Combined Design and Control Optimization: Application to Optimal PHEV Design and Control for Multiple Objectives.
Year: 2012
Type: article
Abstract: This dissertation develops algorithms for optimal design and control solutions of dynamic systems in a computationally efficient manner. These methods are demonstrated by applying them to a Plug-in Hybrid Electric Vehicle (PHEV) powertrain’s optimal design and control. Since a PHEV draws energy from the grid it is important to consider these interactions in its optimal design and control decisions. The battery size also affects the amount of grid energy transferred to propulsion and consequently the on-road power management decisions. Thus, we develop algorithms to determine the optimal PHEV battery size and control decisions considering conditions on the electric grid. First, we develop a Dynamic Programming (DP) based algorithm for optimal on-road power management of a series PHEV. A backward looking implementation of the PHEV powertrain’s dynamic model with the DP algorithm avoided the need to interpolate the value function or enforce constraints through penalty functions, thereby alleviating computational concerns. This algorithm is extended to consider optimal charging on the electric grid by utilizing conditions at the boundaries of the optimal charging and driving problems. The results exposed tradeoffs between the two problems. This algorithm was further applied to determine the optimal CO2 reduction benefits in propulsion depending on wind penetration on the grid. Since PHEVs are expected to address emissions, cost, and participate in grid services, we develop a multi-objective dynamic programming (MODP) algorithm. This algorithm utilizes the idea of crowding distance from Non-Dominated Sort Genetic Algorithms (NSGA) literature to represent the Pareto front with fewer points, easing the computational time and memory requirements. Tradeoffs in achieving minimum CO2 vs. minimum operational costs are discussed. Finally, we utilize the theory on combined optimization of a system’s design and control for PHEV battery sizing considering its optimal charging and power management. Salient features of the algorithm such as calculation and use of a sensitivity term and the reduction of computational effort are demonstrated by solving a beam mass reduction and vibration attenuation problem. Then, this algorithm is applied for optimal battery sizing of a PHEV. The sensitivity values provided insights for optimal PHEV battery sizing while considering the optimal control.
Author: Rakesh M. Patil
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Keywords: plug-in hybrid electric vehicles, powertrain design optimization, battery sizing, dynamic programming, multi-objective optimization, Pareto front, crowding distance, electric vehicle charging, CO2 emission reduction, wind power penetration, sensitivity analysis, vibration attenuation, optimal control
Open Access status: green
OpenAlex ID: W100756261
