
lightning, water heater, and AC, for different customers are
added in a hierarchical manner to obtain the composite load.
11

ECpE Department
3. Long – Term Load Forecasting
• Composite load in a utility service area shows a gradual
increase in load over years.
• Composite load in small areas follow an “S” curve with
three parts: dormant, growth, saturation periods as shown
in figure (a).

Fig. a: S‐curve showing load growth in a small area.

12

ECpE Department
3. Long – Term Load Forecasting
• Combining the composite loads of all small areas in within a
utility exhibits linear characteristics, reflecting overall load
growth for a utility service area as shown in figure (b).
• Breaking a utility service area into smaller areas is useful for
spatial load forecasting.

Fig. b: Load growth of a service area over the years.

13

ECpE Department
4. Load Characteristics
4.1 Customer Classes
• Utilities serve various customer classes:
residential, commercial, industrial,
agricultural.
• Different classes have distinct electricity
usage patterns, leading to varying load
demands.
• Within a class, customers exhibit similar load
patterns.
• For example, residential load is low at night,
rises in morning, peaks during the day,
declines in the evening.
• Commercial and industrial loads: High during
the day, low at night.
• Commercial load variation larger due to
extended operating hours for some
14
industries.

ECpE Department
4. Load Characteristics
4.1 Customer Classes
• Industrial loads: Some industries
operate 24/7, contributing to
relatively steady demand.
• Agricultural load demands vary
based on irrigation needs for
different crops.
• Load demands within each class
also have seasonal fluctuations.
• Utilities maintain load demand
databases for each class to be used
in planning. 15

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House
• Changes in residential loads:
o More efficient appliances.
o New types of loads: computers, entertainment devices.
o Transformation in lighting: incandescent to CFLs to LED
lamps.

16

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House
• 2015 Energy Information Administration (EIA) survey data:
o Average monthly electric energy consumption for US
residential customers: 901 kWh.
o Average monthly bill: $114.03.
o Average price: $0.1265/kWh.
o State variations: Washington (lowest) $0.0909/kWh, Hawaii
(highest) $0.296/kWh, Connecticut (lowest 48 states)
$0.2094/kWh.
o Highest consumption: Louisiana (1286 kWh), lowest: Maine
(556 kWh).
o Louisiana: High AC use due to hot, humid climate. Maine: Low
electricity consumption due to gas-based heating.
17

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House
• Energy consumption breakdown (2015):
o Electricity: 47%, natural gas: 44%.
o Half of home energy consumption is heating, cooling,
ventilation.
• Key energy-consuming uses (2015):
o Air conditioning: largest energy consumption, followed by
space cooling, water heating.
o Lighting: 10%, refrigerators: 7%, TVs: 7%, clothes dryers:
5%.

18

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House
• Key energy-consuming uses (2015):
o Miscellaneous appliances (12%): freezers, dishwashers,
washers, cooking ranges, microwaves, etc.
o Other appliances (13%): computers, home audio, coffee
makers, etc.
• Specific appliances:
o HVAC: 1-6 kW
o Refrigerator: about 600 W, cycles on/off for temperature
maintenance.
o Stand-alone freezer: typical rating of 500 W.
19

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House
o Clothes washer: about 500 W, dishwasher: about 1.2 kW,
clothes dryer: about 4 kW.
• Trends over years:
o Increase in electronics portion (large flat screen TVs,
computers).
o Decrease in lighting portion (proliferation of LED lighting).

20

ECpE Department
4. Load Characteristics
4.2 Loads in a Modern House

• Following table illustrates the energy usage in the US

homes in 2015. Source: Data from U.S. Energy Information Administration

21

ECpE Department
4. Load Characteristics
4.3 Time Aggregation
• Instantaneous power: Voltage × current, not useful for
planning due to large fluctuations.
• Demand (in kW): Average power over selected time period,
used for planning.
• Typical utility time periods: 1 min, 5 min, 15 min, 30 min, 1
hour.
• Calculate demand: Energy consumed (kWh) divided by time
period (hours).
• Example: Load consumes 2 kWh in 15 min (0.25 h), demand =
2/0.25 = 8 kW.
22

ECpE Department
4. Load Characteristics
4.3 Time Aggregation
• Another example: Load consumes 7.5 kWh in 1 hour, demand
= 7.5/1 = 7.5 kW.
• Reported demand at a given time is average load for the
previous time period.
• Example: Demand reported at 2:00 a.m. (one-hour basis) is
average load between 1:00 a.m. and 2:00 a.m.
• For 15-minute reporting: Reported value at 2:00 a.m. is
average load between 1:45 a.m. and 2:00 a.m.

23

ECpE Department
4. Load Characteristics
4.3 Time Aggregation
• Single-house demand characteristics within a short time period
(Figure: a) show significant fluctuations.
• Fluctuations result from random switching of loads: automatic
or by occupants.
• Automatic switches: AC, central heating, refrigerators, water
heaters.

Fig. a: Demand of a house on 15‐minute basis. 24

ECpE Department
4. Load Characteristics
4.3 Time Aggregation
• Occupant-controlled switches: lighting, cooking appliances,
washer, dryer, computers, entertainment loads.
• Loads may vary in power consumption even when on.
• Load demand smoothens with longer time periods, such as the
load demand characteristics on 30-minute basis (Figure: b).

Fig. b: Demand of a house on 30‐minute basis. 25

ECpE Department
4. Load Characteristics
4.3 Time Aggregation
• Load demand characteristics on 1-hour basis (Figure: c).
• Time interval larger than 1 hour is not used to determine the
load demand characteristics of the for planning purposes.
• Notice that time aggregation reduces peak load demand value.

Fig. c: Demand of a house on 15‐minute basis. 26

ECpE Department
 Diversity Factor
By definition, diversity factor (DF) is the ratio of the maximum
noncoincident demand of a group of customers to the maximum
diversified demand of the group. With reference to the transformer
serving four customers, the DF for the four customers would be:
Max_noncoincident_demand = 6.18 + 6.82 + 4.94 + 7.05 = 24.98kW
Max._noncoincident_demand 24.98
Diversity_factor = = = 1.5458
Max._diversified_demand 16.16

27

ECpE Department
Diversity Factor

Fig.1 24-hour demand curve for customer#1 Fig.2 24-hour demand curve for customer#2

Fig.3 24-hour demand curve for customer#3 Fig.4 24-hour demand curve for customer#4

28

ECpE Department
Diversity Factor
Table 1 is an example of the DFs for the number of customers
ranging from 1 up to 70. The table was developed from a different
database than the four customers that have been discussed
previously.
Table 1 Diversity Factors

29

ECpE Department
 Diversity Factor
A graph of the DFs is shown in Fig.5.

Fig.5 Diversity Factor

Note that, in Table 1 and Fig. 5, the value of the DF basically leveled out when the number of
customers reached 70. This is an important observation because it means, at least for the
system from which these DFs were determined, that the DF will remain constant at 3.20 from
70 customers and above. In other words as viewed from the substation, the maximum
diversified demand of a feeder can be predicted by computing the total noncoincident
maximum demand of all of the customers served by the feeder and dividing by 3.2. 30

ECpE Department
Application of Diversity Factors
The definition of the DF is the ratio of the maximum noncoincident
demand to the maximum diversified demand. DFs are shown in
Table 2. When such a table is available, then it is possible to
determine the maximum diversified demand of a group of
customers such as those served by a distribution transformer. That
is, the maximum diversified demand can be computed by:
Max._noncoincident_demand
Diversity_factor =
Max._diversified_demand

Max. _noncoincident_demand
Max. _diversified_demand =
𝐷𝐷𝐷𝐷𝑛𝑛

31

ECpE Department
But how could we know peak demand?
(w/o demand mtr)
Many times the maximum demand of individual customers will be
known either from metering or from knowledge of the energy (kWh)
consumed by the customer. Some utility companies will perform a load
survey of similar customers in order to determine the relationship
between the energy consumption in kWh and the maximum kW
demand. Such a load survey requires the installation of a demand
meter at each customer's location.

32

ECpE Department
But how could we know peak demand?
(w/o demand mtr)
Relate energy consumption to peak demand through study:

• Similar type of customers (residential)

• Metering on each customer for study

The plot of points for 15

customers along with the resulting
equation derived
Max. _kW_demand
= 0.1058 + 0.005014 ∗ 𝑘𝑘𝑘𝑘𝑘

Fig.6 kW demand versus kWh for

residential customers 33

ECpE Department
4. Load Characteristics
4.4 Diversity and Coincidence
• Temporal aggregation reduces peak demand by averaging
over a selected time duration.
• Aggregating demands of multiple houses smoothens out load
demand characteristics (Figures (a-f) shown in the next slide).
• Load diversity (LD) in utility terminology: Houses have
different load on/off timings.
• Peak demand of a group of houses at any time is lower than
an individual house's peak.
• For conductor sizing, use peak value of combined demand on
hottest (or coldest) day.

34

ECpE Department
4. Load Characteristics
4.4 Diversity and Coincidence
• Transformer sizing considers how many houses are served,
using typical load demand characteristics.
• Coincidence factor adjusts aggregated and maximum loads of
houses for coincident peak demand. Mathematically,
𝐷𝐷𝑚𝑚𝑚𝑚
𝐶𝐶 = 𝑛𝑛
∑𝑖𝑖=1 𝐷𝐷𝑚𝑚𝑚𝑚
where, 𝐷𝐷𝑚𝑚𝑚𝑚 is the peak demand of the ith load in a group of n
loads.
and, 𝐷𝐷𝑚𝑚𝑚𝑚 is the Coincident peak demand of the group.

35

ECpE Department
 4. Load Characteristics
4.4 Diversity and Coincidence

Fig. a: Demand of a house on 30‐minute basis.. Fig. b: Demand of a house on one‐hour basis. Fig. c: Fifteen‐minute average load of two houses.

Fig. d: Fifteen‐minute average load of 5 houses. Fig. e: Fifteen‐minute average load of 10 houses. Fig. f: Fifteen‐minute average load of 20 houses.

36

ECpE Department
 4. Load Characteristics
4.4 Diversity and Coincidence
• Coincidence factor (CF) ranges
from 0 to 1.
• CF decreases as the number of
houses in a group increases.
• Common range for CF: 0.3–0.6
for larger groups.
• CF graph (similar to Figure on the
right) shows CF variation based Fig. Coincidence factor as a function of number of houses.
on the number of houses (n).
• CF decreases quickly for lower n
values, then saturates.
• Saturation point typically
reached around n = 15–20.
37

ECpE Department
4. Load Characteristics
4.4 Diversity and Coincidence
• The inverse of the coincidence factor is defined as the
diversity factor (D).
• Diversity Factor gives an indication of the separation in
individual house's peak demands in comparison to the peak
demand of the group. Mathematically,
1
𝐷𝐷 =
𝐶𝐶
• Further, the difference between the sum of the individual
peak demands and the coincident peak demand of the group
is defined as LD,
𝐿𝐿𝐿𝐿 = ∑𝑛𝑛𝐼𝐼=1 𝐷𝐷𝑚𝑚𝑚𝑚 − 𝐷𝐷𝑚𝑚𝑚𝑚
38

ECpE Department
4. Load Characteristics
4.4 Diversity and Coincidence
• The concept of coincidence can also be applied to loads of
different types. For example, if a feeder has to serve
residential, commercial, and industrial loads with given load
demand characteristics, these characteristics can be used to
find the coincident peak demand. This and other factors are
then used to find the size of the feeder that would be
adequate to serve the combined load.

39

ECpE Department
4. Load Characteristics
4.5 Demand Factor
• The ratio of the maximum load demand over a period to the
connected load for any load or class of load is defined as the
demand factor (DemandF).
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷
𝐷𝐷𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 =
𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿
• Connected load: Sum of rated values of all loads in a house.
• Not all loads used simultaneously; only a portion used at
once.

40

ECpE Department
4. Load Characteristics
4.5 Demand Factor
• Example with light bulbs: Total ratings ~2 kW, simultaneous
operation ~600 W.
• Diversity Factor for lighting load: 0.3 (0.6/2 =0.3).
• Example with air conditioner: Rated 6 kW, runs 30 min on
hottest day. Average hourly demand: 3 kW; Air conditioner’s
DemandF: 0.5 (3/6).
• DemandF applies to individual loads or a group of loads.

41

ECpE Department
4. Load Characteristics
4.6 Load Duration Curve
• Load in a distribution system changes over time. Examination
of hourly loads for short durations (e.g., a day) provides
highest, lowest loads, their occurrence times, and load
fluctuation pattern.
• For longer durations (e.g., a year), examining hourly load for
each day is a tedious task.
• Load duration curve simplifies this by graphing load vs. time
for which the load is higher than the specified value.

42

ECpE Department
4. Load Characteristics
4.6 Load Duration Curve
• We can easily obtain the two extreme ends of the load
duration curve by looking at the highest and the lowest
values of the load over a period of time.
• Load is never higher than the peak load (Lp), so the load > Lp
for 0% of the time.
• Load is always ≥ lowest load (Ll), so the load ≥ Ll for 100% of
the time.
• Intermediate points obtained by quantizing load in equal
steps.
• Steps' size is crucial: Too large provides few data points; too
small provides too many data points.
43

ECpE Department
4. Load Characteristics
4.6 Load Duration Curve
• Load duration curve effectively
summarizes load behavior over an
extended period.
• For example: In figure (a), we see
that Lp = 17 MW and Ll = 4 MW
• Illustration using 2 MW steps for
load duration curve.
• Count hours load is higher than Fig. a: Hourly load at a substation.
selected load (e.g., >10 MW for 10
h).

44

ECpE Department
4. Load Characteristics
4.6 Load Duration Curve
Table: Data for load duration curve obtained from Fig. a

45

ECpE Department
 4. Load Characteristics
4.6 Load Duration Curve
• From the table, we can get the load
duration curve as shown in figure
(b).
• Load duration curves, especially for a
Fig. a: Hourly load at a substation.
year are:
o Vital for planning studies.
o Used to plan required system
equipment.
o Used to estimate losses.
• Load duration curves provide
valuable insights for effective system
planning and resource allocation. Fig. b: Load Duration Curve corresponding
to the Load Characteristic.
46

ECpE Department
4. Load Characteristics
4.7 Load Factor
• Load Factor: Relates load changes over a time period to the
peak load.
• Measures how closely load aligns with peak load.
• High load factor: Load consistently near peak load for a
significant portion of the time.
• Mathematically, it is the average load over a period of time
divided by the peak load within that time duration,
𝑁𝑁 𝐿𝐿
∑𝑖𝑖=0 𝑖𝑖
𝐿𝐿𝑎𝑎 𝑁𝑁
𝐿𝐿𝐿𝐿 = =
𝐿𝐿𝑝𝑝 𝐿𝐿𝑝𝑝
where, 𝐿𝐿𝑖𝑖 is the hourly load at hour i, N is the total number of
hours in the time period under consideration, and 𝐿𝐿𝑖𝑖 is the peak
load in that time period. 47

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• Loss Factor (LsF): Ratio of average
hourly losses to peak hourly losses in
the system.
• Very useful for calculating total losses
over a period if losses at the peak load Fig. a: Discrete load duration curve
are known.
• For load characteristics in which peak
load exists for a very short duration,
an approximate value for the LsF: can
be obtained by taking the square of
the load factor.
Fig. b: Discrete loss duration curve.
48

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• Consider a discrete version of load
duration curve as shown in Fig. a,
where Lp is the peak load, and L1, L2,
L3 are the other load levels for
specified durations. The total time
Fig. a: Discrete load duration curve
duration is T hours.
• Resistive losses (predominant in
distribution systems) proportional to
square of current.
• Assuming fixed voltage, losses
proportional to square of load.
• Loss characteristics for the same
Fig. b: Discrete loss duration curve.
duration (Fig.b): 49

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• In figure b, the notations are defined as,
o Lsp: Average hourly losses at peak load.
o Ls1, Ls2, Ls3: Losses at other load levels.
• Loss factor and loss characteristics provide
insights into distribution system losses and Fig. a: Discrete load duration curve
efficiency.
• From fig. a, we get,
𝐿𝐿𝑝𝑝 𝑡𝑡1 +𝐿𝐿3 𝑡𝑡2 −𝑡𝑡1 +𝐿𝐿2 𝑡𝑡3 −𝑡𝑡2 +𝐿𝐿1 𝑇𝑇−𝑡𝑡3
LF =
𝑇𝑇𝐿𝐿𝑝𝑝

• Similarly,
𝐿𝐿𝑠𝑠𝑝𝑝 𝑡𝑡1 +𝐿𝐿𝑠𝑠3 𝑡𝑡2 −𝑡𝑡1 +𝐿𝐿𝑠𝑠2 𝑡𝑡3 −𝑡𝑡2 +𝐿𝐿𝑠𝑠1 𝑇𝑇−𝑡𝑡3
𝐿𝐿𝐿𝐿𝐿𝐿 =
𝑇𝑇𝑇𝑇𝑠𝑠𝑝𝑝
Fig. b: Discrete loss duration curve.
50

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• For a special case where 𝐿𝐿1 , 𝐿𝐿2 , and 𝐿𝐿3 are equal, we get
𝑡𝑡1 𝐿𝐿1 𝑇𝑇 − 𝑡𝑡1
LF = +
𝑇𝑇 𝐿𝐿𝑝𝑝 𝑇𝑇
and
𝑡𝑡1 𝐿𝐿𝑠𝑠1 𝑇𝑇 − 𝑡𝑡1
LsF = +
𝑇𝑇 𝐿𝐿𝑠𝑠𝑝𝑝 𝑇𝑇

51

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• Now, if we consider the peak load of a very short duration or
𝑡𝑡1 ≪ 𝑇𝑇,
𝑡𝑡1 𝑇𝑇 − 𝑡𝑡1
→ 0 and →1
𝑇𝑇 𝑇𝑇
• Hence,
𝐿𝐿1 𝐿𝐿𝑠𝑠1
LF = and LsF = .
𝐿𝐿𝐿𝐿 𝐿𝐿𝑠𝑠𝑝𝑝
2 2
• With 𝐿𝐿𝑠𝑠1 = 𝑘𝑘 𝐿𝐿1 and 𝐿𝐿𝑠𝑠𝑝𝑝 = 𝑘𝑘 𝐿𝐿𝑝𝑝 , we get

𝐋𝐋𝐋𝐋𝐋𝐋 = (𝐋𝐋𝐋𝐋)𝟐𝟐

52

ECpE Department
4. Load Characteristics
4.8 Loss Factor
• Approximation improves as time duration (T) increases, e.g.,
for a year.
• Peak load usually exists for a short period (1-2 hours).
• Square relationship between load factor and LsF is an
extreme simplification.
• Reality: off-peak loads vary.
• Despite simplifications, this relationship is acceptable for
planning purposes.
• Other approximations are also available in literature.

53

ECpE Department
5. Design Criteria and Standards
• Planning engineer's role: Minimize distribution system
equipment cost, ensure design criteria and standards are met.
• Key issues in distribution system planning and design:
o Voltage and service quality.
o Equipment loading.
o Safety.
• Design criteria and standards set by:
o Equipment manufacturers.
o Standards organizations.
o Utilities (often more stringent).

54

ECpE Department
5. Design Criteria and Standards
• Equipment loading:
o Manufacturers specify limits to prevent equipment
damage.
• Safety:
o Involves insulation, clearance between equipment.
• Voltage and service quality:
o Directly affect customers' experience.

55

ECpE Department
5.1 Voltage Standards
• ANSI C84.1 standard specifies service voltage for U.S. utilities:
o Range A: 114–126 V (on a 120-V nominal), preferred
range.
o Range B (emergency): 110–127 V.
o Building wiring drop allowed: 4 V.
o Utilization voltage for customers: 110–126 V under normal
operation.
• Equipment manufacturers design within ANSI range;
operation outside this range shortens equipment life.
• Utilities aim to keep service voltage ≥ 120 V for a larger safety
margin.
56

ECpE Department
5.1 Voltage Standards
• Providing ANSI-specified voltage to customers is a challenge
due to:
o Voltage drop over conductors from substation to
customers.
o Voltage drop fluctuates with load changes.
o Proximity to substation affects voltage drop.
• Utilities use tools like load tap changers, line regulators,
capacitors to
o Maintain service voltage within ANSI range;
o Ensure proper system design and voltage control.

57

ECpE Department
5.2 Conservation Voltage Reduction
• Lowering voltage reduces real and reactive power demand,
energy consumption.
• Utilities can operate within lower ANSI range (e.g., 114 V) for
energy savings.
• Utilities conducted experiments (1970s-1980s) on voltage
reduction's effects.
• Some success in energy reduction, peak power control.
• Voltage control for conservation (CVR) and peak power
reduction practiced by a few utilities.

58

ECpE Department
5.2 Conservation Voltage Reduction
• Challenges include load reduction uncertainty, voltage-related
complaints.
• Recent interest grows due to:
o National energy conservation focus.
o Environmental concerns and carbon emission reduction
efforts.
o Positive impact on utility's "green image."
• Studies show real load-to-voltage sensitivities vary (0.4 - 2.5),
average: 0.8 - 1.0.
• Reactive load-to-voltage sensitivity range: 3 - 5, average around
4.
• Energy sensitivity over 24 hours: 0.6 - 1.237. 59

ECpE Department
 6. Distribution System Design
6.1 Substation Design
• Distribution substation marks
transition from transmission/sub-
transmission to distribution.
• Power enters at 34.5 - 230 kV,
reduced to 2.4 - 34.5 kV in
distribution substation.
• Distribution voltage determined by
load density, power delivery
distance.
• Substations often positioned for Fig: Typical Layout of a Substation
(NC, normally closed and NO, normally open)
uniform service area coverage,
except rural settings.
60

ECpE Department
 6. Distribution System Design
6.1 Substation Design
• Example substation layout shown in
the figure.
• Three 115-kV incoming lines
connected to 115-kV bus.
• 115-kV bus normally operates as
contiguous bus with 4 NC switches.
• Different configurations possible in
emergencies.
• Each incoming line equipped with
Fig: Typical Layout of a Substation
circuit breaker. (NC, normally closed and NO, normally open)

61

ECpE Department
 6. Distribution System Design
6.1 Substation Design
• Various bus arrangements (main &
transfer bus, ring bus, breaker-and-a-
half bus schemes) for reliability
enhancement. Complex arrangements
more expensive, can complicate
protection systems.
• Distribution substations can have 1 to 3
transformers.
Fig: Typical Layout of a Substation
(NC, normally closed and NO, normally open)

62

ECpE Department
 6. Distribution System Design
6.1 Substation Design
• Example substation (Figure) has two
12/16/20 MVA transformers, 115kV-
Δ/12.47kV-Y-grounded.
• Equal capacity transformers preferred
based on substation's total load.
• Tie breaker on 12.47 kV bus normally
open to avoid parallel operation of
transformers.
Fig: Typical Layout of a Substation
• Transformers loaded to 50% of highest (NC, normally closed and NO, normally open)

rating during peak load for redundancy.

• If 3 transformers, loaded to 67% for
load sharing on failure.
63

ECpE Department
6. Distribution System Design
6.1 Substation Design
• Multiple transformers enhance
loading and reliability.
• Automation and switch operation
enable load transfer during failures.
• Distribution substation transformers
usually have Δ/Y configuration to
provide neutral on the low-voltage
side for distribution of single-phase
power to customers using one of
the phases and the neural. Fig: Typical Layout of a Substation
(NC, normally closed and NO, normally open)

64

ECpE Department
6. Distribution System Design
6.1 Substation Design
• The rule of keeping Y-connection on
the high-voltage side of the Δ/Y
transformers is overruled in this case
to make the neutral available on the
low-voltage side.
• 12.47 kV bus on low-voltage side split
into two segments with NO switch.
• Each segment has 1 to 4 feeders
based on load and transformer
capacity.
Fig: Typical Layout of a Substation
• Typical feeder designed for 4-8 MVA. (NC, normally closed and NO, normally open)
• Example substation (Figure) has three
feeders per transformer, a total of six
feeders. 65

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• Primary feeders are crucial for
distributing power from substations to
customers.
• Distribution systems typically have a
radial structure, except in densely
populated areas.
• Radial structures have power flowing
in one direction, aiding simplicity.
• Radial systems are cost-effective,
easier to operate, and simpler to
protect.
66

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• Distribution systems mainly use radial
configuration with tree-like feeder
arrangement.
• Substation is the root of the feeder
tree, with a main three-phase primary
feeder or trunk.
• Primary feeder may split into sub-
feeders or remain as one throughout
service area.
• Single and multi-feeder layouts are
depicted in Figures on the right,
respectively.
67

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• Load on primary feeders decreases as distance from
substation increases.
• Conductor sizes for primary feeders can be tapered based on
radial load distribution.
• Primary main feeders in distribution systems often have a
N.O. (Normally Open) switch or circuit breaker called the tie
switch at their ends.
• Tie switches are closed during permanent faults to restore
power to some customers on faulted feeders.

68

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• The feeder on the other side of the tie switch could be
connected to the same transformer, or to a different
transformer from the same substation, or to a transformer
from a different substation, each of them provides a different
level of reliability and has different costs.
• Feeding arrangements impact backup possibilities in case of
transformer or feeder failures.
• Primary lateral feeders are single-phase, phase and neutral
conductors branching off primary mains.

69

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• Connected to the primary
main feeders are the primary
lateral feeders, which are
usually single phase, with
phase and neutral
conductors.
• Primary lateral feeders carry
power from primary mains to
customers.

Fig. Feeders along with the roads in a

typical distribution system in the United States. 70

ECpE Department
6. Distribution System Design
6.2 Design of Primary Feeders
• Primary main feeders are
commonly along city streets,
but the lateral feeders are
mostly in the utility easement
behind houses in residential
subdivisions.
• Phase selection for lateral
feeders maintains rough
balance among the three
phases.
• Figure on the right illustrates
feeders and streets in a Fig. Overhead feeders along with the roads in a
typical US city. typical distribution system in the United States. 71

ECpE Department
 6. Distribution System Design
6.3 Design of Secondary Systems

Fig. Illustration of a secondary system for service to eight customers.

• The secondary system is the part of the distribution system closest to

customers.
• Components of the secondary system include distribution transformers,
secondary feeders, and service drops.
• Secondary system is single-phase for residential customers and three-phase
for large industrial and commercial customers. 72

ECpE Department
 6. Distribution System Design
6.3 Design of Secondary Systems
• Distribution transformers are small (5 to 100 kVA) with 120
V/240 V secondary.
• Transformer size at a location depends on the number and
class of customers it serves.
• Transformers typically serve four to eight medium-sized single-
family homes.
• If homes are very large, a transformer may serve only one or
two homes.
• A single transformer may serve several apartments in a
complex.

73

ECpE Department
 6. Distribution System Design
6.4 Underground Distribution Systems
• Overhead distribution systems are predominant worldwide due
to lower cost compared to underground systems.
• Underground systems are 5–10 times more expensive than
overhead systems.
• Densely populated areas and business districts often use
underground systems despite the higher cost.
• Some cities require underground systems in new housing
developments for aesthetic reasons.

74

ECpE Department
 6. Distribution System Design
6.4 Underground Distribution Systems
• Developers may choose underground distribution in high-cost
subdivisions even without mandates. In some cases,
underground systems have underground laterals and secondary
systems, but the primary main remains overhead.
• Overhead distribution systems are more vulnerable to failure
due to exposure to the environment.
• Overhead systems offer the advantage of easier fault location
and repairs.
• Underground systems have fewer faults, but locating and
repairing faults is challenging and time-consuming.

75

ECpE Department
 6. Distribution System Design
6.5 Rural vs. Urban Systems
• Distribution systems can be categorized into urban, semi-
urban, and rural based on load density.
• Urban systems have high load density, smaller substation
spacing, larger transformers, and bigger feeders.
• Urban systems are mostly thermally limited. The total load that
a substation can supply is based on the thermal loading
capacity of transformers and feeders. Voltage drop is usually
not a problem.

76

ECpE Department
 6. Distribution System Design
6.5 Rural vs. Urban Systems
• Rural systems have low load density, larger substation spacing,
smaller transformers, and smaller feeders.
• Semi-urban systems fall between urban and rural systems in
terms of characteristics.
• Urban systems often have multiple power transformers and
sufficient NO tie points, leading to higher reliability.
• Rural systems are usually voltage drop-limited due to long
feeder lengths.
• Semi-urban systems may experience a combination of thermal
and voltage drop limitations based on design choices.

77

ECpE Department
 7. Cold Load Pickup (CLPU)
7.1 CLPU Fundamentals
• Categories of Residential Feeder Loads:
o Thermostatically controlled devices (e.g., air conditioners,
heaters, refrigerators).
o Manually controlled loads (operated by occupants based on
needs).
• Contribution of Thermostatically Controlled Devices:
o Largest share of total load in typical houses.
o Contribution depends occupants' lifestyle.

78

ECpE Department
 7. Cold Load Pickup (CLPU)
7.1 CLPU Fundamentals
• Diversity Among Loads in Groups of Houses:
o Aggregate load lower than connected load under normal
conditions.
o Impact of extended interruption: Immediate activation of
thermostatically controlled devices upon power
restoration.
o Higher post-restoration load from manually controlled
devices due to user demand.
• Cold Load Pickup (CLPU) Phenomenon:
o Experienced after extended power interruption.

79

ECpE Department
 7. Cold Load Pickup (CLPU)
7.1 CLPU Fundamentals
o Four phases: inrush, motor starting, motor running, and
enduring.
o First three phases <15 s, current may reach 5–15 times of
pre-outage current.
o Enduring phase until normal load diversity is restored,
lasting hours.
o Influenced by outage time, temperature, device type, and
ratings.
• Historical Context of CLPU:
o Originated in 1940s due to high inrush currents hindering
reenergization.
o Solutions include inverse characteristic relays and
distribution system sectionalization. 80

ECpE Department
 7. Cold Load Pickup (CLPU)
7.1 CLPU Fundamentals
• Increasing Presence of Thermostatically Controlled Devices:
o Growing penetration in distribution systems.
o CLPU restoration issues preceding serious overloading
problems.
• Significance of Sustained Load After Restoration:
o Important consideration for loading limitations of
distribution equipment.

81

ECpE Department
 7. Cold Load Pickup (CLPU)
7.2 CLPU Models
• Main idea: develop a
physically‐based load model
for individual loads. Then
use aggregation of these
loads to find the total Fig. Load upon restoration following a long outage. The dots
show the average load for 15 min prior to that time.
demand.
• Aggregated load behavior
can be determined based
on these models using
numerical techniques by
solving partial differential
equations or by Monte Fig. Delayed exponential model for cold load pickup. SU is the
undiversified load and SD is the diversified load.
Carlo simulation. 82

ECpE Department
 7. Cold Load Pickup (CLPU)
7.2 CLPU Models
• An example after restoration
around midnight during winter is
given. The load prior to
interruption was 749 kW.
o The load in the first 15‐min
Fig. Load upon restoration following a long outage. The dots
period is 1451 kW, around two show the average load for 15 min prior to that time.
times the load prior to
interruption.
o The load returns to normal in
the third interval.
• Simulations and experimental data
suggest that aggregate load in
distribution system during CLPU
Fig. Delayed exponential model for cold load pickup. SU is the
can be represented by a delayed undiversified load and SD is the diversified load.
exponential model. 83

ECpE Department
 7. Cold Load Pickup (CLPU)
7.3 Impacts of CLPU
• Impact on Operation and Design:
o CLPU affects operation and design of distribution systems.
o Impacts vary based on system type.
• Overloading and Excessive Voltage Drops:
o CLPU leads to higher load than normal.
o Overloading of transformers and conductors.
o Excessive voltage drops on feeders.
o Urban systems: Overloading of transformers is critical.
o Rural systems: Excessive voltage drops are critical.

84

ECpE Department
 7. Cold Load Pickup (CLPU)
7.3 Impacts of CLPU
• Step-wise Restoration Approach:
o Full load restoration may not be possible in one step.
o Remote sectionalizing switches on main feeder for system
division.
o Sections restored incrementally.
• Load Behavior and Restoration Sequence:
o Load dynamics of sections are crucial.
o Restoration sequence impacts procedure.
o Objectives: Meeting restoration goals, limiting transformer
and feeder loading, preventing voltage-drop violations.
85

ECpE Department
 7. Cold Load Pickup (CLPU)
7.3 Impacts of CLPU
• Restoration Objectives:
o Minimize customer interruption duration.
o Directly enhance system reliability.
o Shorter interruptions lead to higher system reliability.

86

ECpE Department
 7. Cold Load Pickup (CLPU)
7.4 Operating Limits
ANSI/IEEE PC57.91‐1994 Standard:
• Transformer Load and Loss of Life:
o Transformer load can exceed rated load in emergencies.
o Transformer loss of life should not exceed 4%.
• Temperature Limits for Transformer:
o Maximum top-oil temperature: 110 °C.
o Maximum hottest spot winding temperature: 180 °C.
o Maximum short time loading: Up to two times the
maximum normal rating.
• Short Time Emergency Loading of Conductors:
87

ECpE Department
 7. Cold Load Pickup (CLPU)
7.4 Operating Limits
ANSI/IEEE PC57.91‐1994 Standard:
• 133% of normal full load. Voltage Limits and Voltage Drop:
o Supply voltage to customers within ANSI C84.1 limits.
o Voltage drop breakdown:
 Lateral under normal operation: ~2.5%.
 Lateral under emergency operation: 3.54%.
• Total Permissible Voltage Drop:
o Upper end of zone at substation.
o Main feeders:
 Normal operation: Up to 7.5%.
 Emergencies: Up to 10.62%. 88

ECpE Department
 7. Cold Load Pickup (CLPU)
7.4 Operating Limits
ANSI/IEEE PC57.91‐1994 Standard:
• Voltage Drop Limit During CLPU Restoration:
o Voltage drop should not exceed 10.62%.
Asset Management Overview:
o Optimal utilization of assets in planning and operation.
o Balancing decisions for replacement and updating of equipment.
o Considering age and current operating status.
o Accounting for risk associated with actions.
• Planning Phase:
o Decisions on equipment replacement or updates.
o Balancing timing to avoid disastrous consequences.
o Considering risk and cost-effectiveness.
89

ECpE Department
8. Asset Management
• Operation Phase:
o Real-time decisions on equipment loading.
o Considering equipment conditions and status.
o Balancing risk of overloading and curtailed service.

90

ECpE Department
8. Asset Management
• Balancing Risk and Service:
o Preventing equipment failure from sustained overloading.
o Ensuring uninterrupted service to customers.
o Risk assessment and mitigation.
• Challenges with Aging Infrastructure:
o Existing distribution and transmission infrastructure is old.
o Some equipment exceeds 40 years in age.
o Complete replacement is cost-prohibitive.
o A planned replacement strategy is required.

91

ECpE Department
Thank You!

92

ECpE Department