Informs 2015

447 / 552

INFORMS Philadelphia – 2015
445
3 - Optimal Assembly Planning to Manage Manufacturing and Supply
Chain Complexity Due to Product Variety
Jeonghan Ko, University of Michigan and Ajou University,
1205 Beal Ave., Ann Arbor, MI, United States of America,
Heng Kuang, Ehsan Nazarian
The complexity due to product variety has been an important issue for
manufacturing and supply chains. Assembly planning determines the
manufacturing process sequence and logistics, and plays a crucial role in
minimizing such complexity. We propose a new complexity measure of assembly
plans, verify the index through manufacturing and supply chain models, and
provide a strategy for improved assembly plans. Our result illustrates the
advantages of delayed differentiation and balanced planning.
4 - How Much Efficiency Is Enough?
Mohammad Ali Asudegi, University of Tennessee, 525 John D.
Tickle Engineering Building,, 851 Neyland Dr, Knoxville, TN,
37996, United States of America,
Rupy Sawhney
For many years the focus of many companies has been on efficiency metrics
while other metrics have been ignored to gain higher efficiency. To define an
optimal level of efficiency in a facility the interaction between efficiency and all
other metrics should be studied. In this study, a framework is offered to define
optimal level of efficiency considering other financial and non financial metrics
for a better use of available resources.
5 - Managing Electricity Peak Loads in Make-to-Stock
Manufacturing Lines
Felix Papier, Associate Professor, ESSEC Business School, Avenue
Bernard Hirsch, Cergy, 95021, France,
Our research is motivated by new manufacturing systems that smoothen
electricity consumption to avoid expensive and carbon-emission-intensive peak
loads. We study the control of stochastic make-to-stock manufacturing lines in
the presence of peak loads. We show that standard methods in manufacturing
research are not effective in this setting and we develop a new control policy. We
derive analytic properties of the control policy for 2 workstations and develop a
heuristic for more workstations.
WC57
57-Room 109B, CC
Power System Design and Optimization
Sponsor: ENRE – Energy I – Electricity
Sponsored Session
Chair: Amir Mousavian, Assistant Professor, Clarkson University, 8
Clarkson Avenue, Potsdam, NY, 13699-5790, United States of America,
1 - Optimizing Primary Response in Preventive Security-constrained
Optimal Power Flow
Hrvoje Pandzic, Faculty of Electrical Engineering and Computing
University of Zagreb, Unska 3, Zagreb, Croatia,
, Pierre Henneaux, Yury Dvorkin,
Daniel Kirschen
Preventive Security-Constrained (PSC) Optimal Power Flow (OPF) dispatches
controllable generators at minimum cost while ensuring that operating
constraints on generation and transmission assets are respected during both the
pre- and post-contingency states without relying on post-contingency redispatch.
A new PSCOPF model that optimizes the droop coefficients of the synchronized
generators will be described.
2 - Optimal Resilient Distribution Grid Design using a 3-phase
Unbalanced AC Power Flow
Russell Bent, Los Alamos National Laboratory, P.O. Box 1663, Los
Alamos, NM, 87545, United States of America, ,
Emre Yamangil, Harsha Nagarajan, Scott Backhaus
Modern society is critically dependent on the services provided by power systems.
Here we discuss an optimal electrical distribution grid design problem that
improves the resiliency of such systems. To improve tractability, we introduce two
ways of approximating the 3-phase AC power flow equations and L-shaped
infeasibility cuts. Our experiments show that, unless the network impedance
profile is unrealistically modified, our algorithms remain tractable and the
solutions provide good results.
3 - Dispatchability Maximized Energy and Reserve Dispatch
Wei Wei, Tsinghua University, 3-211, West Main Building,
Tsinghua Univ, Beijing, 100084, China,
, Shengwei Mei, Jianhui Wang
The dispatchability of the affine policy based energy and reserve dispatch (AF-
ERD) is shown to be a polytope. The mathematical model of AF-RED is proposed.
An efficient SOCP based algorithm is developed to solve the proposed model. The
generalized Gauss inequality is adopted to evaluate the probability of infeasible
real-time dispatch in the absence of the exact probability distribution of wind
power.
4 - Optimal Energy Scheduling in a Microgrid Including Electric
Vehicle Parking Lots
Ebrahim Mortaz, Auburn University, Auburn University, Auburn,
AL, 36849, United States of America,
Jorge Valenzuela
Microgrids are small-scale versions of the current bulk power grids. Microgrids
allow customers to participate in energy trading and demand response programs.
In this presentation, we consider scheduling electric power in a microgrid that
includes thermal generators, renewable energy, and a parking lot with electric
vehicles. We propose a mathematical programming model that minimizes the
total expected operation cost. Results confirm that the proposed energy
scheduling reduces generation costs.
WC58
58-Room 110A, CC
Optimal Power Flow in Electric Power Systems II
Sponsor: ENRE – Energy I – Electricity
Sponsored Session
Chair: Andy Sun, Assistant Professor, Georgia Institute of Technology,
755 Ferst Drive, Atlanta, GA, 30332, United States of America,
1 - Power Markets with Real and Reactive Power Based off a
Sequential Linear Approximation to the ACOPF
Paula Lipka, Graduate Student Researcher, University of
California, Berkeley, 4141 Etcheverry Hall, Berkeley, CA, 94720,
United States of America, , Shmuel Oren
This work shows how to use a successive linear program that solves the ACOPF to
run a real-time power market. It discusses how to calculate load payments,
generation rents, congestion rents, and flowgate prices and how these items are
different from the DC terms due to voltage and reactive power. It compares
market outcomes under these new settings to those from a DCOPF approach with
several examples.
2 - Impact of ACOPF Constraints on Security-constrained
Unit Commitment
Anya Castillo, Federal Energy Regulatory Commission,
888 1st Street NE, Washington DC, United States of America,
Richard O’Neill, Carl Laird,
Jean-Paul Watson, Cesar Silva Monroy
We propose a unit commitment formulation with alternating current optimal
power flow (ACOPF) constraints and solve the problem as a mixed-integer
sequential linear program. This approach accounts for voltage requirements and
enables the commitment of units for reactive power compensation in addition to
economically satisfying the real power balancing requirements. We compare our
results to unit commitment approaches that either ignore network effects or
approximate lossless power flows.
3 - First Order Loss Approximation for LMP Calculation
Brent Eldridge, Operations Research Analyst, Federal Energy
Regulatory Commission, 888 First St. NE, Washington, DC,
United States of America, , Richard O’Neill,
Anya Castillo
The following paper discusses an improvement method for estimating line losses
when solving a DC optimal power flow (DCOPF) with endogenous line loss
estimation. We present a DCOPF model and propose a method called First Order
Improvement of Losses, or FOIL, which uses only linear constraints and does not
require additional solutions to AC power flow equations. We compare FOIL to the
initial solution and to results using successive linearization.
4 - Robust Feasibility and Stability Analysis in Power Flow Problems
Krishnamurthy Dvijotham, California Institute of Technology,
Pasadena, CA, United States of America,
Recent research has shown that convex relaxations often find provably optimal
solutions of ACOPF problems. However, they do not provide guarantees of
feasibility and dynamic stability in the presence of uncertainty (contingencies,
stochastic generation, load etc.). In this work, we develop efficient algorithms to
certify stability and feasibility of power flow solutions under various forms of
uncertainty. The certificates cover practical operating conditions for several test
cases.
WC58