Informs Annual Meeting 2017

WC78

INFORMS Houston – 2017

WC78

optimally aggregate PCPs into a larger “box” with the expectation that, 1) a large fire would be contained within the box by fire line construction efforts along the box boundary; 2) point protection could be implemented in a box to protect homes and infrastructures. This model directly aggregates polygons into a contiguous patch. Test results illustrate how the size and shape of the box, along with the proportion of suppression effort devoted to line versus point protection activities, varies with ignition and weather scenarios. 2 - A Two-stage Model to Allocate Financial among Landowners for Controlled Burn Hugh Medal, Mississippi State University, MS, United States, Medal, Hugh This research aims to reduce the potential damage caused by a wildfire by encouraging controlled burns on private lands through a cost-share program. An agency having a limited budget offers financial assistance to the landowners with uncertainty about the landowner’s decision of accepting or rejecting the offer. The objective is to optimally allocate the financial assistance among the landowners to minimize the expected total acreage burned. We model this problem as a two- stage stochastic program; the first-stage determines the agency’s budget allocation decision among the landowners and the second-stage models the spread of fire through the landscape. 3 - Evaluating the Performance of the Aerial Forest Fire Detection System in the Province of Ontario, Canada David L.Martell, University of Toronto, Faculty of Forestry, 33 Willcocks Street, Toronto, ON, M5S.3B3, Canada, david.martell@utoronto.ca Forest fire management agencies often use fixed-wing detection aircraft that carry detection observers that search for undetected forest fires as the pilot flies planned routes. We describe how we used data collected by on-board GPS units that track the aircraft to estimate probability that a detection observer will find undetected fires and to evaluate the performance of the aerial detection system. 381C Data-analytic Methods in Renewable Energy Sponsored: Energy, Natural Res & the Environment, Energy Sponsored Session Chair: Andrea Staid, Sandia National Labs, Sandia National Labs, Albuquerque, NM, 87108, United States, astaid@sandia.gov 1 - Synchronicity Assessment for Wind Farm Siting: Wind Drought Dynamics with System Load Kristen Schell, University of Michigan, 1891 Beal St., Ann Arbor, MI, 48109, United States, krschell@umich.edu, Seth Guikema, Brent McRoberts There are over 10,000 individual wind turbines currently operating in the state of Texas. Tax credits have helped spur investment in the Texas wind industry, which anticipates a further 11 GW of new capacity additions in the near future. While this increase in wind capacity at times provides a remarkable percentage of load with renewable generation, variable wind power output is not often synchronous with peak demand, which raises the issue of its contribution to overall resource adequacy. Clustering the results of spectral analyses of wind droughts illustrate locations which can better contribute to resource adequacy, by providing wind power more highly synchronous with system load. 2 - Modeling Sustainable Energy Supply Case Study for Renewables and Natural Gas Synergy Ebisa Wollega, Assistant Professor, Colorado State University Pueblo, 2200 Bonforte Blvd Room 260, Pueblo, CO, 81001, United States, ebisa.wollega@csupueblo.edu, Hiba Baroud, Vitor Winckler The penetration of renewables into the power grids has significantly increased over the past decade. However, the supply uncertainty of the renewables results in complexities when modeling the supply and demand dynamics of power in the future. In this paper, we present a stochastic model of power networks in which we combine the renewables with natural gas through large scale data processing where the renewables are used as the main energy supply sources and the natural gas is used as a backup in order to minimize the supply uncertainty. 3 - Integrated Prediction of Wind Speed and Power Production Jingxing Wang, University of Michigan, 1205 Beal Ave., Ann Arbor, MI, 48109, United States, jeffwjx@umich.edu, Abdullah Alshelahi, Eunshin Byon, Romesh Saigal, Mingdi You We present a new integrated prediction methodology for wind power production under the assumption that the underlying dynamics follow the Geometric Brownian Motion. The proposed predictive model is a novel integration of the sequential adaptive regularized learning power curve modeling, theories in real options and dual Kalman filtering. Our method provides promising results, achieving lower prediction errors than alternative methods. WC80

381A Power Network Transformation in Smart & Connected Communities Sponsored: Energy, Natural Res & the Environment Electricity Sponsored Session Chair: Vignesh Subramanian, vigneshs@mail.usf.edu Co-Chair: Tapas Das, das@usf.edu 1 - Harnessing Flexibleand Reliable Demand Response under Customer Uncertainties Joshua Comden, Stony Brook University, Jericho, NY, 11753, United States, na, Zhenhua Liu This work first models the problem of joint capacity planning and demand response program design by a stochastic optimization problem and then proposes online DR control policies based on the optimal structures of the offline solution. A distributed algorithm is developed for implementing the control policies without efficiency loss, which is further enhanced by allowing flexibilities into the commitment level. Numerical simulation results based on real-world traces demonstrate that the proposed algorithms can achieve near optimal social costs and significant social cost savings. 2 - Menu-based Pricing for Charging of Electric Vehicles with Vehicle-to-grid Service Vaneet Aggarwal, Purdue University, 315 N. Grant St., West Lafayette, IN, 47906, United States, vaneet@purdue.edu, Arnob Ghosh The paper considers a bidirectional power flow model of the electric vehicles (EVs) in a charging station where the EVs can inject energies by discharging via a Vehicle-to-Grid (V2G) service. However, frequent charging and discharging degrade battery life. A proper compensation needs to be paid to the users to participate in the V2G service. We propose a menu-based pricing scheme, where the charging station selects a price for each arriving user for the amount of battery utilization, the total energy, and the time (deadline) that the EV will stay. The user can accept one of the contracts or rejects all depending on their utilities. We explore social welfare and profit maximization strategies. 3 - Joint Strategies for Dynamic Pricing of Electricity and Demand Response in Smart and Connected Communities Vignesh Subramanian, 5006, Bordeaux Village Pl, Apt #201, Tampa, FL, 33617, United States, vigneshs@mail.usf.edu, Tapas Das A model-based methodology developed to aid policy makers to understand the joint behavior of the power producers, power systems operator, and S&CC aggregator. In this paper, we present a model that yields joint strategies for dynamic pricing (DP) and demand response (DR). The goal of DP is to offer consumers a stable advance price signals that are as close as possible to the real- time settlement price that results from the aggregator’s cost minimizing DR decisions. 4 - Coordinating Procurement Decisions with a Dispatch Strategy Mark Husted, Colorado School of Mines, 816 20th Street, Golden, CO, 80401, United States, mhusted@mines.edu A optimization model designs and operates a hybrid power system consisting of diesel generators, photovoltaic cells and battery storage to minimize fuel use at remote sites subject to meeting variable demand profiles subject to the following constraints: (i) power generated must meet demand in every time period; (ii) power generated by any technology cannot exceed its maximum rating; and (iii) best practices should be enforced to prolong the life of the technologies. We solve instances of the model to within 2% of optimality and show that a model with minute time fidelity is able to meet highly variable energy demand using the procurement strategy dictated by a model with hourly time fidelity. 381B Forestry II - Wildland Fire Management Sponsored: Energy, Natural Res & the Environment Forestry Sponsored Session Chair: David L. Martell, University of Toronto, Toronto, ON, M5S 3B3, Canada, david.martell@utoronto.ca 1 - A MIP Model to Support the “Confine and Contain” Strategy in Large Fire Management Yu Wei, Associate Professor, Colorado State University, Department of FRWS, Forestry 102, Fort Collins, CO, 80523, United States, yu.wei@colostate.edu, Matthew Thompson We first delineate a landscape into potential fire control polygons (PCP) following potential fire control lines. We introduce a new mixed integer program (MIP) to WC79

530