IEEE Power & Energy Magazine - July/August 2016 - 37

Grid planning should anticipate how generation investments might
shift in response to transmission investment, which would represent
a proactive or anticipative transmission planning paradigm.

gained from applying stochastic programming to transmission planning.

Stochastic Programming Is Feasible
The results from the JhU-weCC studies show that it is
practical to identify economically optimal grid additions
considering multiple scenarios simultaneously using stochastic programming. The model was able to recommend
a set of initial (years 1-10) transmission investments considering multiple scenarios, together with later (years
11-20) investments, which adapted to the particular scenario that was realized while anticipating how generator
investment would react to those grid enhancements. The
model was solved on a lab-level desktop, and the most
sophisticated case (300-bus network, 20 long-run scenarios, 4 million constraints, 3 million continuous variables,

Objectivies

and 1,113 binary variables for transmission investments)
took four hours to run. This means a more detailed network with more scenarios can be solved on an enterpriselevel server/cluster.
In other words, deriving an adaptive transmission plan
for an uncertain future is very practical. what makes such
a model particularly useful is that, once it is set up, a single
analyst can alter assumptions and develop a new optimal
transmission-generation investment scenario quickly. In
contrast, traditional planning processes, which develop
generation investment build-out scenarios and then evaluate a number of alternative transmission configurations
with production models, can take person-weeks of effort
per scenario. stochastic optimization can, therefore,
allow analysts to consider the impact of many more sets
of assumptions on transmission recommendations.

Uncertainties
Will Cause
Will Limit
Will Change
Will Affect

Transmission
Expansion Cost

Operation Cost
Fixed Variable
O&M
O&M

Fuel

Carbon
Tax

Generation
Investment Cost

Decision Variables
Choices of
Candidate
Lines

Operating
Phase Angle

Production

Invested
Generation
Capacity

Constraints
Transmission
* Kirchhoff's Laws
* Thermal Limits
* Path Limits;
Effects of New
Line Additions
on Limits

Generation
* Existing Generation
Capacity
(Pmax, Pmin)
* Hourly Variability of
Renewable Energy
* Ramp Rates

Policy
* State and
Federal
Renewable
Portfolio
Standards
* Carbon

Siting Limits:
* Regional
Maximum Wind
and Solar Energy
Availability
* New Gen
Capacity
* Maximum Line
Additions by
Corridor

Economy:
* Load Growth
* Peak Growth
Fuel Price:
* Natural Gas
* Coal
Technology:
* Generation Capital Cost
* DR, DG, and
Storage Growth
Policy:
* Renewable Portfolio
Standards
* Carbon Tax
Climate Change:
* Hydro Power
Availability
Permitting:
* "Failure to Launch" of
Planned Tran Line

figure 5. The interactions among uncertainties and optimization model components. O&M: operations and maintenance;
DR: distributed resources; DG: distributed generation.
38

ieee power & energy magazine

july/august 2016
Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - July/August 2016
