Minimizing the Cost for Charging Plug-in Electric Vehicles Using BFGS Quasi-Newton Method
Open Access
- Author:
- Li, Zhiran
- Graduate Program:
- Electrical Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 24, 2022
- Committee Members:
- Kultegin Aydin, Program Head/Chair
Yan Li, Thesis Advisor/Co-Advisor
Donald E Ebeigbe, Committee Member - Keywords:
- Minimizing the cost of charging
Optimization methods
Quasi- Newton method
Newton’s method
Broyden–Fletcher–Goldfarb–Shanno (BFGS)
Plug-in Electric Vehicles (PEVs).
Plug-in Electric Vehicles (PEVs) - Abstract:
- This thesis presents a Broyden–Fletcher–Goldfarb–Shanno (BFGS)-enabled Quasi- Newton method to minimize the cost for charging Plug-in Electric Vehicles (PEVs). Firstly, the objective function for minimizing the cost for charging PEVs is introduced, as well as the constraints on the charger, the state of charge, and the voltage levels. Secondly, the mathematical formulas of the BFGS Quasi-Newton algorithm are presented to calculate the voltage of a node topology. After that, the charging results for using both Newton’s method and BFGS Quasi-Newton method are compared in two scenario cases. The grid system in the first scenario is a traditional power grid that consists of 18 power nodes. And the power node is a PEV charging station that can handle bidirectional power flow. And the grid system in the second grid topology is a micro-grid system that consists of 8 power nodes and 10 Distributed Energy Resources (DER) generation buses including photo- voltaic, micro-turbine and fuel cell etc. The test result also shows BFGS-enabled method has a lower cost in charging PEVs than using the Newton’s method in same grid topology and same situation. Therefore, the comparison between the BFGS-enabled Quasi-Newton method and Newton’s method verifies that the BFGS- enabled method can better minimize the cost for charging PEVs.