A new stochastic algorithm for proton exchange membrane fuel cell stack design optimization
Source abstract
Abstract
Abstract This paper develops a new stochastic heuristic for proton exchange membrane fuel cell stack design optimization. The problem involves finding the optimal size and configuration of stand-alone, fuel-cell-based power supply systems: the stack is to be configured so that it delivers the maximum power output at the load's operating voltage. The problem apparently looks straightforward but is analytically intractable and computationally hard. No exact solution can be found, nor is it easy to find the exact number of local optima; we, therefore, are forced to settle with approximate or near-optimal solutions. This real-world problem, first reported in Journal of Power Sources 131 , poses both engineering challenges and computational challenges and is representative of many of today's open problems in fuel cell design involving a mix of discrete and continuous parameters. The new algorithm is compared against genetic algorithm, simulated annealing, and (1+1)-EA. Statistical tests of significance show that the results produced by our method are better than the best-known solutions for this problem published in the literature. A finite Markov chain analysis of the new algorithm establishes an upper bound on the expected time to find the optimum solution.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Fuel Cells and Related Materials · Electrocatalysts for Energy Conversion · Advanced Battery Technologies Research
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (15)
Show source-provided references
1. An Adaptive Penalty Method for Genetic Algorithms in Constrained Optimization Problems
2. Bean · 1992
3. Chakraborty · Energy · 2009 · 34 · 74010.1016/j.energy.2009.02.012 ↗
4. Chakraborty · Energy · 2012 · 40 · 38710.1016/j.energy.2012.01.039 ↗
5. Friede · IEEE Transactions on Power Electronics · 2004 · 19 · 123410.1109/TPEL.2004.833449 ↗
6. 2000
7. M. Gen, R. Cheng, A Survey of Penalty Techniques in Genetic Algorithms, Proceedings of IEEE International Conference on Evolutionary Computation, 1996, 804–809.10.1109/ICEC.1996.542704 ↗
8. Johnson · 2000
9. Kemeny · 1960
10. Kirkpatrick · Science · 1983 · 220 · 67110.1126/science.220.4598.671 ↗
11. Larminie · 2003
12. Michalewicz · 1996
13. Michalewicz · 2000
14. Mohamed · Journal of Power Sources · 2004 · 131 · 14210.1016/j.jpowsour.2004.01.012 ↗
15. Tessema · IEEE Transactions on Systems, Man, and Cybernetics – Part A · 2009 · 39 · 56510.1109/TSMCA.2009.2013333 ↗