Is there any systemic way of determining a better guess while solving DAEs using ODE15s?

1 次查看（过去 30 天）

显示更早的评论

Dursman Mchabe 2018-8-15

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/415002-is-there-any-systemic-way-of-determining-a-better-guess-while-solving-daes-using-ode15s

评论： Star Strider 2018-8-16

simultaneousEquations1.m

When running the attached code (or code on the dropbox link):

https://www.dropbox.com/s/7sz6gli891k0u4l/simultaneousEquations1.m?dl=0

I get the error message:

"Error using daeic12 (line 166) Need a better guess y0 for consistent initial conditions.

Error in ode15s (line 310) [y,yp,f0,dfdy,nFE,nPD,Jfac] = daeic12(odeFcn,odeArgs,t,ICtype,Mt,y,yp0,f0,...

Error in simultaneousEquations1 (line 79) [t,y] = ode15s(@(ti,yi)revisedModelode(ti,yi),tspan,y0,options);

>> "

Is there any systemic way of determining a better guess?

6 个评论
显示 4更早的评论隐藏 4更早的评论

Dursman Mchabe 2018-8-16

Hello Torsten, I have checked the equations, at this stage I con not see the mistake yet, probably because I developed them. However the original equations were taken from two papers:

(1) https://pubs.acs.org/doi/abs/10.1021/ie9709446 Equations 41 to 50.

(2) https://www.sciencedirect.com/science/article/pii/S0378382010002547 Equations 1 to 54.

It usually takes me a while to spot my own mistakes. However I'm also trying different approaches as well. E.g. I have split the set of equations into two:

(a) I solve 7 unknowns from 7 ODEs (ODE45).

(b) Then I also solve 9 unknowns from 9 algebraic (fsolve).

The equations are solved in time-steps. Nine variables (taken from fsolve solution) are kept as constant parameters while solving the differential equations. Similarly, 7 variables from ODE45 is used as input constant parameters in the next time step of the fsolve. The loop continues like that.

I am currently doing it manually, and comparing the outcome with the experimental results.

I hope I will be able to loop the two functions using matlab.