Energy management for automotive power nets

Kessels, JTBA (John) · 2007 · OpenAlex-citations

DOI: 10.6100/ir617399

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This PhD thesis addresses the challenge of optimizing energy management (EM) in automotive power nets, specifically focusing on Hybrid Electric Vehicles (HEVs) and conventional vehicles. Motivated by the rising cost of fossil fuels and the increasing electrical power demand in modern vehicles due to safety, comfort, and hybridization trends, the research aims to minimize fuel consumption. The study identifies that while HEVs offer superior tank-to-wheel efficiency compared to traditional gasoline vehicles, realizing this potential requires intelligent control strategies that integrate concepts such as regenerative braking, engine stop/start, motor assist, and load scheduling. The central problem is developing a causal, on-line EM strategy that mimics global optimal performance without requiring complex computational resources or prior knowledge of future driving cycles, which are limitations of existing dynamic programming and heuristic approaches. The methodology involves developing quasi-static vehicle models for series (S-HEV), parallel (P-HEV), and series/parallel (S/P-HEV) configurations, as well as conventional vehicles. These models account for internal combustion engine fuel maps, electric machine dynamics, battery behavior, and drivetrain losses. The author derives an on-line adaptive EM strategy based on optimal control theory, utilizing a "fuel equivalent" term to balance immediate fuel consumption against battery energy changes. This approach provides both mathematical and physical interpretations for the weighting factors, ensuring charge-sustaining operation. The strategies are validated through simulations using environments like CarSim and evaluated against baseline strategies and global optimal benchmarks. Additionally, the thesis explores the integration of "electronic horizon" data, where statistical road information allows for energy scheduling between road segments, and applies quadratic programming and model predictive control to conventional vehicle cases. The results demonstrate that the proposed on-line adaptive strategies achieve fuel economy performance very close to the global optimal solution, despite not having future driving cycle information. The strategies successfully manage power flow between the internal combustion engine and electric machine, optimizing operating points to avoid inefficient fuel consumption areas. For HEVs, the adaptive control effectively handles the trade-offs between engine idling, motor assist, and battery charging. In the case of conventional vehicles, the application of model predictive control with quadratic programming showed significant fuel savings by optimizing engine operating points. The inclusion of electronic horizon data further improved performance by allowing the system to anticipate road gradients and traffic conditions, enabling more efficient energy scheduling. The significance of this work lies in providing a robust, computationally efficient framework for real-time energy management in automotive applications. By deriving a causal strategy that approximates optimal control without the need for complex optimization routines or perfect future predictions, the thesis offers a practical solution for vehicle manufacturers. The findings imply that significant fuel economy improvements can be achieved through intelligent power net management, contributing to reduced global oil demand and emissions. The research establishes design rules and control laws that are applicable across various vehicle topologies, bridging the gap between theoretical optimal control and practical on-line implementation in automotive engineering.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• eco driving