Calibrate Guidelines AWWA

Calibration Guidelines for Water Distribution System Modeling

byEngineering Computer Applications Committee

Hydraulic network simulation models are widely used by planners, water utilitypersonnel, consultants, and others involved in the analysis, design, operation, andmaintenance of closed-conduit hydraulic systems. Quite possibly the largest applicationof hydraulic network models lies with the municipal water supply industry. The resultsof network models have been used to assist in long-range master planning, short-termproject design, fire flow studies, daily operations, emergency response, energymanagement, rehabilitation, troubleshooting, operator training, and water qualityinvestigations. Clearly multi-million dollar decisions can be and have been based on theresults provided by hydraulic network models. Consequently the results from the modelmust bear close resemblance to the actual performance of the hydraulic system. In otherwords, the computer model must be calibrated. The purpose of this paper is to presentguidelines for network model calibration. The degree of accuracy needed for modelcalibration will be discussed within the context of the intended use of the model. Forexample, a higher degree of calibration is necessary for a model that is used to examinedaily operations than is needed for a model used in long-range planning studies. Thepaper will also discuss calibration approaches and testing procedures that can be used toaid in the calibration effort.

Copyright 1999 American Water Works Association.Reprinted from Proceedings of AWWA 1999 ImTech Conference, by permission.

Calibration Guidelines for Water Distribution System Modeling

By

Engineering Computer Applications Committee

Introduction

A computer model of a water distribution system is a valuable tool which can assistengineers and planners in analyzing the hydraulic performance of water delivery systems.Computer models of water supply systems are constructed for a variety of uses including:

1) Identifying system deficiencies,2) Analyzing the impacts of proposed development or long-term growth within the

system,3) Determining the ability of the system to deliver adequate fire flows,4) Sizing pipes, selecting pumps, locating tanks,5) Evaluating operating strategies,6) Assessing pipeline rehabilitation methods,7) Developing emergency response plans,8) Training new operators,9) Estimating the quality of water throughout the system.

A computer model of a water distribution system is a mathematical representation of areal physical system. Data describing the physical characteristics of the system as well asloading conditions and boundary information are supplied to a computer program thatsimulates the behavior of the real system. Physical data includes information such aspipe length, diameter and roughness, pump characteristics and minor loss coefficients.Loading conditions reflect demands that are placed on the system while boundaryinformation describes reservoir and tank levels, valve settings and pump on/off status.The computer program uses this information to develop a set of equations which is thensolved by the program. The solution to the set of equations provides information onpressures and flows throughout the system. Information on pressures and flows can thenbe used to aid in decision-making.

It is not unusual for the results provided by a computer simulation model to be used incapital projects involving several million dollars. As a result, it is imperative that theresults provided by the model bear close resemblance to reality. If this is not the case,then the results provided by the model will be of limited value. Consequently, the modelmust be calibrated. Computer model calibration can best be defined as the process ofadjusting data describing the mathematical model of the system until observedperformance, typically pressures and flow rates, are in reasonable agreement withcomputer-predicted performance over a wide range of operating conditions.


There have been a number of papers published on the subject of hydraulic network modelcalibration. These papers generally discuss errors that contribute to discrepanciesbetween observed and computer predicted performance, provide detail on data collectionand testing methods and provide some insight into what should be adjusted to achieve asuitable match. However a void exists in the United States network modeling communitydealing with an acceptable accuracy of the calibration effort. In other words, when is themodel calibrated well enough?

The purpose of this paper is to provide a little deeper background on the various sourcesof error that will produce differences between measured and computed systemperformance. Armed with a strong understanding of the errors inherent in a networkmodel, we will briefly discuss the level of effort that can be involved in performing acomprehensive network calibration. Finally we propose some calibration guidelines andattempt to establish some criteria indicating a suitable level of calibration based upon theintended use of the model. The purpose of this paper is not to establish standards formodel calibration. Rather members of the Engineering Computer ApplicationsCommittee hope that the proposed criteria presented here will foster meaningfuldiscussion among the United States modeling community regarding the need and validityof calibration standards.

Sources of Error

Before discussing how accurately a computer model should be calibrated, it is valuable toclosely examine why a computer model might not exactly match the field performance ofa real hydraulic system. With a computer simulation model we are trying to reproducethe behavior of a real system that acts continuously over space and time. We do so bysupplying data that depicts the physical characteristics of the system and by providinginformation that, to some degree, represents the continuous loads (system demands)placed on the system. Calibrating a hydraulic network model involves more than justadjusting pipe roughness values and nodal demands. True model calibration is achievedby adjusting whatever should be adjusted within the model until a reasonable agreementbetween model-predicted behavior and actual field behavior is obtained.

Errors in Input DataIn computer modeling, any data that is supplied to the model is a candidate foradjustment. There are two sources of error that can be directly associated with input data:1) typographical errors and 2) measurement errors. Generally speaking, typographicalerrors are more easily corrected than measurement errors assuming, of course, that theycan be identified. An example of a typographical error would be typing in a value of2250 ft for a 250 ft pipe segment. A measurement error, on the other hand, might occurbecause of the limited precision of measuring devices combined with the scale used onsystem maps. For instance, all else being equal a length measurement on a map having ascale of 1”=50’ would have more precision than a length measured from a map having ascale of 1”= 2000’.


A critical piece of information required of all computer simulation models is the diameterof system pipes. Many times modelers ask the question “Which pipe diameter should beused – the actual diameter or the nominal diameter?” Nominal diameters have values like6”, 8”, 12”, etc. The actual pipe diameter must be measured in the field. For older pipessuch as unlined cast iron it is likely that the actual internal diameter will vary along thelength of pipe.

Consider the pipe shown in Figure 1. Notice the build-up, called turburculation, on theinterior pipe walls. The actual diameter of this pipe might be closer to 5 ½” or 5 ¼”instead of 6”. Because the build-up on the pipe walls is totally random and irregular, it ishighly doubtful that the actual diameter, regardless of what it is, will remain the samethroughout the pipe length. Generally speaking it is usually best to use the nominaldiameter in a computer model and adjust pipe roughness values to achieve calibration.

Unknown Internal Pipe Roughness ValuesIn the paragraph above it was stated that it is usually best to supply the nominal pipediameter and adjust pipe roughness values until a suitable match is obtained. But whatroughness values should be used? There are a number of tables in various texts thatprovide estimates of pipe roughness, usually Hazen-Williams C-Factors, as a function ofpipe material, size and age. However pipe roughness is also a function of water quality[1]. In other words, there is no guarantee that the internal roughness of a 12” 40-year oldunlined cast iron pipe in New York City will be the same as the internal roughness of a12” 40-year old unlined cast iron pipe in Seattle.

Simulation models require a value for pipe roughness for each individual pipe segment.This can result in quite a bit of information if one considers that it is not unusual for amodel of a moderately sized system to contain 500-1,000 pipes with larger systemsincorporating 2,000 or more pipelines. Certainly observed pressures can point themodeler in the correct direction with regard to pipe roughness values. Nonethelesscompensating errors could cause incorrect pipe roughness values to be used.

Consider the simple parallel pipe system shown in Figure 2. Suppose that pressuremeasurements have been taken at the node on each end of the pipe segment and the flowthrough the system is known. The unknowns are the internal pipe roughness values foreach pipe. Table 1 shows the results of a simple analysis performed on this system.Column 1 represents an assumed internal roughness for Pipe #1. Column 2 is theresulting flow due to the measured head loss, the pipe characteristics and the assumedroughness. Column 3 is the flow in Pipe #2 assuming a total system flow of 1,350 Gpm.Finally column 4 represents the pipe roughness in Pipe 2 that is due to the head loss, thepipe characteristics and the pipe flow.

Clearly there are multiple values of pipe roughness in pipes 1 and 2 that produce the samehead loss across the system. So the real question becomes which is the correct set of piperoughness values? The only way this question can be answered is to measure the flow inone of the pipes. With the flow known the correct roughness values can be established.Frequently in the United States actual pipeline flows throughout the system are not


measured in support of model calibration. Instead only system pressures at a severalselected locations are measured.

The problem above illustrates the complexity for a simple two pipe system. The problemcomplexity is compounded in real systems. For the system shown in Figure 2, let’sassume that the flow into the system is not known and that the pressure at the upstreamnode is also unknown. Then the number of combinations of flow, pipe roughness andupstream pressure that would produce a pressure of 47 psi at the downstream node is verylarge. Now consider that there are many parallel paths that water may take in realsystems. With the above illustration it is clear how compensating errors can result inincorrect model parameters even though an exact match between computer predicted andobserved values is obtained.

Effect of System DemandsAlthough simulation models require that water use be applied at a single point called anode, in reality water use occurs along the entire length of a pipe as shown in Figure 3.Water use is assigned to nodes because this approach simplifies the complexity of themodeling problem. In Figure 3 the water use associated with the eight homes closest toJ-23 was assigned to this node. Likewise, the water use for the 10 homes closest to J-24is assigned to junction node J-24.

Grouping or lumping water use at junction nodes instead of placing it at the actuallocation where water is withdrawn from the system will produce some differencesbetween computer predicted and actual field performance. The differences should berelatively minor assuming, of course, that the customer demand assigned to a particularjunction node is not located far away from the node. A bigger reason for the discrepancybetween computer predicted and field measured results lies with the spatial and temporalnature of water use.

Water use within a home, and to a lesser extent, a commercial establishment is mostlikely stochastic or random. In other words, one cannot predict with absolute certaintywhen and how much water will be used at a particular residence. Perhaps the closest onecan get to certainty in residential water use lies with those homes that have automaticlawn irrigation systems designed to operate at the same time each day. If one were tometer the instantaneous water use in a specific residence the usage might look somethingthat shown in Figure 4. Industrial water use has a tendency to be a somewhat lessvariable than residential or commercial usage. Any difference between the actual wateruse in a distribution system and the estimates of that use supplied to a computer modelwill generate errors between computer predicted and field measured performance.

Errors in System MapsWater system maps are the primary source of model data for the physical characteristicsof many water distribution systems. Pipeline lengths and diameters are typically takenfrom water system maps. The presence and types of fittings and appurtenances thatcontribute to minor losses can also be found from system maps. Finally the layout of thesystem and the pipe/node connectivity is usually determined from the system maps.


Water system maps come in a variety of formats. Some systems may have very accurate,nicely detailed maps generated using sophisticated CAD or GIS systems. The watersystem maps of other utilities may consist of a set of rolled-up plans that have not beenunrolled in several years even though some system improvements have taken place. Forsome systems, the system caretaker may be the sole individual responsible for recordkeeping including maintaining the system maps – that is if they even exist! Indeed, it isentirely possible that the extent of system mapping for some systems may be theknowledge that is kept in the head of the system caretaker.

As changes to a system are made then the system maps should be modified to reflect thechanges that have been made. Most water utilities are very good at making these changesthough some are not. Despite the best efforts of the mapping department, sometimessystems grow at a faster pace than the mapping department can make the changes.Obviously the accuracy of system maps can have a pronounced effect on the accuracy ofa model calibration effort.

Node ElevationsFrequently in hydraulic network simulation ground elevations will be used to findpressures at locations throughout the hydraulic network. For most systems, pipes areburied at a uniform distance below the ground surface usually anywhere between 3-6 ftdeep (1.3-2.6 psi). The pressure difference associated with using the ground level versusthe actual pipe centerline elevation is usually not that great. Moreover, it is easier togather elevation data from topographic maps than it is to scour through As-Builtdrawings to find the actual elevations.

When pressure measurements are taken for model calibration they are usually taken atfire hydrants or at some other direct connection to the water system. The elevation of thepressure gauge at the hydrant can be several feet higher than the ground elevation orseveral feel lower than the ground in the case of a meter vault. For calibration purposesthis elevation difference can make a difference. Therefore one should always use theelevation of the pressure gauge, that is, the location where the pressure measurement istaken when calibrating a hydraulic network model.

Effect of TimeIf the hydraulic simulation is an extended period simulation (EPS), then the calibrationmust also consider adjustments that should be made for time-varying conditions.Generally this will mean adjusting model parameters until there is a reasonableagreement between computer predicted and observed tank water levels. In addition totank water levels, time-varying pressures and pipeline flows can also be compared.

Many times information on tank water levels will be obtained from circular charts such asthe one shown in Figure 5. Notice that there can be fairly substantial changes in tankwater level over a short period of time. This is especially true for small storage tankssuch as standpipes or if the tank capacity is too small for the system it is supplying. If toolarge a time step is used in hydraulic calculations then the true time-varying behavior of


the system between time intervals may not be captured even though a suitable agreementis obtained at individual time steps.

Model DetailWhen a computer model of an existing system is constructed, quite often a simplified orskeletal version of the system will be developed instead of one which contains every pipein the system. A skeletal version of the system is one that does not include smalldiameter pipes or does not include lines that have an insignificant influence thehydraulics of the overall system. A skeletal model is one which is manageable and yetaccurate – when properly calibrated. Interestingly, Cesario, et al state that the economicsof model construction make a strong case for including all pipes in the mathematicalmodel [2].

Although skeletalization can have an impact on the calibration process, the effects ofskeletalization on calibration are, for the most part, unknown. If one is having difficultyin achieving a reasonable calibration with a highly skeletal system, then perhaps it maybe necessary to include more lines in the model. For example, omitting a dense grid ofsmall diameter mains from a model simply because of their size may be inappropriate if,as a group, they present a sizeable hydraulic impact on the system.

Geometric AnomaliesEven if high quality information on the physical attributes of the system is available andgood estimates of nodal demand is provided to the simulation model, there can still bedifferences between computer predicted and observed performance. Anomalies in thegeometry of the system are usually to blame in these cases. For example, suppose thatwe have a condition where two pipes cross one another as illustrated in Figure 6. Fromthe plan view it may appear that these two lines are connected. The cross-section viewshows otherwise. Obviously the model will not be calibrated if these pipes were assumedto be connected.

Another issue related to geometric anomalies that could cause differences betweenobserved and predicted behavior lies with system isolation valves. Most of the time incomputer models gate valves, butterfly valves, etc. are assumed to be fully open. In somesystems this may not be the case. If one has to use unrealistically low roughness values toobtain a suitable calibration then perhaps there are closed or partially closed isolationvalves in the system.

Outdated Pump Characteristic CurvesIf there are pumps in the system then it is necessary to supply information on the pumpcharacteristics. Most hydraulic models use some type of curve fit using three or morepoints from the actual pump head-discharge relationship to reproduce the curve. Errorsin the numerical curve fit could contribute to discrepancies. However the more likelycause of error for pumps is due to old or outdated pump curves. Over time impellers canwear and change the characteristics of the pump. It is possible that new impellers mayhave been installed on a pump without the pump curve being updated to reflect the


change. Clearly pump curves used in hydraulic simulation should represent the in-situpump characteristics of the unit.

Not only can inaccurate pump curves cause a discrepancy between computer predictedand observed system performance, but errors in any boundary element can causedifficulty with the calibration. These boundary elements include regulating valvesettings, tank levels and pressure zone boundaries. Clearly an incorrect setting forpressure reducing valve will produce pressures in the computer model that areinconsistent with field measured values.

Poorly Calibrated Measuring EquipmentThere are many ways calibration data is gathered including direct measurement usingpressure gauges and flow recording devices. Many of today’s water supply systems haveSupervisory Control and Data Acquisition (SCADA) systems that can provideinformation on tank levels and possibly pressures and flows at selected locations.However in the United States pressure and flow measurement using SCADS systems isgenerally limited to pump stations and boundaries with other systems.

Whenever a measurement is taken we must be sure that the equipment or device that wasused to take the measurement is itself calibrated. Pressure measurements that were takenwith a poorly calibrated pressure gauge or pressure transducer are of limited value andwill frequently be discarded. Remember not only should the equipment used to takemeasurement be calibrated but the pressure, flow, and tank level monitors for theSCADA system should also be calibrated.

Even if measuring and monitoring equipment is well calibrated, other problems can ariseto give poor data. Suppose tank water levels is obtained from circular charts such as thatshown in Figure 5. Notice how the chart indicates the tank water level at a given time.We need to make sure that the timing mechanism in the chart recorder is working. Inother words, does the level on the circular chart at 6:00 a.m. really correspond to 6:00a.m. or some other time because the timing mechanism is slow or because the chart slipsor because someone installed the chart incorrectly. Issues such as these must beconsidered when using analog monitoring equipment. Of course the Y2K issue maycreate problems for digital systems that are not compliant.

Calibrating for Water Quality

In the preceding paragraphs we discussed some of the data requirements for hydraulicnetwork models. Key among this information is internal pipe roughness values for eachindividual pipe segment. Providing accurate roughness coefficients for moderate to largemodels can be a considerable undertaking. In fact, it is possible that each individualinternal pipe roughness value could be a candidate for adjustment during the calibration.

When calibrating for water quality, the level of effort required to obtain a suitablecalibration can be even more imposing. Most water quality models require information


on coefficients describing reactions that take place within the bulk fluid and reactions thattake place between the fluid and the pipe wall. In essence this means that two additionalparameters for each pipeline are required.

Obtaining bulk reaction coefficients is not that difficult and procedures are available toperform the necessary tests. Finding wall reaction coefficients can be much morechallenging. Just as each pipeline can have a unique internal roughness, it is possible thateach pipeline can have a unique wall reaction coefficient. Unlike roughness coefficients,no well-established test is available for measuring reaction coefficients. Unlike roughnesscoefficients, very little data is available for typical reaction coefficients. In fact, there issome question within the modeling community if “typical” reaction coefficients evenexist.

In the case of water quality models, a match between some observed and predicted waterquality parameter such as chlorine or fluoride concentrations is usually used forcalibration purposes. However in the case of nonconservative species such as chlorine notonly are system-wide concentrations dependent upon the network hydraulics, but they area function of any reactions that take place in the system. Clearly one can see thatcalibrating for water quality can greatly increase the level of effort required to obtain asuitable match between observed and computer predicted performance. Interestingly asGrayman points out, a water quality model can be used to help calibrate a hydraulicmodel [5].

Calibration Methods

There have been a number of papers written on the subject of hydraulic network modelcalibration and several methods have been developed to assist with the calibration effort.However the traditional method of trial-and-error (trial-and-effort?) still seems to be thepredominate method of choice. In other words, data describing the model is adjusted –sometimes randomly – until a suitable match is obtained.

The lack of calibration methods is not surprising once the full complexity of the level ofeffort required for a comprehensive calibration is understood. Ormsbee and Reddyprovide a nice summary of the steps necessary to calibrate a hydraulic network computermodel [3]. In their paper they note that direct solution techniques or optimizationapproaches offer some potential for simplifying the calibration effort. However formoderate to large systems calibrating a model can be a considerable undertaking.

Direct parameter calculation methods and optimization techniques are not immune to theproblem of compensating errors. This problem can be addressed somewhat withoptimization methods by placing constraints on the decision variables, e.g. piperoughness values, water demand, valve setting, etc. With the direct parameter calculationmethods engineering judgment is usually the filter which captures unrealistic modelparameter values. Nonetheless, the problem of compensating errors may still exist.


Perhaps the best way to address the problem of compensating errors short of measuringheads and flows at numerous points in the system is to calibrate for multiple demandconditions. These demands conditions may reflect maximum day, average day andminimum day demands. Generally speaking the physical characteristics of the systemwill be the same between the various demand days. Only the loading and boundaryconditions will change. If the same or nearly the same roughness characteristics producea suitable match between observed and predicted pressures then chances are good thatthese roughness values are the correct ones. However, there is no substitute formeasuring flows in throughout the system which, unfortunately, is not commonly done inthe United States.

Calibration Accuracy in the US vs. the UK

It should be clear by now that the level of effort required to obtain an acceptable level ofcalibration can be considerable. Water quality calibrations can be even more challenging.Many modelers in the United States agree that the level of effort required to calibrate ahydraulic network model will depend upon the intended use of the model [2, 3, 4]. Inother words, it may not be reasonable to expect that a model used for long-range planningpurposes will have the same level of accuracy than a model intended for operations orwater quality investigations. Walski provides some suggestions on a suitable level ofcalibration as a function of the use of the model [4].

In the United Kingdom, however, standards or performance criteria that modeler shouldstrive for have been established. The criteria for flow and pressure are shown below.Additional criteria also exist including those for extended period simulations [6].

1. Flows agree to:a) 5% of measured flow when flows are more than 10% of total demand

(transmission lines),b) 10% of measured flow when flows are less than 10% of total demand (distribution

lines).2. Pressures agree to:

a) 0.5 m (1.6 ft) or 5% of head loss for 85% of test measurements,b) 0.75 m (2.31 ft) or 7.5% of head loss for 95% of test measurements,c) 2 m (6.2 ft) or 15% of head loss for 100% of test measurements.

The manner by which hydraulic networks are operated in the United Kingdom goes along way to explaining why calibration criteria exist in the UK. For example, in the UKit is not unusual for water utilities to provide a lower level of fire protection than isprovided in the United States. Perhaps one of the bigger differences lies in metering. Inthe UK most individual service connections are not metered. Rather large networks aretypically divided into smaller “demand management areas” each serving about 1,000 –2,000 connections. The demand areas are independent service areas that are metered atall inflow and outflow points. As a result, there will be more “in-line” flow metering


than exists in the US. Cesario, et al provide a summary of the differences in water utilityoperations between the United States and the United Kingdom [2].

Calibration Accuracy

Despite the many variables that contribute to differences between observed and predictedsystem performance and even though calibrating a hydraulic network model can be atime consuming and sometimes frustrating task, the time is right in the United States toconsider adopting some calibration criteria. The results of computer simulations arefrequently used as the basis for capital projects involving many millions of dollars.Moreover, the 1997 EPA Drinking Water Needs Survey has placed the price tag over thenext 20 years of bringing water distribution systems up to compliance with the SafeDrinking Water Act at nearly $140 billion [7]. It is likely therefore that in the near futurecomputer simulation will be even more widely used for planning, design, operation andwater quality investigations.

As stated earlier, the primary purpose of this paper is to suggest calibration guidelinesthat can be used in the United States to determine a suitable level of model calibration fora particular use of the model. The purpose of this paper is not to establish calibrationstandards. The Engineering Computer Applications Committee hopes that by providingthese criteria some discussion among the network modeling community in the UnitedStates (and abroad) will ensue regarding the need for such criteria and the validity of thecriteria.

Minimum calibration criteria are presented in Table 2. Note that while a hydraulicnetwork model may have a number of different uses, four basic use categories arepresented in Table 2. Most of the uses of hydraulic models should fit into one of thefollowing categories: 1) Planning, 2) Design, 3) Operations and 4) Water Quality. Forexample training system operators, developing emergency response plans, or performingenergy efficiency studies would all fall under the Operations category.

Six calibration criteria have been identified. The level of detail criteria relates the degreeof skeletonization that a model may have while the type of time simulation reflectswhether the model should be a steady-state or extended period simulation model. Thenumber of pressure readings is expressed as a percentage of the number of nodes in themodel while the number of flow readings is presented in terms of the number of pipes inthe model.

Note that the criteria represent minimum criteria. For example, based on the proposedcriteria a model used for long-range planning may have a low level of detail. Howeverthere is nothing to prevent a model used for long range planning from being a model witha high degree of detail. Note that the level of detail and the type of time simulation will,in large measure, depend upon the use of the model.


References

[1] Walski, Thomas M., Sharp. Wayne W. and Shields, F. Douglas Jr. 1988 “PredictingInternal Roughness in Water Mains,” Miscellaneous Paper EL-88-2, US Army EngineerWaterways Experiment Station, Vicksburg, Miss.

[2] Cesario, A. L., Kroon, J.R., Grayman, W., and Wright, G. “New Perspectives onCalibration of Treated Water Distribution System Models,”

[3] Ormsbee, L.E. and Lingireddy, S. “Calibrating Hydraulic Network Models,” JournalAmerican Water Works Association, Vol. 89, No. 2, pp. 42-50, 1997.

[4] Walski, Thomas M. “Standards for Model Calibration,” 1995 AWWA ComputerConference, Norfolk, VA.

[5] Grayman, Walter M. “Use of Tracer Studies and Water Quality Models to Calibratea Network Hydraulic Mode,” Essential Hydraulics and Hydrology, Haestad Press,Waterbury, CT., 1998.

[6] Water Authorities Association and WRc, 1989, “Network Analysis – A Code ofPractice,” WRc, Swindon, England.

[7] U.S. Environmental Protection Agency. 1997. “Drinking Water InfrastructureNeeds Survey,” EPA/012/R-97/001. Office of Water, Washington, D.C.


List of Tables

Table 1

Pipe Roughness Values and Flow for Simple Two-Pipe System

Pipe #1Roughness

Pipe #1Flow (Gpm)

Pipe #2Flow (Gpm)

Pipe #2Roughness

80 660.75 689.25 166.8890 743.34 606.66 146.88

100 825.93 524.07 126.88110 908.53 441.47 106.89120 991.12 358.88 86.89130 1073.71 276.29 66.89140 1156.31 193.69 46.90


Table 2

Minimum Criteria for Hydraulic Network Model Calibration

Intended Use Level of Detail Type of TimeSimulation

Number ofPressure

Readings1

Accuracy ofPressureReadings

Number ofFlow Readings

Accuracy ofFlow Readings

Long-RangePlanning

Low Steady-StateOr EPS

10% of Nodes ± 5 Psi for 100%of Readings

1% of Pipes ± 10%

Design Moderate toHigh

Steady-StateOr EPS

5% - 2% ofNodes

± 2 Psi for 90%of Readings

3% of Pipes ± 5%

Operations Low to High Steady-StateOr EPS

10% - 2% ofNodes

± 2 Psi for 90%of Readings

2% of Pipes ± 5%

Water Quality High EPS 2% of Nodes ± 3 Psi for 70%of Readings

5% of Pipes ± 2%

Notes:1. The number of pressure reading is related to the level of detail as illustrated in the table below.

Level of Detail Number ofPressureReadings

Low 10% of NodesModerate 5% of Nodes

High 2% of Nodes


List of Figures

Figure 1

P = 54 Psi P = 47 Psi

Pipe 1L = 2,300 Ft

D = 10”

Pipe 2L = 2,800 Ft

D = 8”

1350 Gpm

Figure 2


J-23 J-24

These Demands AssignedTo J-24

These Demands AssignedTo J-23

Q Q Q Q QQ Q Q Q

Q Q Q Q Q Q Q Q Q

Figure 3

Time (Hrs)

Wat

er U

se

Figure 4


2-Hrs

A

B

Mid

6 am

Noon

6 pm

Figure 5

Top View

SideView

Figure 6


Documents

Calibrate Guidelines AWWA