Page 1 of 1

query regarding errors in model

Posted: Wed Nov 16, 2016 6:54 pm
by GOKILA VANI A R
hello,
i have made the model for water demand estimation. i am some errors like floating point errors. what is floating point error? how to fix it?
anybody can help me? here am attaching the model.

Re: query regarding errors in model

Posted: Wed Nov 16, 2016 7:51 pm
by Administrator
Start with the population and get that working properly first.

You've got an error in the equation for "population". It should be "INTEG ( births -deaths), but you're adding the population in each time step as well. This is shown by the units error.

Then the birth rate is a little strange as well. It's 2.4 people per year, per member of the population. That seems very high to me. Same issue with the death rate.

I think if you fix all the unit errors, things might start to work for you.

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 6:23 am
by GOKILA VANI A R
Hello,
i have tried out with the solution on which you advised. it has corrected over the population alone. it shows the floating point error in rainwater consumption rate(rainwater - RW). how to fix it? anybody can help me to arrive the solution for total demand?

i hope your suggestion will be helpful to achieve the expecting demand.
i thank you in advance

gokila vani

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 6:26 am
by GOKILA VANI A R
Hello,
i have posted my model. can anybody check the equations which i have given in model?
if you find any mistakes then please help me to correct that.

thanks in advance

gokila vani

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 7:59 am
by Administrator
Can you upload the model with the corrected population calculations?

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 2:37 pm
by GOKILA VANI A R
hello,
here am uploading the corrected population model. But still am getting error in rainwater consumption rate. could you please the correct errors in model? and can u help to arrive the actual demand?
Thank you in advance sir.
gokila vani

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 2:39 pm
by GOKILA VANI A R
hello,
here am uploading the corrected population model. But still am getting error in rainwater consumption rate. could you please the correct errors in model? and can u help to arrive the actual demand?
Thank you in advance sir.
gokila vani

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 2:48 pm
by Administrator
I don't think the population part of the WD model is correct. The population rate is "1", so for each member of the population (both male and female), they will have one child per year.

Total HH demand should also not be a level.

HH with basic supply is also wrong, you are adding in "HH with basic supply" each time step. The equation should be "netgrwth of HH with BS-connection".

"netgrwth of HH with BS" probably should not go negative.

Have a go at fixing all the unit errors, these clearly show where things are wrong. Also look at the tabular values for population and go forward from there.
WD.mdl
(13.56 KiB) Downloaded 306 times

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 3:09 pm
by GOKILA VANI A R
Hello,
Yes , i have checked over that model. i am getting negative values for HH with basic supply for future years. and why am not able to forecast for 30 years? it shows only for 7 years. Sir, please i request to correct all the errors occurs in the model. i have tried many times. i am not able to correct all the errors in model.
Thank you in advance

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 3:28 pm
by Administrator
I cannot correct the errors, I don't even know what the model is supposed to be showing (or any of the terminology). For example, I assume HH is households, but it's just a guess.

The population goes from 1.4M to 2.8M in 7 years, is this correct?

This leads on to "total no of HH". This in turn is used to calculate "total no of informal HH". This goes negative at time = 3. Is the equation for that parameter correct?

I'm sure if you start with the population and follow it forwards (as I have done), you'll find the errors. The units check highlights all of them.

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 4:20 pm
by GOKILA VANI A R
Hello,
The expansion for HH is household only which you assumed correctly. This model has been made to find the water demand for the year of 2040. i have prioritized the factors then i formed the model. The base year data i have taken is 2011. i have done the population estimation using some other manual methods. For 7 years ,the answer arrived using this vensim is almost correct. like wise, i need to forecast the water demand for 2040. The equation i have given for total no of HH is correct. but it gives negative answer from 3rd year.. The objective of the water demand model, i have mentioned in the above points. I hope,this will be helpful to solve the errors. If you can correct the errors in model then please correct it and send me to my mail id. gokila.1086@gmail.com :)

Re: query regarding errors in model

Posted: Thu Nov 17, 2016 4:56 pm
by Administrator
What about "total no of informal HH"? As I mentioned in the last post, this goes negative at time = 3. Is the equation for that parameter correct? As I said, I'm just guessing that there is an error here, I have no knowledge of this industry/area.

You need to spend some time looking a the numbers from here and see if they make sense. If not, correct and then move onto the next part of the model (the error for "total no of informal HH" leads on to "HH with basic supply").

Re: query regarding errors in model

Posted: Sat Nov 19, 2016 6:26 pm
by GOKILA VANI A R
Hello,
i have one more issue. i have cleared the errors as much as possible. i have some errors regarding the function used INTEG in equation. if possible ,clarify this error. and i have tried to do simulation like floating point error on time which i mentioned for rainfall. it affects the projection. here i attach the mail. anybody go through and help to clear the errors.

thanks in advance

Re: query regarding errors in model

Posted: Mon Nov 21, 2016 11:09 am
by Administrator
First problem, you've got rainfall = seasonal rainfall/Time. At time = 0, this causes a floating point error.

The next problem is caused by "RW storage rate" going negative.

Have a look at "HH with basic supply", is this correct? This has a knock on effect on everything else. The equation is definitely incorrect, it should be "integ(netgrwth of HH with BS-connection,initial hh with basic supply)".

As I keep saying, the unit errors in your model highlight all the problems. Once you've corrected the unit errors, check the values that you are calculating to make sure they are reasonable.

Re: query regarding errors in model

Posted: Tue Nov 22, 2016 7:21 am
by GOKILA VANI A R
Hello,
i have given the time period for 100 years. why it takes the time '0'? what's there to change in time? is that equation correct?

Re: query regarding errors in model

Posted: Tue Nov 22, 2016 8:04 am
by Administrator
i have given the time period for 100 years. why it takes the time '0'? what's there to change in time?
You can change the constant INITIAL TIME or FINAL TIME.
is that equation correct?
I don't think so. But I don't know what you are trying to do here so cannot comment any further.