(637b) MILP Formulation for Optimal Planning of Electric Power Infrastructure

Energy systems planning models allow the evaluation of alternate scenarios of future growth, providing information to support the decision-making process and the selection of technology pathways in the power sector. Although different models [1,2] have been developed to plan the expansion of generation capacity and transmission infrastructure for long-term time horizons, there are major challenges that have not been fully addressed yet, such as the aging of the infrastructure, and the spatial and temporal multi-scale detail required to adequately model the increasing contribution of intermittent renewable generation to the power grid.

In this paper we propose an optimization modeling framework to evaluate the changes in generation and transmission infrastructure required to meet the projected demand for electricity over the next few decades while taking into account detailed operational constraints (e.g. unit commitment) and the variability and intermittency of renewable generation sources. The modeling framework, which is based on mixed-integer linear programming (MILP), takes the viewpoint of a central planning entity whose goal is to identify the source (nuclear, coal, natural gas, wind and solar), type, location and capacity of future power generation technologies and transmission infrastructure that can meet the projected electricity demand while minimizing the amortized capital investment of all new generating units and transmission lines, the operating and maintenance costs of both new and existing units, and suitable environmental costs (e.g. carbon tax).

The proposed formulation is applied to a case study in the region managed by the Electric Reliability Council of Texas (ERCOT) for a 30 year planning horizon, resulting in a large-scale MILP model with more than 100,000 discrete variables and constraints. Through a combination of judicious model formulation strategies such as time sampling [3] and generator clustering [4] as well as the use of specialized decomposition strategies [3], this work demonstrates that these large-scale MILP problems can be solved in a reasonable amount of time. The results for hourly and sub-hourly level of information are compared and, in both cases, it shows that the future growth will be met by a portfolio of different generation technologies.

[1] ReEDS model by NREL http://www.nrel.gov/analysis/reeds/

[2] EPAâ??s Power Sector Modeling https://www.epa.gov/airmarkets/power-sector-modeling-platform-v515

[3] Mitra, S., Pinto, J. M., & Grossmann, I. E. (2014). Optimal multi-scale capacity planning for power-intensive continuous processes under time-sensitive electricity prices and demand uncertainty. Part I: Modeling. Computers & Chemical Engineering, 65, 89-101.

[4] Palmintier, B. S., & Webster, M. D. (2014). Heterogeneous Unit Clustering for Efficient Operational Flexibility Modeling. IEEE Trans. Power Syst. IEEE Transactions on Power Systems, 29, 1089-1098.