Abstract
A hybrid evolutionary algorithm, consisting of a genetic algorithm (GA) and particle swarm optimization (PSO), is proposed. Generally, GAs maintain diverse solutions of good quality in multi-objective problems, while PSO shows fast convergence to the optimum solution. By coupling these algorithms, GA will compensate for the low diversity of PSO, while PSO will compensate for the high computational costs of GA. The hybrid algorithm was validated using standard test functions. The results showed that the hybrid algorithm has better performance than either a pure GA or pure PSO. The method was applied to an engineering design problem - the geometry of diesel engine combustion chamber reducing exhaust emissions such as NOx, soot and CO was optimized. The results demonstrated the usefulness of the present method to this engineering design problem. To identify the relation between exhaust emissions and combustion chamber geometry, data mining was performed with a self-organising map (SOM). The results indicate that the volume near the lower central part of the combustion chamber has a large effect on exhaust emissions and the optimum chamber geometry will vary depending on fuel injection angle.
Original language | English |
---|---|
Pages (from-to) | 1-16 |
Number of pages | 16 |
Journal | Engineering Optimization |
Volume | 40 |
Issue number | 1 |
DOIs | |
Publication status | Published - 2008 Jan |
Keywords
- Combustion chamber design
- Data-mining
- Hybrid evolutionary algorithm
- Low exhaust emission
- Self-organising map (SOM)
ASJC Scopus subject areas
- Computer Science Applications
- Control and Optimization
- Management Science and Operations Research
- Industrial and Manufacturing Engineering
- Applied Mathematics