2016 INFORMS Annual Meeting Program

SC05

INFORMS Nashville – 2016

2 - Gas & Power Markets: Forecasting Prices In An Evolving Energy Landscape Connie S. Trecazzi, Tennessee Valley Authority, cstrecazzi@tva.gov Environmental regulations in the energy sector paired with a tsunami of shale gas have changed power and natural gas market operations. The shift in regional gas supply is driving infrastructure changes. Lower renewable costs are impacting capacity decisions and affecting reliability requirement decisions. Electricity demand has gone through a paradigm shift as steps taken to improve energy efficiency are realized, changing views on how to model future growth. In this environment, having tools to evaluate the impact of changes in both the electricity and gas markets and pass detailed information between the models is essential to understanding how each assumption impacts both markets. 3 - The Clean Power Plan: The Art And Science Of Quantifying Its Impacts Using Integrated Gas-power Modeling Rahul Dhal, Developer, EPIS, LLC, 13535 72nd Ave., Ste. 165, Tigard, OR, 97223, United States, rahuldhal@epis.com Energy policies are, by nature, complex. The mechanisms through which policies attempt to bring about changes in the market regularly involve a large number of stakeholders. Given the decentralized and interconnected nature of U.S. energy sectors, it is very important to develop methods for evaluating the impact of the complex energy policies. In this talk we present a method for evaluating energy policies. Our methodology that integrates industry-standard modeling frameworks for gas and power markets. The integrated gas-power framework allows for evaluation of a wide-range of energy policies. We employ this framework to quantify the impact of the Clean Power Plan on both power and gas markets. 4 - Integrated Natural Gas And Electricity Modeling With RBAC GPCM And GE Maps Leah Kaffine, Senior Engineer, GE Energy Consulting Group, Schenectady, NY, United States, Leah.Kaffine@ge.com Natural gas has seen a steady increase in its market share as a fuel for power generation, with continued growth expected. GE Energy Consulting has integrated Multi Area Production Simulation Software (MAPS) with Gas Pipeline Competition Model (GPCM) in order to provide a more detailed spatial and dynamic understanding of the gas-power interaction. While existing models can capture some market dynamics in isolation, the integration of MAPS with GPCM allows for a comprehensive approach to understanding interdependent issues. Ultimately Energy Consulting’s integrated modeling allows for a consistent view of the future natural gas demand for power between the two models. SC05 101E-MCC Power Transmission Planning under Uncertainty Sponsored: Energy, Natural Res & the Environment, Energy I Electricity Sponsored Session Chair: Rodrigo Moreno, University of Chile, Av. Tupper 2007, Santiago, 8370451, Chile, rmorenovieyra@ing.uchile.cl 1 - A Comparison Of Stochastic And Adaptation Programming Methods For Electric Infrastructure Planning Patrick Maloney, Iowa State University, Ames, IA, United States, patrickm@iastate.edu, Ali Jahanbani-Ardakani, James McCalley In this work a recently developed mathematical programming formulation called adaptation is compared with traditional stochastic programming methods in the context of electric infrastructure expansion planning. While the adaptation formulation structure resembles that of a generic stochastic program it deviates from the temporal conventions of traditional expansion planning formulations. Structural comparisons and simulations are investigated to better understand differences in the methods. 2 - Value Of Model Sophistication On Transmission Expansion Planning Qingyu Xu, Johns Hopkins University, Baltimore, MD, 21218, United States, qxu25@jhu.edu, Saamrat Kasina, Benjamin Field Hobbs A set of transmission expansion plans for the western North America interconnection are optimized based on several variants of a 300-bus co- optimization model with a range of levels of sophistication, including DC optimal power flow, unit commitment and stochastic planning. The economic benefits of increasing model realism are estimated. The results show consistent impacts of sophistication upon transmission and generation investments, with load flow representations mattering most.

3 - A Five-level Milp Model For Flexible Transmission Network Planning Under Uncertainty: A Min-max Regret Approach Alexandre Moreira, Imperial College London, a.moreira14@imperial.ac.uk Goran Strbac, Rodrigo Moreno, Alexandre Street, Ioannis Konstantelos The benefits of network planning solutions have to be explicitly considered in the context of uncertainty in future realizations of generation infrastructure. Hence this talk presents a novel five-level model to determine optimal transmission expansion plans under generation expansion uncertainty in a min-max regret fashion, when considering flexible network options and n-1 security. In order to solve the five-level model on large-scale networks, we propose an effective outer algorithm. Benefits of transmission network planning significantly depend on deployment patterns of electricity generation that are characterized by severe uncertainty. In this context, this talk presents various approaches to solve the transmission expansion planning problem under generation expansion uncertainty. In particular, we compare robust and stochastic methods, and discuss about their suitability to properly balance benefits of economies of scale against risks of stranded assets. SC06 102A-MCC INFORMS 2016 Data Mining Best Student Paper Awards Sponsored: Data Mining Sponsored Session Chair: Mustafa Gokce Baydogan, Bogazici University, Istanbul, Turkey, baydoganmustafa@gmail.com SC07 102B-MCC Joint Session DM/AI: Data Mining for Decision Making Sponsored: Data Mining Sponsored Session Chair: Iljoo Kim, Saint Joseph’s University, Philadelphia, PA, United States, ikim@sju.edu 1 - Studying Agenda Setting Influence Of Online Newspaper Comments Iljoo Kim, Saint Joseph’s University, ikim@sju.edu In this continued work, we study online comments and their influence in online news articles. Using text-mining techniques, we attempt to explain and/or predict influence of online newspaper comments on the context of the original article or even on creating a new agenda through the discussions among commenters. This is done based on the textual signals embedded within comments as well as news articles. 2 - Crowdiq: Aggregating Crowd Opinions For Stock Price Predictions The Wisdom of Crowds (WoC) theory explains how crowd opinions should be aggregated in order to improve the performance of decision making. Diversity, independence, decentralization, and aggregation are important factors to crowd wisdom. Existing opinion aggregation methods fail to collectively consider all the factors of crowd wisdom. We propose a new opinion aggregation method, namely CrowdIQ, to evaluate crowd wisdom using all four factors. We apply CrowdIQ to a stock prediction task using user-generated stock tweets. The result shows that CrowdIQ outperforms baseline methods. Qianzhou Du, Virginia Tech, 100 Otey Street, Room 301, Blacksburg, VA, 24061, United States, qiand12@vt.edu Hong Hong, Alan Wang, Weiguo Fan 4 - Uncertainty In Strategic Network Investment: Stochastic vs. Robust Min-max Approaches Rodrigo Moreno, University of Chile, Santiago, Chile, rmorenovieyra@ing.uchile.cl, Goran Strbac

70