PNNL-SA-41618 1
Simulating Price Responsive Distributed Resources
N. Lu Member IEEE, D. P. Chassin, Member IEEE, and S. E. Widergren, Sr. Member, IEEE
Given a large penetration of price responsive DER with its
associated market transactions, a new approach is also
proposed, based upon more holistic statistical mechanic
techniques. Such an approach holds promise in revealing the
fundamentals of complex system behavior by elevating
analysis from the overwhelming details to higher levels of
abstract behavior.
Abstract — Distributed energy resources (DER) include
distributed generation, storage, and responsive demand. The
integration of DER into the power system control framework is
part of the evolutionary advances that allow these resources to
actively participate in the energy balance equation. Price can
provide a powerful signal for independent decision-making in
distributed control strategies. To study the impact of price
responsive DER on the electric power system requires generation
and load models that can capture the dynamic coupling between
the energy market and the physical operation of the power
system in appropriate time frames. This paper presents modeling
approaches for simulating electricity market price responsive
DER, and introduces a statistical mechanics approach to
modeling the aggregated response of a transformed electric
system of pervasive, transacting DER.
II. DEMAND RESPONSE MODEL FOR RESIDENTIAL LOADS
As one example of price responsive DER, we investigated
the models for appliances calibrated against a database of
extensively monitored residential distribution feeders. The
simulation of the dynamic response of residential loads for use
in demand response programs requires models that can
interact with price, voltage, and frequency input signals.
The diversity of the appliances over time must be taken into
account in the load models. Factors of diversity can be
divided into two types: spatial and temporal. Spatial diversity
is caused by differences in household characteristics, such as
the size and type of the appliances. Temporal diversity is the
result of the random behavioral variations among different
households. Such variations may include occupancy
schedule, likelihood and frequencies of opening doors and
windows, thermostat settings, and other lifestyle-related
factors [1].
Physically based load modeling methodologies have been
widely used because they are able to predict the individual
load dynamic response and allow one to obtain the aggregated
response of these loads with reasonable accuracy [2].
Index Terms—demand response, load model, load synthesis,
thermostatically controlled appliance, system simulation,
distributed energy resources.
I. INTRODUCTION
s market-based approaches have dominated changes in
the operation of the bulk power system, initiatives have
also begun to involve distributed energy resources (DER)
in this process. Controllable load, distributed generation, and
storage at the finger tips of the system hold promise to
enhance cost effective operation and reliability. These
resources represent half of the energy balance equation that
traditionally exhibits uncontrolled, but generally predictable
behavior. Though individually small contributors, by using
appropriate signals or incentives the control of vast numbers
of DER can have significant operational impact.
A
Demand management programs have been implemented for
some time, but more recently, initiatives that provide price
incentives to DER operation have appeared. As these
programs become more prevalent, the need to understand their
impact on reliable system operation becomes important. The
following discussion reviews work done to model price
responsive load from the details of individual electric
appliances interacting at the distribution feeder level, to price
responsive feeder equivalent models appropriate for regional
system level studies. Inherent in the price responsive DER
modeling is the simulation of market-based systems that allow
these resources to interact with bulk energy markets to drive
decision-making in economically efficient directions.
Fig. 1: A bottom-up approach
The load model we used for price responsive residential
load simulation takes a bottom-up approach in which detailed
physically-based models of each type of appliance are
developed, as shown in Fig.1. An aggregated load model
using a state queuing model (SQ) approach [4] was developed
based on modeling group appliance behaviors to improve the
computation speed. The SQ model serves as a simplified
reduced order feeder equivalent model to the detailed
household-based load model, as shown in Fig. 2.
This work is supported by the Pacific Northwest National Laboratory
operated for the U.S. Department of Energy by Battelle under Contract DE-
AC06-76RL01830.
N. Lu, D. P. Chassin, and S. E. Widergren are with the Energy Science and
Technology Division, Pacific Northwest National Laboratory, P.O. Box 999,
MSIN: K5-20, Richland, WA - 99352, USA (e-mail: ning.lu@pnl.gov ,
david.chassin@pnl.gov , steve.widergren@pnl.gov )
© 2004 IEEE. Reprinted from Proceedings of 2004 PES PSCE Meeting