THE NUMERICAL-SOLUTION OF SOME KINETICS MODELS WITH VODE AND CHEMKIN-II

被引:38
作者
BYRNE, GD [1 ]
DEAN, AM [1 ]
机构
[1] EXXON RES & ENGN CO,ANNANDALE,NJ 08801
来源
COMPUTERS & CHEMISTRY | 1993年 / 17卷 / 03期
关键词
D O I
10.1016/0097-8485(93)80011-2
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
We describe some numerical experiments we have performed in solving chemical kinetics models with CHEMKIN II and two high quality ordinary differential equations (ODE) packages-VODE and LSODE. For this mode of kinetics modeling, we give performance figures for these codes and for three chemical systems of varying size and behavior. In all cases, the Jacobian matrix is internally computed by the ODE solver, owing to the complexity of the models. Although these results are somewhat more conservative than others we have obtained, th y do show that significant performance improvement can be obtained by using VODE instead of LSODE for these problems. We believe that the problems are typical of those that one of us (AMD) has been working with for several years. The main objective of this report is to make kineticists and other practitioners aware of VODE and its potential advantages in such applications. Performance results are given. For example, a small mechanism (17 species and 31 reactions) required 0.48 s CPU time with VODE vs 0.83 s with LSODE. A larger mechanism (135 species and 497 reactions) showed a more dramatic difference: 60 s CPU time for VODE vs 213 s with LSODE.
引用
收藏
页码:297 / 302
页数:6
相关论文
共 9 条
[1]   REDUCED STORAGE MATRIX-METHODS IN STIFF ODE SYSTEMS [J].
BROWN, PN ;
HINDMARSH, AC .
APPLIED MATHEMATICS AND COMPUTATION, 1989, 31 :40-91
[2]   MATRIX-FREE METHODS FOR STIFF SYSTEMS OF ODES [J].
BROWN, PN ;
HINDMARSH, AC .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 1986, 23 (03) :610-638
[3]   VODE - A VARIABLE-COEFFICIENT ODE SOLVER [J].
BROWN, PN ;
BYRNE, GD ;
HINDMARSH, AC .
SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING, 1989, 10 (05) :1038-1051
[4]  
BYRNE GD, 1992, RECENT DEVELOPMENTS IN NUMERICAL METHODS AND SOFTWARE FOR ODES/DAES/PDES, P137
[5]  
BYRNE GD, 1982, NUMERICAL ORDINARY D
[6]  
BYRNE GD, 1973, ACM T MATH SOFTWARE, V1, P71
[7]  
Hindmarsh A.C., 1983, SCI COMPUT, P55
[8]  
HINDMARSH AC, 1980, ACM SIGNUM NEWSLETTE, V15, P10, DOI DOI 10.1145/1218052.1218054
[9]  
KEE RJ, 1989, SAND898009 NAT LAB R