Application of genetic programming to forecast streamflow

[qays707]

Good day, friends. I am currently working on a research which is based on the forecasting of streamflow using genetic programming. My input variables will be rainfall, evaporation and temperature while the output variable will be the resulting streamflow.
I would be glad if I can get help on how to go about it.
Thanks in advance.

 

[IRstuff]

Student posting is not allowed

TTFN
faq731-376
Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[qays707]

I am not s student. It's a freelance project. I have already carried out the same work using artificial neural networks. I am just curious and want to understand genetic programming.
Thanks.

 

[IRstuff]

Seems to me to be a less than ideal choice for this sort of algorithm, given the variability of the basic forecasts to begin with. If you already know the actual rainfall, etc., then you probably already have the actual streamflow. The statistical probability of predicting the streamflow from an advanced rainfall prediction alone seems dubious, since streamflow would seem to me to partly dependent on what has happened in the past, i.e., if there had been a big storm recently, the ground could still be saturated and runoff would be higher, which would drive streamflow. That's not something that comes directly out of a single storm's rainfall. Now, if you could add in a history component and/or a ground cover component, that would seem to be more interesting.

In California, that's potentially a big issue, due to changes in ground cover from fires, which can drastically alter the runoff.

TTFN
faq731-376
Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[qays707]

Thanks a lot. I will try to find more variables.

 

[BigInch]

I'd think that Neural Networks would be a better way to go. If you have previous data, it will learn from that and make future predictions based on changes to the input.

 

[zelgar]

In addition to the items you've listed there are a lot of other factors to consider (e.g., , soil types & properties, run-off coefficients (depending on land use), drainage area, channel characteristics, topography (e.g., slope), seasonal changes (e.g., soil saturation, crop growth, frost/frozen soil), items that could effect flow (e.g., bridges, obstructions, rapids))