Novel optimal control algorithms with application to the parallel hydraulic hybrid vehicle power train.

The parallel Hydraulic Hybrid Vehicle (HHV) power train is quickly becoming a viable option among large (class 7-10) vehicles. This is due to its potentially vast improvements in fuel economy over non-hybrid power trains. Optimal control of the parallel HHV power train is critical to overall vehicle performance and is largely responsible for gains in efficiency. The research presented in this thesis aims to answer the question of how to best operate the power train to achieve maximum efficiency during driving intervals when vehicle speed is unspecified, except at boundary points. A state-space model of the parallel HHV power train is derived in the energy domain using a classical Lagrangian approach. Two optimal control algorithms are developed and applied to the vehicle. The first algorithm is gradient descent based and is derived using the calculus of variations. The second algorithm discretizes the optimal control problem in time and converts it to a non-linear program. Several optimal control problems are solved and the results offer valuable insight into efficient operation of the parallel HHV power train.

Author dc:creator

• Ertel, Robert Gregory

• Control
• Hybrid
• Optimal
• Parallel
• Mechanical Engineering

Language dc:language.iso

en_US

Repository record dc:identifier.uri

http://purl.umn.edu/104813

OAI identifier oai:identifier

oai:conservancy.umn.edu:11299/104813