IEEE Power & Energy Magazine - November/December 2014 - 84

figure 5. Interstate and intraprovincial pipelines serving
the study region.

represent production regions and supply basins, pipeline
zones, interconnects, storage facilities, delivery points, and
either specific large customers or groupings of smaller customers. arcs represent gas transactions and flows, either spatial or
temporal. the arcs have appropriate constraints that represent
the actual pipeline flow capabilities between adjoining nodes.
gPcM uses partial-equilibrium economics to reach a solution in which supply and demand are balanced at each node
at the lowest possible cost. it is a partial-equilibrium model
because it singles out the natural gas industry rather than
performing economic trade-offs and optimizations for the
economy as a whole. Key inputs to gPcM include pipeline
and storage tariffs, production basin supply curves, pipeline
capacities and interconnections, Lng terminal storage and
daily vaporization capacities, and demand curves by sector
and consumption area. Model outputs include the flow through
each pipeline segment represented in the model and the price
at each node. While gPcM is generally used as a monthly
model encompassing average monthly price and quantity variables, given the study objectives set forth in targets 2, 3, and
4, Lai-with technical support from rBac-modified the
input and output conventions to capture daily flow dynamics.
gPcM is a complex, mathematical model, but it is not a
"black box," as it draws key input information from various
reports filed with Ferc as well as data systematically posted
on pipeline electronic bulletin boards. certain of the model
inputs governing constraints across the pipeline systems or
pipeline segments linking supply with demand are exogenous
inputs. Unlike hydraulic models that simulate actual pipeline
operations, gPcM does not explicitly capture the pressure
and flow relationships that affect how pipeline operators actually manage line pack throughout the day to serve rci and
power generation loads. in order to test the transient response
of the gas delivery system to changing operating conditions,
a hydraulic model of the physical pipeline system is needed.
For the eiPc study, Lai used WinFlow (steady-state) and
Wintran (transient) hydraulic modeling software provided by
84

ieee power & energy magazine

gregg engineering, inc. extensive hydraulic detail can be produced using the WinFlow-based pipeline simulation model of
the interconnected pipelines and storage infrastructure across
the study region. technical input parameters to the steadystate model include pipeline diameters, segment lengths,
compressor horsepower, discharge temperatures, velocities,
maximum allowable operating pressures, elevations, and gas
demands. the schematic diagram in Figure 4 illustrates the
level of detail included in the hydraulic models.
the typical modeling process is to validate the model
using known (generally peak day), steady-state operating
conditions with WinFlow and then to use that model representation in Wintran to simulate transient conditions. WinFlow's modeling features include the ability to display the
modeled system as a map featuring color-coded infrastructure information. this interface lets Lai zoom in or out and
scroll for maximum visualization. Formulation of the steadystate and transient flow models provides the PPas with a
dynamic planning tool that reveals the gas-fired generation at
risk when adverse events occur across the consolidated network of pipelines and storage facilities in the study region.

Target 1: A Baseline Gas
Infrastructure Assessment
energizing north america with natural gas requires a complex and multifaceted supply chain, from the wellhead to the
burner tip. the supply chain includes production, midstream,
transmission, storage, and distribution facilities. Figure 5
shows the network complexity of the interstate and interprovincial pipelines operating in the study region. although not
shown here, the study also identified the gas-fired electric
generators, underground storage fields, and Lng facilities in
the study region, as well as the Ldcs and intrastate, intraprovincial pipelines serving the gas-fired electric generators.
summary statistics regarding generating capacity and connectivity to pipelines and Ldcs are presented in table 1.
in addition to delineating the natural gas infrastructure
and electric interfaces, the target 1 research also examined
the storage and transportation options available to the electric
sector from pipelines and Ldcs, generator contracting and
fuel assurance practices, and capacity release and secondary
markets for gas transportation.

Target 2: Ability of Gas to
Meet Electric System Demands
the focus of target 2 is the forecast of peak-day rci and
generator gas demands throughout the study region and the
subsequent identification of potential points of stress in the
gas delivery system-points where gas-fired generators cannot obtain sufficient transportation to support the forecast
energy production level and profile. consistent with eiPc's
study design, the winter and summer peak days in 2018 and
2023 have been tested. these work efforts were centered on
two modeling platforms, the aUroraxmp chronological
electric system model and the gPcM gas network model.
november/december 2014

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - November/December 2014