SWMM-USERS Digest - 31 Jan 2012 to 5 Feb 2012 (#2012-10)

8 views

Skip to first unread message

SWMM-USERS automatic digest system

unread,

Feb 6, 2012, 12:00:44 AM2/6/12

to SWMM-...@listserv.uoguelph.ca

There are 4 messages totaling 788 lines in this issue.

Topics of the day:

1. Pumping in SWMM - timeseries reporting (4)

**********************************************************
* To sign off, email to: list...@listserv.uoguelph.ca *
* In the body of the message type: signoff swmm-users *
**********************************************************

----------------------------------------------------------------------

Date: Mon, 6 Feb 2012 09:36:13 +1000
From: Nick Andrewes <Nick.A...@GHD.COM>
Subject: Pumping in SWMM - timeseries reporting

Hi there,

I am after some advice on the use of pumps in EPA SWMM.

I have been using EPA SWMM to model a storm water scenario that includes
pumping between two open storages. Modelling has been undertaken on a 6
min timestep over 3 years.

I have two variable speed drive 'type 2' pumps operating based on start up
and shutoff depths at the 'source' storage as follows:

Pump 1

Initial status: OFF
Startup Depth: 1
Shutoff Depth: 1.3

Depth (m) Flow (CMS)
0 0
0.999 0
1 0.015
1.2 0.015
1.201 0.03
1.3 0.03

Pump 2

Initial status: OFF
Startup Depth: 1.301
Shutoff Depth: 3

Depth (m) Flow (CMS)
0 0
1.3 0
1.301 0.23
1.4 0.23
1.401 0.43
3 0.43

I am concerned the pumps are not operating correctly based on a time
series comparison of depths at the 'source' storage and pump flow rate.

The pump flow time series shows that pumping generally occurs at the
intended rates when the depth of the 'source' storage is appropriate.
However in some cases the time series indicates no pumping occurs when the
depth time series indicates there should be pumping.

Additionally, when there is zero flow in the pumping time series in some
cases the depth of the 'source' storage appears to reduce at a rate
similar to what would be expected if pumping was occurring. However in
other cases where pump flow is zero and the depth is within pumping range,
the depth of the 'source' storage does not reduce.

Pump 1 and pump 2 are reported in the 'highest flow instability indexes'
table with values of 31 and 1 respectively. I have tried other pump
setups including using a single pump and a force main pipe instead of a
direct pump connection between storages but have not been able to achieve
a more stable setup.

Neither the storages being pumped from or two are reported as having high
continuity errors.

At this stage I am unsure if this is a reporting or modelling issue (or
both) and would appreciate any feedback from people who may have
encountered this before or could suggest anything I might try.

Some additional information is included below:

1. Model numerical precision is 4 decimal places for depth and flow
2. Flow routing continuity error is 0.025 %
3. Pumps are directly connected between storages
4. A control rule is in operation for another pump in the model. The
control rule specifies pump operation based on month of the year. This
does not correlate to when pump 1 and pump 2 appear to be malfunctioning
as per the time series.
5. Other than pumping losses occur through monthly evaporation @ a peak of
approximately 7 mm/day.

Thanks for your help.

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

_____________________
This email and all attachments are confidential. For further important information about emails sent to or from GHD or if you have received this email in error, please refer to http://www.ghd.com/emaildisclaimer.html .
_____________________
This e-mail has been scanned for viruses by MessageLabs.

------------------------------

Date: Sun, 5 Feb 2012 16:57:16 -0700
From: Robert Dickinson <Robert.E....@INNOVYZE.COM>
Subject: Re: Pumping in SWMM - timeseries reporting

Hi Nick,

A few comments:

1. Have you looked at the Pump Summary Table to see how many times the pumps turn on and what percent of the time the pumps are high or low off the pump curve - you can find this table at the very end of the RPT file,
2. You mention using a 3 year simulation which sort of implies you are saving information every 15 minutes or longer and thereby missing most of what happens in your model visually in the graphs.
3. Is your hydraulic time step 6 minutes for the dynamic wave solution?
4. How long do your pumps turn on for the average pump start - this might be a better time step to use.

I would lower your time step, run the model for a few weeks at a very small report time step and then do some node vs pump graphs to satisfy yourself the pumps and node depths are in sync at small report time steps. Once you understand visually how matters stand for a short period simulation you can go back to a 3 year model but realize the report time step is important in understanding your model.

Robert Dickinson
Product Sector Leader
Innovyze Inc.
9340 Pontiac Drive Tel: 813-712-0664
Tampa, Florida USA 33626
robert.d...@innovyze.com
www.innovyze.com

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@LISTSERV.UOGUELPH.CA] On Behalf Of Nick Andrewes
Sent: Sunday, February 05, 2012 6:36 PM
To: SWMM-...@LISTSERV.UOGUELPH.CA
Subject: [SWMM-USERS] Pumping in SWMM - timeseries reporting

Hi there,

I am after some advice on the use of pumps in EPA SWMM.

I have been using EPA SWMM to model a storm water scenario that includes
pumping between two open storages. Modelling has been undertaken on a 6
min timestep over 3 years.

I have two variable speed drive 'type 2' pumps operating based on start up
and shutoff depths at the 'source' storage as follows:

Pump 1

Initial status: OFF
Startup Depth: 1
Shutoff Depth: 1.3

Depth (m) Flow (CMS)
0 0
0.999 0
1 0.015
1.2 0.015
1.201 0.03
1.3 0.03

Pump 2

Initial status: OFF
Startup Depth: 1.301
Shutoff Depth: 3

Depth (m) Flow (CMS)
0 0
1.3 0
1.301 0.23
1.4 0.23
1.401 0.43
3 0.43

I am concerned the pumps are not operating correctly based on a time
series comparison of depths at the 'source' storage and pump flow rate.

Neither the storages being pumped from or two are reported as having high
continuity errors.

At this stage I am unsure if this is a reporting or modelling issue (or
both) and would appreciate any feedback from people who may have
encountered this before or could suggest anything I might try.

Some additional information is included below:

Thanks for your help.

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

------------------------------

Date: Mon, 6 Feb 2012 10:42:16 +1000
From: Nick Andrewes <Nick.A...@GHD.COM>
Subject: Re: Pumping in SWMM - timeseries reporting

Hi Robert,

Thanks for your very fast response!

In response to your comments:

1. The pump summary table indicates that the percent utilised for Pump 1
(low flow pump) is 11.04 % with a total of 221 ML pumped. Pump 2 (high
flow pump) is 0.55% utilised with 180 ML pumped.
The % time off the high curve for Pump 1 is 17.1% and 0% off the low
curve. The % time off the high and low curve for Pump 2 is 0% and 0 %
respectively.

The latter seems consistent with the modelled pump curves (as per my
previous post).

2. The model is reporting on a 6 min time step.
3. The model also includes a pipe network upstream of the storages. It is
setup to run using dynamic wave routing at a 0.5 second timestep.
4. Based on the number of start-ups for pump 1 (17326472) and
considering the time series lengths, this equates to one start up every
5.46 seconds!

I fully agree that care must be taken to understand how the output time
step may portray the modelling results. However in this case my thoughts
were that the reporting time step should be sufficient to capture the
change in the 'source' storage depth and therefore the pumping rate, which
is being used as the pump control based on depth.

Thanks again for your comments/suggestions, based on my response would
you alter anything else with regards to the model setup?

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

From:
Robert Dickinson <Robert.E....@INNOVYZE.COM>
To:
SWMM-...@LISTSERV.UOGUELPH.CA
Date:
06/02/2012 10:59 AM
Subject:
Re: Pumping in SWMM - timeseries reporting
Sent by:
SWMM-USERS <SWMM-...@LISTSERV.UOGUELPH.CA>

Hi Nick,

A few comments:

1. Have you looked at the Pump Summary Table to see how many times the
pumps turn on and what percent of the time the pumps are high or low off
the pump curve - you can find this table at the very end of the RPT file,
2. You mention using a 3 year simulation which sort of implies you are
saving information every 15 minutes or longer and thereby missing most of
what happens in your model visually in the graphs.
3. Is your hydraulic time step 6 minutes for the dynamic wave solution?
4. How long do your pumps turn on for the average pump start - this might
be a better time step to use.

I would lower your time step, run the model for a few weeks at a very
small report time step and then do some node vs pump graphs to satisfy
yourself the pumps and node depths are in sync at small report time steps.
Once you understand visually how matters stand for a short period
simulation you can go back to a 3 year model but realize the report time
step is important in understanding your model.

Robert Dickinson
Product Sector Leader
Innovyze Inc.
9340 Pontiac Drive Tel: 813-712-0664
Tampa, Florida USA 33626
robert.d...@innovyze.com
www.innovyze.com

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@LISTSERV.UOGUELPH.CA] On Behalf Of
Nick Andrewes
Sent: Sunday, February 05, 2012 6:36 PM
To: SWMM-...@LISTSERV.UOGUELPH.CA
Subject: [SWMM-USERS] Pumping in SWMM - timeseries reporting

Hi there,

I am after some advice on the use of pumps in EPA SWMM.

I have been using EPA SWMM to model a storm water scenario that includes
pumping between two open storages. Modelling has been undertaken on a 6
min timestep over 3 years.

I have two variable speed drive 'type 2' pumps operating based on start up

and shutoff depths at the 'source' storage as follows:

Pump 1

Initial status: OFF
Startup Depth: 1
Shutoff Depth: 1.3

Depth (m) Flow (CMS)
0 0
0.999 0
1 0.015
1.2 0.015
1.201 0.03
1.3 0.03

Pump 2

Initial status: OFF
Startup Depth: 1.301
Shutoff Depth: 3

Depth (m) Flow (CMS)
0 0
1.3 0
1.301 0.23
1.4 0.23
1.401 0.43
3 0.43

I am concerned the pumps are not operating correctly based on a time
series comparison of depths at the 'source' storage and pump flow rate.

depth time series indicates there should be pumping.

the depth of the 'source' storage does not reduce.

Neither the storages being pumped from or two are reported as having high
continuity errors.

At this stage I am unsure if this is a reporting or modelling issue (or
both) and would appreciate any feedback from people who may have
encountered this before or could suggest anything I might try.

Some additional information is included below:

approximately 7 mm/day.

Thanks for your help.

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

_____________________
This email and all attachments are confidential. For further important
information about emails sent to or from GHD or if you have received this
email in error, please refer to http://www.ghd.com/emaildisclaimer.html .
_____________________
This e-mail has been scanned for viruses by MessageLabs.

_____________________
This e-mail has been scanned for viruses by MessageLabs.

------------------------------

Date: Sun, 5 Feb 2012 18:59:33 -0800
From: Jonathan Welke <jwe...@KAMLOOPS.CA>
Subject: Re: Pumping in SWMM - timeseries reporting

Nick,

I've always had issues with pumps and storage in SWMM. The larger the storage the more issues. Here are some tips that have worked for me in the past.

Instead of using a variable pumping set up series of fix flow pumps varying start/stop levels to achieve the same thing.Superficially lower you storage so that water levels do not back water any incoming elements.Run the model for the upstream elements to develop the incoming hydrograph to the storage node. Then use this in a separate model of the storage and pumps.
These may work for you, or may not. I'm not sure why they have worked in the past, but suspect backwatering over multiple elements complicated things (even though this should not be an issue with dynamic wave). I'm more of a practical user rather than knowing the intricacies of the engine. So when I find something that works just run with it.

Definitely test it out with some shorter runs as Robert suggested to proof things out.

Good luck!

Jonathan Welke, P.Eng.
Water and Sewer Engineer
City of Kamloops
Direct Line: 250-828-3467

>>> Nick Andrewes 02/05/12 4:42 PM >>>
Hi Robert,

Thanks for your very fast response!

In response to your comments:

The latter seems consistent with the modelled pump curves (as per my
previous post).

Thanks again for your comments/suggestions, based on my response would
you alter anything else with regards to the model setup?

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

From:
Robert Dickinson
To:
SWMM-...@LISTSERV.UOGUELPH.CA
Date:
06/02/2012 10:59 AM
Subject:
Re: Pumping in SWMM - timeseries reporting
Sent by:
SWMM-USERS

Hi Nick,

A few comments:

1. Have you looked at the Pump Summary Table to see how many times the
pumps turn on and what percent of the time the pumps are high or low off
the pump curve - you can find this table at the very end of the RPT file,
2. You mention using a 3 year simulation which sort of implies you are
saving information every 15 minutes or longer and thereby missing most of
what happens in your model visually in the graphs.
3. Is your hydraulic time step 6 minutes for the dynamic wave solution?
4. How long do your pumps turn on for the average pump start - this might
be a better time step to use.

I would lower your time step, run the model for a few weeks at a very
small report time step and then do some node vs pump graphs to satisfy
yourself the pumps and node depths are in sync at small report time steps.
Once you understand visually how matters stand for a short period
simulation you can go back to a 3 year model but realize the report time
step is important in understanding your model.

Robert Dickinson
Product Sector Leader
Innovyze Inc.
9340 Pontiac Drive Tel: 813-712-0664
Tampa, Florida USA 33626
robert.d...@innovyze.com
www.innovyze.com

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@LISTSERV.UOGUELPH.CA] On Behalf Of
Nick Andrewes
Sent: Sunday, February 05, 2012 6:36 PM
To: SWMM-...@LISTSERV.UOGUELPH.CA
Subject: [SWMM-USERS] Pumping in SWMM - timeseries reporting

Hi there,

I am after some advice on the use of pumps in EPA SWMM.

I have been using EPA SWMM to model a storm water scenario that includes
pumping between two open storages. Modelling has been undertaken on a 6
min timestep over 3 years.

I have two variable speed drive 'type 2' pumps operating based on start up

and shutoff depths at the 'source' storage as follows:

Pump 1

Initial status: OFF
Startup Depth: 1
Shutoff Depth: 1.3

Depth (m) Flow (CMS)
0 0
0.999 0
1 0.015
1.2 0.015
1.201 0.03
1.3 0.03

Pump 2

Initial status: OFF
Startup Depth: 1.301
Shutoff Depth: 3

Depth (m) Flow (CMS)
0 0
1.3 0
1.301 0.23
1.4 0.23
1.401 0.43
3 0.43

I am concerned the pumps are not operating correctly based on a time
series comparison of depths at the 'source' storage and pump flow rate.

depth time series indicates there should be pumping.

the depth of the 'source' storage does not reduce.

Neither the storages being pumped from or two are reported as having high
continuity errors.

At this stage I am unsure if this is a reporting or modelling issue (or
both) and would appreciate any feedback from people who may have
encountered this before or could suggest anything I might try.

Some additional information is included below:

approximately 7 mm/day.

Thanks for your help.

Regards

Nick Andrewes
Environmental Engineer | Waterways & Stormwater Service Group

Please consider the environment before printing this email

_____________________
This email and all attachments are confidential. For further important
information about emails sent to or from GHD or if you have received this
email in error, please refer to http://www.ghd.com/emaildisclaimer.html .
_____________________
This e-mail has been scanned for viruses by MessageLabs.

_____________________
This e-mail has been scanned for viruses by MessageLabs.

Our corporate mission is...
MAKING KAMLOOPS SHINE

------------------------------

End of SWMM-USERS Digest - 31 Jan 2012 to 5 Feb 2012 (#2012-10)
***************************************************************

Reply all

Reply to author

Forward

0 new messages