View Full Version : Field Data Collection for Calibration

Darrell D
March 2, 2017, 08:53 AM
Our utility uses InfoWater to model our reclaimed water distribution system. We recently completed the installation of a SCADA system on the reclaimed system which will help in calibrating the model. In a potable water system, fire flow tests can be used for calibration but there are no hydrants available in the reclaimed system. Has anyone here had experience in using some type of data recorders in the field to collect pressure and flow information and then used it for model calibration? If so, can you please share as much detail as you're willing to and your experience with the devices.

Patrick Moore
March 6, 2017, 10:06 AM

One thing you can always generally do is to use the available SCADA data to compare known Flows ( usually at pump station discharges) known pressures, often at Pump station discharge and in some cases pump station suction pressures or remote pressure locations if known), and lastly comparing Tank levels. Most comparisons are generally done against and EPS scenario with a 24 hour period being the most common used to compare and this generally done on a day as close as possible to the Maximum day usage.

You will need to set up your model to have diurnal curve water usage patterns, demands, and controls based on the day you are comparing to best use for calibration purposes. If you find you wish to have additional remote pressure or flow recorders added, this would best supplement the available SCADA data. Generally, most calibrations find that the known SCADA values are often sufficient, but if a key sensor is down or there is an area with uncertainty supplementing available SCADA data with a data logger is always an option.

Since this system is likely installed in the last 15 years it is probably using pipeline materials that don't corrode (only unlined Cast Iron and steel pipes generally are the only materials expected to have internal corrosion). In that case you likely may not need to make c-factor adjustments to the pipelines in the model. Generally you may find that the EPS calibration will help validate the demand allocation, the system diurnal curves, the model pump curves used, and the system control set points for pumps and other controllable features and it is only those features that you may generally adjust for calibration purposes. Just make sure to only adjust something if you know is justifiable to change. In other words, you want to only change factors that have uncertainty in them and make sure whatever you change you can justify making that change.

One final note, since the SCADA system is newly installed, you may wish to wait to calibrate until you can get data corresponding to when the high usage period for that system occurs. This may mean waiting a few months until you have a whole month of peak water usage days you can choose form and then selecting the best day to calibrate to from that period. Just a thought to keep in mind as the higher the usage, the more stressed the system will be and the more likely you may identify differences in the model and field data so that you can make adjustments.

Hope this helps give you at least a few ideas on what you can do with what you currently have.

Patrick Moore