Optimizing deep decarbonization pathways in California with power system planning using surrogate level-based Lagrangian relaxation
Osten Anderson, Mikhail A. Bragin, Nanpeng Yu
Abstract
With California’s ambitious goals to decarbonize the electrical grid by 2045, significant challenges arise in power system investment planning. Existing modeling methods and software focus on computational efficiency, which is currently achieved by simplifying the associated formulations. The simplifications, such as linear relaxation of binary decisions, may lead to significant inaccuracies in the cost and constraints of generation operations and may affect both the timing and the extent of investment in new resources, such as renewable energy and energy storage. To address this issue, this paper develops a more detailed and rigorous mixed-integer linear programming model where the discrete nature of the problem is properly captured. Further, a solution methodology utilizing surrogate level-based Lagrangian relaxation is adapted to address the combinatorial complexity that results from the enhanced level of model detail. Our methodology is capable of efficiently handling the decarbonization model with approximately 12 million binary and 100 million total variables in under 48 h, representing a massive leap in the ability to solve complex planning models in acceptable time. The solution obtained by our method leads to an investment plan, which is then compared with the plan produced by E3’s RESOLVE software currently employed by the California Energy Commission, California Public Utilities Commission as well as by three major California utilities: SCE, PG&E, and SDGE. Our model produces an investment plan that leads to substantial savings to the State of California of over 4 billion dollars over the investment horizon as compared to the existing method.