Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TC48

3 - Improving the Fidelity of Large Electric System Production Cost Models Clayton Barrows, NREL, 15013 Denver West Parkway, Golden, CO, 80401, United States The evolution of electrical power systems is driven by analysis using power system planning and design software tools. Many of these tools involve production cost models for optimizing scheduling decisions. This presentation highlights a suite of four interrelated Grid Modernization Laboratory Consortium projects that involve improving the fidelity and computational efficiency of production cost models, and using those improvements to gain insights into infrastructure plans and market design. 4 - Warm Starting for Security Constrained Deterministic Unit Commitment Suresh Bolusani, Lehigh University, 200 West Packer Avenue, Bethlehem, PA, 18015, United States, Feng Qiu, Ted K. Ralphs, Audun Botterud, Kibaek Kim The deterministic unit commitment problem is usually formulated as a mixed integer linear optimization problem and solved from scratch multiple times a day resulting in a loss of useful information from previous solution processes. This loss can be avoided by warm starting the solution process. Warm starting is a process of collecting information from the solution process of one instance and reusing this information to accelerate the solution process of a closely related instance. In this talk, we present a warm starting methodology for solving deterministic unit commitment problem with transmission line constraints. We also present our computational results with an open-source implementation. Chair: Amy R. Ward, University of Southern California, Marshall School of Business, Bridge Hall BRI 401H, Los Angeles, CA, 90089- 0809, United States 1 - Bayesian Optimization Peter Frazier, Cornell University, School of Operations Research, and Information Engineering, Ithaca, NY, 14853, United States Bayesian optimization is widely used for tuning deep neural networks and optimizing other black-box objective functions that take a long time to evaluate. In this tutorial, we describe how Bayesian optimization works, including the Bayesian machine learning model it uses to model the objective function, Gaussian process regression, and three common acquisition functions: expected improvement, entropy search, and knowledge gradient. We then describe applications at Yelp and Uber, explain techniques important for making it work well in practice, and survey techniques for solving “exotic” Bayesian optimization problems. n TC48 North Bldg 229B Sustainability Sponsored: Energy, Natural Res & the Environment Environment & Sustainability Sponsored Session Chair: Gulver Karamemis, University of Rhode Island, Kingston, RI, 02881, United States 1 - Modeling Economic Impacts of Climate Change Induced Weather Events to Natural Gas Sector in California Duan Zhang, University of California, Santa Cruz, CA, 95062, United States, Yihsu Chen We studied the regional economic impacts of weather events such as sea level rise and wildfire with a focus on natural gas sector. A computable general equilibrium (CGE) model with bilateral trade flow was constructed. The model integrates county-level details for California and state-level data for the rest of US. We used IMPLAN data to acquire social account matrices that characterize different sectors and regions. Several cases were simulated to represent impacts of sea level rise on downstream consumption of natural gas and impacts of wildfire on upstream supply. 2 - Manufacturers’ Competition and Cooperation in Sustainability: Stable Recycling Alliances Fang Tian, Pepperdine University, 2720 Kensington Ave, Thousand Oaks, CA, 91362, United States, Greys Sosic, Laurens G. Debo We study the stability of producers’ strategies emerging under Extended Producer Responsibility. In our paper, the producers compete with multiple, differentiated products in consumer markets, but may consider cooperating when recycling n TC47 North Bldg 229A Bayesian Optimization Emerging Topic Session

of many renewables and the lack of provision for algorithms that would ensure a successful operation of the renewable plants during extreme conditions. We explore the technical capabilities of hybrid (renewable and storage) units to be used as Black Start resources. We develop an optimization-based model that will recommend such systems to maximize system resilience. We illustrate our approach in a number of test power systems. n TC45 North Bldg 228A Frontiers on Combined Cycle Modeling for Electric Market Clearing Sponsored: Energy, Natural Res & the Environment/Electricity Sponsored Session Chair: Yongpei Guan, University of Florida, Gainesville, FL, 32611, United States Co-Chair: Yonghong Chen, Midwest ISO, Midwest ISO, Carmel, IN, 46032, United States 1 - A Tight Configuration-component Based Hybrid Model for Combined-cycle Units in Miso Day-ahead Market Lei Wu, Clarkson University, 8 Clarkson Avenue, P.O. Box 5720, Potsdam, NY, 13676, United States 2 - Frontiers on Combined Cycle Modeling for Electric Market Clearing Yonghong Chen, Midwest ISO, 720 City Center Drive, Carmel, IN, 46032, United States Combined cycle Modeling represents one of the most complicated resource Modeling in the electricity market clearing process. This panel discusses current advances in the modeling and solution approaches for systems with large number of combined cycles. Panelists Yongpei Guan, University of Florida, 303 Weil Hall, P.O. Box 116595, Gainesville, FL, 32611, United States Jim Ostrowski, University of Tennessee, 11421 Old Colony Pkwy, Knoxville, TN, 37934, United States Bowen Hua, University of Texas at Austin, Austin, TX, United States n TC46 North Bldg 228B Grid Modernization Lab Consortium Sponsored: Energy, Natural Res & the Environment/Energy Sponsored Session Chair: Feng Qiu, Argonne National Laboratory, Lemont, IL, 60439, United States Co-Chair: Clayton Barrows, NREL, Golden, CO, 80401, United States 1 - Energy Storage Siting And Sizing for Economic Power Dispatch Considering Battery Degradation Liu Su, University of South Florida, ENG 302, Tampa, FL, 33620, United States, Kibaek Kim, Audun Botterud, Changhyun Kwon Energy storage can alleviate the uncertainty of energy resources, provide flexible power system scheduling and reduce the costs of energy generation for power grids. We propose a long-term planning model for siting and sizing of energy storage under the consideration of battery degradation and formulate it as a mixed-integer programming (MIP) problem. To solve the large-scale MIP, we implement the temporal decomposition in the parallel decomposition framework. We will present the computational results on Reliability Test System - Grid Modernization Lab Consortium. 2 - Scalable Capacity Expansion for Explicit Representation of Intermittent Generation Devon Sigler, National Renewable Energy Laboratory, Golden, CO, United States, Gord Stephen, Bethany Frew, Wesley Jones Capacity expansion models inform power system infrastructure planning decisions so future electrical power demand on the grid is met economically and reliably. With a growing amount of intermittent renewable energy being used on the grid, the number of operational scenarios that must be considered to meet these goals is growing. Historically, the number of operational scenarios considered has been limited due to the resulting increase in problem complexity from doing so. We present a scalable capacity expansion model which uses progressive hedging to solve the model via parallel computing. Planning decisions computed considering a large number of operation scenarios are presented.

323