(554a) Power Systems Infrastructure Planning with High Renewables Penetration

The design of energy systems is increasingly focusing on systems that involve renewable energies. As of 2019, 64% of the electric generation capacity additions come from solar and wind [8]. The increased penetration of renewables has brought up new challenges to power systems infrastructure planning. First, renewable generations are subject to weather conditions, which makes their power output volatile. Second, the renewable generating units are usually placed in remote areas that are not well connected with load demand.

Power systems planning is generally solved as two independent optimization problems, generation expansion planning and transmission expansion planning. Generation expansion planning (GEP) of power systems involves determining the optimal size, location, and construction time of new power generation plants, while minimizing the total cost over a long-term planning horizon [2,3,5]. Transmission expansion planning (TEP) refers to installing new transmission lines or expanding the capacities of existing transmission lines in a power system. The recognition of transmission’s interaction with generation expansion has motivated the development of co-optimization methods to consider the tradeoffs between generation and transmission expansion. Several works have been reported to simultaneously optimize generation and transmission expansion planning (GTEP) [1,4,6].

The penetration of renewable in remote areas with high wind and solar capacity factors requires coordination of generation and transmission, which makes it even more meaningful to solve the GTEP problems. However, GTEP models with renewable generations can lead to large-scale problems that are intractable with state-of-the-art commercial solvers.

This presentation proposes a GTEP model that not only includes the investment decisions on the thermal and renewable generators, but also considers the detailed operating conditions of the generators to guarantee the feasibility of the design subject to volatile weather conditions. To limit the size of the GTEP model, both spatially and temporal simplifications are made. We aggregate the generators that use the same technology assuming that they have the same design parameters. We also spatially aggregate regions with similar climate and load profiles. We compare three different formulations for transmission expansion, i.e., the big-M formulation, the hull formulation and the alternative big-M formulation [7]. We prove that the alternative big-M (ABM) formulation has the same feasible region as the big-M formulation (BM) when projected onto the space of the variables involved in the big-M formulation. Computational experiments are performed as well for the three formulations. It is shown that none of the three formulations is superior as each has its strengths and weaknesses.

Despite the spatial and temporal simplifications, the MILP model still has tens of millions of variables, which cannot be solved by the state-of-the-art commercial solvers directly. Two solution techniques, a nested Benders decomposition algorithm and a tailored Benders decomposition algorithm, are proposed. The tailored Benders decomposition algorithm outperforms the nested Benders decomposition in our computational experiments, and is able to solve the large MILP model within reasonable computational time.

An ERCOT case study is used to demonstrate the GTEP model and the solution techniques. The tailored Benders decomposition is able to solve the 20 year planning problem with up to 15 representative days. The capacity expansion mix for ERCOT will mainly include solar and wind capacities in the West and Panhandle regions. The transmission lines are mainly built to transfer power from solar and wind rich regions to the South and Northeast regions of ERCOT, which shows that the generation and transmission decisions are correlated. Co-optimization of generation and transmission has the potential to bring additional value to the system operator/regulator than solving the two planning problems independently.

[1] Aghaei, J., Amjady, N., Baharvandi, A., Akbari, M.A.: Generation and transmission expansion planning: MILP–based probabilistic model. IEEE Transactions on Power Systems29(4), 1592–1601 (2014)

[2] Conejo, A.J., Baringo, L., Kazempour, S.J., Sissiqui, A.S.: Investment in Electricity Generation and Transmission - Decision Making under Uncertainty. Springer International Publishing (2016). DOI 10.1007/978-3-319-29501-5

[3] Koltsaklis, N.E., Dagoumas, A.S.: State-of-the-art generation expansion planning: A review. Applied Energy230, 563–589 (2018)

[4] Krishnan, V., Ho, J., Hobbs, B.F., Liu, A.L., McCalley, J.D., Shahideh-pour, M., Zheng, Q.P.: Co-optimization of electricity transmission and generation resources for planning and policy analysis: review of concepts and modeling approaches. Energy Systems7(2), 297–332 (2016).

[5] Lara, C.L., Mallapragada, D.S., Papageorgiou, D.J., Venkatesh, A., Grossmann, I.E.: Deterministic electric power infrastructure planning: Mixed-integer programming model and nested decomposition algorithm. European Journal of Operational Research271(3), 1037–1054 (2018)

[6] Pozo, D., Sauma, E.E., Contreras, J.: A three-level static MILP model for generation and transmission expansion planning. IEEE Transactions on Power systems28(1), 202–210 (2012)

[7] Bahiense, L., Oliveira, G.C., Pereira, M., Granville, S.: A mixed integer disjunctive model for transmission network expansion. IEEE Transactions on Power Systems16(3), 560–565 (2001)

[8] U.S. Energy Information Administration. New electric generating capacity in 2019 will come from renewables and natural gas. URL= eia.gov/todayinenergy/detail.php?id=37952