Demand Side Management

1
 Generation Side Management
• The traditional power grids are top-down oriented and
only try to manage the system by controlling resources at
the generation side.
• The electric power must be generated, transmitted, and
consumed at the same time.
• Traditionally, only the generating units are adjusted to keep the
demand and supply balanced instantaneously.
• Variable renewables bring new challenges.
• Extra flexibility and more advanced control methods are
desired.
• Demand side management (DSM) provides system
operators an additional option to manage the grid.
2
 Demand Side Management
• Demand side management (DSM) program consists of a
group of strategies that are designed to encourage
consumers to modify their consumption level and pattern
through a variety of methods such as financial incentives.
• DSM can change and flatten the load profile.
• DSM encourages end-consumers to shift their
consumption from peak hours to off-peak hours.
• The objectives of DSM are to reduce cost, enhance grid
reliability, and facilitate the integration of renewable
energy sources.
• DSM helps defer the need for new sources of power and
network upgrade. 3
 Demand Side Management
Demand side management (DSM) programs:
• Energy efficiency
• Time of use (TOU)
• Demand response (DR)
• Spinning reserve (SR)

Benefits of DSM programs:

• Alleviate congestions
• Commit less generators
• Reduce cost
• Improve grid reliability
• Defer investments

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
4
on industrial informatics 7.3 (2011): 381-388.
 Energy Efficiency
• Energy efficiency is key to achieving a sustainable energy
system.
• The goal is to reduce the amount of electric power to
provide the same quality of services and products.
• Many customers have hidden energy waste problems:
• Dirty AC filters
• Compressed air leakages
• Low efficiency equipment
• Appliances that draw power even when it's not needed
• Simultaneous heating and cooling in the same building.
• Energy efficiency: efficient energy use.

5
 Energy Efficiency
• Energy efficiency is achieved with more efficient technologies or
processes.
• If a house has well located windows, well-sealed doors, basement slabs, and
thermal insulation of walls, then much less energy is used in heating and
cooling to maintain the same temperature.
• Energy-efficient bulbs (e.g. LED bulbs) use less electricity than traditional
light bulbs (e.g. incandescent bulbs).

An energy-efficient 5W LED bulb

saves more than 80% energy
than a 40W traditional bulb but
provides the same level of light.

CFL: compact fluorescent lamp 6

http://www.sunlinkpv.com/products/accessories/led_lights.html LED: light-emitting diodes
 Time of Use
• Time of use (TOU) tariff charges a high price over certain
periods of time (peak hours) but offer a low price over
other periods.
• TOU prices vary for different hours, days, and seasons.
• TOU encourages customers to adjust their energy
consumption habits.
• The goal of TOU is to flatten the load profile by shifting
some peak loads.

Time-based rate programs

7
 Demand Response
• Demand response (DR) is a change in the electricity
consumption to balance supply and demand.

• DR can lower the electricity price in the wholesale

markets and improve system reliability.

• It provides consumers an extra electricity plan/option:

they would get paid when asked to reduce or shift their
electricity usage.

8
 Demand Response
• Market DR: real-time pricing, price signals and incentives.
• It relies on certain markets where prices are formed and
products are traded.
• When the price of real-time markets (RTM) is high and RTM
pricing is forwarded to consumers, they may reduce their
electricity usage in a timely manner.
• In some places, the RTM pricing only exists in wholesale energy
markets; end-use consumers have a fixed electricity plan.

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
9
on industrial informatics 7.3 (2011): 381-388.
 Demand Response
• Physical DR: grid management and emergency signals.
• DR contracts will make consumer reduce electricity usage.
• It sends out signals for mandate demand reduction when the
grid is in an insecure situation or some components are facing
outage or reduced performance.
• Direct load control (DLR).

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
10
on industrial informatics 7.3 (2011): 381-388.
 Spinning Reserves by Loads
• Spinning reserve is for addressing unexpected outages.
Typically, spinning reserve only denotes the available
generation of online units in a given time period.
• Loads can act as “virtual” spinning reserve if they
correlate their consumption with the grid state and
system event.
• This can happen in an autonomous way or in a
coordinated way.

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
11
on industrial informatics 7.3 (2011): 381-388.
 Demand Side Management
Impact of improved energy efficiency versus demand response

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
12
on industrial informatics 7.3 (2011): 381-388.
 Demand Response
Incentive-based DR:
• Direct load control (DLC):
• grid operators have full access to some customers’ appliances.
• Interruptible/curtailable rates:
• customers get special contracts with utilities.
• Emergency demand response programs:
• Voluntary/involuntary response to emergency signals.
• Capacity market programs:
• customers guarantee to pitch in when the grid is in need.
• Demand bidding programs:
• customers can bid for curtailing at attractive prices.

Ref: Palensky, Peter, and Dietmar Dietrich. "Demand side management: Demand response, intelligent energy systems, and smart loads." IEEE transactions
13
on industrial informatics 7.3 (2011): 381-388.
 Demand Response
FERC’s definition of Demand Response:
• Changes in electric usage by demand-side resources
from their normal consumption patterns in response to
changes in the price of electricity over time, or to
incentive payments designed to induce lower electricity
use at times of high wholesale market prices or when
system reliability is jeopardized.

The key of DR:

• intentional changes in consumption patterns of
electricity.
FERC: Federal Energy Regulatory Commission
14
https://www.ferc.gov/industries/electric/indus-act/demand-response/dem-res-adv-metering.asp
 Demand Response
Customer Enrollment in Retail Demand Response Programs, by Region

2019 Assessment of Demand Response and Advanced Metering, Staff Report, Federal Energy Regulatory Commission, December 2019. 15
https://www.ferc.gov/legal/staff-reports/2019/DR-AM-Report2019.pdf
 Demand Response

Comfort v.s. Incentive

• Pay regular electricity price • Pay reduced electricity price

• Sufficient power supply • Limited power supply (occasionally)

https://carolinacomfortsc.com/the-top-2-ways-to-keep-your-home-hvac-in-the-zone/ 16
https://imgflip.com/memetemplate/105468009/Sweating-Guy
 Thermostatically Controlled Loads
Flexible thermostatically controlled loads (TCLs)
• Heating, ventilation, and air-conditioning (HVAC) systems
• Heat pumps
• Electric water heaters (w. tanks)
• Refrigerators

TCLs are flexible demand resources

• Possess a brief period where their power utilization can be altered
without any significant impact on customer comfort level.
• Can be aggregated to provide demand response.
• Account for a significant portion of electricity consumption.
• Potential to support power grid, provide grid ancillary services.
17
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 HVAC
TCLs: bang-bang control
• From On to Off versus From Off to On.
• Can be frequently switched on/off.
• Simple, but effective to offer demand response service.
• E.g., HVAC.

HVAC
• The heat transfer process is critical to relate it to the electricity
consumption of the unit.

18
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 Heat Transfer Process of HVAC
Equivalent Thermal Parameter (ETP) model:
• A thermal dynamics model.
• This model is appropriate to be used for residential and small
commercial buildings.

Ambient temperature (°F or °C)

Heat rate for HVAC unit (Btu/°F)
Heat exchange factor for house objects (Btu/°F)
Air heat capacity (Btu/°F or J/°C)
Mass heat capacity (Btu/°F or J/°C)
Thermal resistance of air (°F/Btu*hr or °C/ W)
Thermal resistance of mass (°F/Btu*hr or °C/ W)
Air temperature inside the house (°F or °C)
Mass temperature inside the house (°F or °C)

Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022. 19
S. Katipamula and N. Lu, “Evaluation of residential HVAC control strategies for demand response programs,” ASHRAE Trans, vol. 1, no. 12, pp. 1–12, 2006
 Demand Response Support Time
Demand response support time (DRST)
• A metric to quantify HVAC-enabled demand response.
• Indicating how long an HVAC will be available for demand response
at a given time.
• Be used to identify the set of HVACs that are qualified to provide
DR without compromising consumers’ comfort.
• The HVAC thermodynamic model can be simulated to determine
the DRST for each HVAC.

20
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 Demand Response Support Time
Illustration of the Metric DRST:

• Assume a hot summer day.

• The HVAC is at “on” state during time periods and .
• The HVAC is at “off” state during and .
• With a DR signal is received at , the HVAC can provide DR until it reaches the upper
bound temperature at DRST is the time period between and .
21
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 TM-DRST
Theoretically Maximum DRST (TM-DRST)
• Provide insights about how long each HVAC can support when they
receive DR signals at extreme low points (when cooling) or extreme
high points (when heating).
• When HVAC receives the DR signal right before those extreme low
points such as , the corresponding DRST is the TM-DRST.
• Like DRST, TM-DRST is also affected by several factors including the
ambient temperature.

22
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 TM-DRST
TM-DRST of 100 small HVACs with reasonable random
parameters:

TM-DRST statistics of these 100 HVACs (minutes)

23
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 DRST-HDRQ
DRST-based HVAC-DR quantification (DRST-HDRQ) method

Use HVAC-DR as contingency

response reserve.

denotes the pre-specified

contingency response duration.
denotes the power capacity of
HVAC h (kW).

24
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 Networked Microgrid with HVAC-DR
Microgrid day-ahead scheduling

Scenario 2:
Economic Use HVAC-DR to
Benefits meet the reserve
requirement.

Average forced load shed for scenario 1 (without HVAC as reserve in microgrid day-
ahead scheduling) when tie-line is outaged from hour 21 to hour 22 in real-time.

Reliability
Benefits

25
Ref: Praveen Dhanasekar, Cunzhi Zhao, and Xingpeng Li. "Quantitative Analysis of Demand Response Using Thermostatically Controlled Loads." ISGT 2022.
 Home Area Network
Individual home load (energy) management with
price-based demand response.

The system architecture of a home

area network consists of
• sensors (a smart meter and smart
plugs to connect legacy appliances).
• smart devices (smart appliances,
electricity storage, electric vehicles).
• and a computing unit (the smart
home controller).

Ref: Di Giorgio, Alessandro, and Francesco Liberati. "Near real time load shifting control for residential electricity prosumers under designed and market 26
indexed pricing models." Applied Energy 128 (2014): 119-132.
 Home Area Network

Home load management (HLM)

• Monitor, control, and coordinate the
operation and energy consumption of
individual responsive appliances.

Ref: Safdarian, Amir, Mahmud Fotuhi-Firuzabad, and Matti Lehtonen. "A distributed algorithm for managing residential demand response in smart 27
grids." IEEE Transactions on Industrial Informatics 10.4 (2014): 2385-2393.
 Flexible Loads
Different flexible loads:
• Deferrable Loads: washer, dryer, dishwasher.
• Lighting Loads: can be dimmed.
• EV Loads: essentially batteries.
• Thermostatically Controlled Loads: on/off control.

Different flexibilities:
• Deferrable.
• Interruptible.
• Variable Power.

Ref: Raksha Ramakrishna, Nurullah Karakoc, Kari Hreinsson, Anna Scaglione. "Federating Solar, Storage and Communications in the Electric Grid and 28
Internet of things." ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2020.
 Dispatchable Demand Response
Category of DR:
• Dispatchable demand response
• Controllable load: capable of adjusting consumption
per dispatch signals sent by the utility/ISO.
• Non-dispatchable demand response
from the perspective of grid operator.

29
 Wholesale Energy Market (real-time price of ERCOT):
• Normal price is ~10-40$/MWh (or ~0.01-0.04$/kWh)
• Cap limit is $9,000 MWh (or 9$/kWh)
Retail Electricity Market: With demand side management,
• regular electric plan price is ~0.1$/kWh. we may avoid price spikes/peaks.

Aug 13, 2019

http://www.ercot.com/content/cdr/contours/rtmLmp.html 30
www.ercot.com