ECOSM Conference, Review of Optimal Design Strategies for HEV
•
0 likes•568 views
This paper presents an overview of the existing approaches and algorithms used for optimal design of hybrid electric vehicles (HEV). It also includes an introduction in various hybrid topologies and examples from different transportation sectors.
![Introduction Optimal design Conclusions
Parametric and structural design optimization
Example 4: Size optimization
Mid-size Passenger Vehicle
Optimize a parallel HEV for
maximum fuel economy
Composite driving cycle
Compare for 4 algorithms
Before
Opt. DIRECT
Simulated
Annealing
Genetic
Algorithms
Particle Swarm
Optimization Constraints:
Fuel Economy
[L/100km] 6,7 5,93 5,83 6,26 6,34
Vehicle Weight [kg] 1683 1635 1656 1694 1690
Power rating of the
EM [kW] 65,9 20,2 21,9 24,2 14,8 10 kW - 80 kW
Power rating of the
ICE [kW] 86 83,1 82,4 95,5 87,1
40 kW - 100
kW
0-60 mph [sec] 18,1 15,5 10,8 11,9 11,1 ≤ 18.1 s
40-60 mph [sec] 7 6,8 5 4,4 4,9 ≤ 7 s
0-85 mph [sec] 35,1 30,6 20,7 21,2 20 ≤ 35.1 s
Bat. No of cells 240 245 311 300 238 150-350
120%
100%
80%
60%
40%
20%
0%
Fuel Economy Vehicle Weight Power rating of
the EM
Power rating of
the ICE
0-60 mph 40-60 mph 0-85 mph Bat. No of cells
Percent Change
Before Opt. DIRECT Simulated Annealing (SA) Genetic Algorithms (GA) Particle Swarm Optimization (PSO)
E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV](https://image.slidesharecdn.com/presentatione-cosmconferenceparisfr2012-141028043217-conversion-gate02/85/ECOSM-Conference-Review-of-Optimal-Design-Strategies-for-HEV-11-320.jpg)
ECOSM Conference, Review of Optimal Design Strategies for HEV
- 1. Introduction Optimal design Conclusions Review of Optimal Design Strategies for Hybrid Electric Vehicles Emilia Silva¸s, Theo Hofman and Maarten Steinbuch1 1Department of Mechanical Engineering Eindhoven University of Technology E-COSM, 23-25 October 2012, Rueil-Malmaison, France E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 2. Introduction Optimal design Conclusions Optimization problem Size Control Motorcycle Vehicle Truck Crane Ship Airplane Fuel Emission Performance Comfort, Handling Cost Topology Application Parameters Targets / Constraints (predicted) operation conditions Technology E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 3. Introduction Optimal design Conclusions Parametric and structural design optimization Example 1: Size optimization PHEV city bus A parallel and a series topology Battery size Convex optimization vs. DP Assumptions: gear and engine on/off state Error vs. no. of variable / computation time E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 4. Introduction Optimal design Conclusions Parametric and structural design optimization Example 1: Size optimization PHEV city bus A parallel and a series topology Battery size Convex optimization vs. DP Assumptions: gear and engine on/off state Error vs. no. of variable / computation time E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 5. Introduction Optimal design Conclusions Parametric and structural design optimization Example 2: Size and technology optimization Mid-sized to large family car (torque-assist) Use, initially, DP to find the optimal driving strategy Build a RB method - sizing of the CE, EM (8 driving cycles) Minimize the CO2 emissions Adjusted Gear Ratio CO2 emissions, 1% for the RB method E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 6. Introduction Optimal design Conclusions Parametric and structural design optimization Example 3: Size and topology optimization Submarine 4 sea mission scenarios Multiobjective Genetic Algorithms Five objective functions Max. propeller efficiency Max. electric motor efficiency Min. electric motor size Min. total energy consumption Max. steam turbine efficiency 8% improvement in energy consumption Electric Motor Electric Motor Electric Motor Gearbox Steam Turbine Electric Motor Gearbox (a) (b) (c) Source: http://www.guardian.co.uk E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 7. Introduction Optimal design Conclusions Parametric and structural design optimization Example 3: Size and topology optimization Submarine 4 sea mission scenarios Multiobjective Genetic Algorithms Five objective functions Max. propeller efficiency Max. electric motor efficiency Min. electric motor size Min. total energy consumption Max. steam turbine efficiency 8% improvement in energy consumption Electric Motor Electric Motor Electric Motor Gearbox Steam Turbine Electric Motor Gearbox (a) (b) (c) Source: http://www.guardian.co.uk E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 8. Introduction Optimal design Conclusions Parametric and structural design optimization Example 3: Results E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 9. Introduction Optimal design Conclusions Parametric and structural design optimization Example 4: Size optimization Mid-size Passenger Vehicle Optimize a parallel HEV for maximum fuel economy Composite driving cycle Compare for 4 algorithms Engine Motor/ Generator Transmission Electric Power Controller Battery Electrical connection Mechanical connection E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 10. Introduction Optimal design Conclusions Parametric and structural design optimization Example 4: Size optimization Mid-size Passenger Vehicle Optimize a parallel HEV for maximum fuel economy Composite driving cycle Compare for 4 algorithms HWFET (Highway Fuel Economy Test) FTP-75 (Federal Test Procedure) E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 11. Introduction Optimal design Conclusions Parametric and structural design optimization Example 4: Size optimization Mid-size Passenger Vehicle Optimize a parallel HEV for maximum fuel economy Composite driving cycle Compare for 4 algorithms Before Opt. DIRECT Simulated Annealing Genetic Algorithms Particle Swarm Optimization Constraints: Fuel Economy [L/100km] 6,7 5,93 5,83 6,26 6,34 Vehicle Weight [kg] 1683 1635 1656 1694 1690 Power rating of the EM [kW] 65,9 20,2 21,9 24,2 14,8 10 kW - 80 kW Power rating of the ICE [kW] 86 83,1 82,4 95,5 87,1 40 kW - 100 kW 0-60 mph [sec] 18,1 15,5 10,8 11,9 11,1 ≤ 18.1 s 40-60 mph [sec] 7 6,8 5 4,4 4,9 ≤ 7 s 0-85 mph [sec] 35,1 30,6 20,7 21,2 20 ≤ 35.1 s Bat. No of cells 240 245 311 300 238 150-350 120% 100% 80% 60% 40% 20% 0% Fuel Economy Vehicle Weight Power rating of the EM Power rating of the ICE 0-60 mph 40-60 mph 0-85 mph Bat. No of cells Percent Change Before Opt. DIRECT Simulated Annealing (SA) Genetic Algorithms (GA) Particle Swarm Optimization (PSO) E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 12. Introduction Optimal design Conclusions Parametric and structural design optimization Challenges and trends in hybrid power-trains design Challenges Large design space High computation time Discrete/Continuous variables Non-convexity character of the problem Trends Multi-objective optimization More automated approaches in searching for the global optimum E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 13. Introduction Optimal design Conclusions Parametric and structural design optimization Challenges and trends in hybrid power-trains design Classification of the hybrid power train optimization algorithms mostly used for optimal design Global Deterministic Dynamic Programing DIRECT Genetic algorithms Derivative free Multi-objective Genetic Algorithms Particle Swarm Optimization Simulated Aanealing Sequential Quadratic Programing Convex optimization (subgradient ) Convex optimization (gradient) E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 14. Introduction Optimal design Conclusions Conclusions Summary The optimum power train design is desired for fuel consumption minimization, emissions and dynamic performance. The complexity of the design space makes the design difficult. Future work Develop and analyze an optimization framework for commercial vehicles under given work conditions. E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV
- 15. Introduction Optimal design Conclusions Conclusions Thank you! E. Silva¸s (e.silvas@tue.nl) Review of Optimal Design Strategies for HEV