The Quick-Start tutorial is designed for first-time users of InfoWater Pro Designer and provides a guided tour to core commands and functions used to create and execute a water distribution design optimization run. The Quick Start tutorial will help you become familiar with the core set of InfoWater Pro Designer features and should be used as a launching point to a more comprehensive understanding of the program. The estimated time to complete the Quick Start tutorial is approximately 45 minutes. The Quick Start tutorial will help you become familiar with the following:
During this Quick Start tutorial, you will modify two existing projects. The first project (TwoLoop) is used to illustrate the design of a new system while the second project (NewYork) deals with the rehabilitation of an existing system. SI units are utilized for the first project and US Customary units are employed for the second project. These projects can be downloaded from Quick Start Tutorial Example files.
The TwoLoop project modified in this tutorial illustrates how Designer calculates pipe sizes to meet specified design criteria at the minimum cost. The TwoLoop model schematic is shown below. The model contains one treatment plant, six junction nodes, and eight pipes. All eight pipes have “dummy” diameters of 200 mm and Hazen-Williams roughness coefficients of 100. The objective is to size the existing pipes for the given network layout. A discrete diameter set of 14 values will be specified along with a minimum pressure requirement of 30 m to be maintained at each junction node.
The decision variables for this problem are the eight pipe diameters with 14 different commercially available sizes to choose from. This results in a solution space containing 148 or 1.48 x 109 possible designs.
During this first tutorial, you will be guided through:
Pipe design groups must be specified before specifying other options in the designer. Pipes should be grouped together based on similar characteristics (e.g., material, age, location and diameter) and associated improvement costs. All pipes within a group will possess the same final diameter for the following three design actions: New, Parallel, and Replacement. In this example, eight pipe design groups will be created, one group for each pipe. A new database field must be added to the existing Pipe Information database table and populated with the appropriate group IDs.
The Pipe Group Definition table is now populated.
The next step is to define the desired lower and upper bounds for the pipe diameters along with their associated cost codes and applicable actions for each of the eight design groups. Enter the following in the Pipe Group Definition table for each pipe group (1-8).
Minimum Diameter: 25.4 mm
Maximum Diameter: 609.6 mm
Cost Code: 1
Applicable Actions: NEW
Right-clicking anywhere in the table allows you to access additional commands, including copy and paste.
The Designer-Pipe Group dialog box now appears as follows:
Specify the Design Action for the specified pipe groups using the Actions view. Since the entire system is new, only one Action (NEW) needs to be specified along with the set of commercially available pipe sizes to choose from and their associated costs. A total of 14 discrete pipe sizes with their cost matrix will be inputted. Enter the data as shown in the following table.
Select the Groups view and select the Action New for the eight Pipe Groups, as illustrated in the picture below for the first Pipe Group.
Using the Constraints, target minimum pressures can be defined for the entire system, for all the junction nodes within the specified domain, or for only selected key individual junction nodes. For this example, you will specify a minimum pressure of 30 meters and a maximum pressure of 150 meters to be maintained for all the junctions in the system.
Note that if Specific Constraints for a junction were specified, they would override the Constraints for the Domain. No maximum velocities or hydraulic gradient (headloss per 1,000 unit of length) constraints for pipes will be specified for this example.
Select the Options view and specify the following GA optimization options:
Advanced GA Options should only be modified when you completely understands how Genetic Algorithms work.
All required information for executing a design run is available to the Designer. Click on the
Run
button (or
Run - >
Start) to start the optimization. Click
YES when prompted for Confirmation. The Simulation view appears and the statistics are updated with each trial. The figure below depicts the Simulation view after the optimization process is completed:
A design solution at a cost of approximately $420,000 was reached after 50,002 trials.
Since GA optimization is a stochastic search process, you may encounter local optimal solutions that differ from the global solution shown above. In case such a local solution is reached, simply rerun the design simulation to further explore the solution space to narrow the search towards the global solution.
The design results can be reviewed in the Report section of the Designer. The Report consists of five views: Summary, Pipe Group, Individual Pipe, Junction Constraint, and Pipe Constraint. Note that each view contains separate views for Pipe Group, Individual Pipe, Junction Constraint, and Pipe Constraint. SUMMARY view: Summarizes each of the three best solutions out of 50002. Three solutions were retained because the Allowed Maximum was set to 3. Note that the lowest cost solution has a greater penalty cost than the second lowest cost solution. This view is useful for looking at the total costs. PIPE GROUP view (Solution 1): This view details the pipe group results for solution 1. The determined diameters are shown, and the total length of pipe in each pipe group is shown. The total cost for each pipe group is included as well. INDIVIDUAL PIPE view (Solution 1): This view is very similar to the PIPE GROUP view, except that individual pipes are shown instead of pipe groups. This view can be used to locate pipes that may be determined to be too costly; these pipes can be placed in an alternate or new pipe group for a subsequent Designer run. JUNCTION CONSTRAINT view (Solution 1): This view contains each junction with pressure range constraints for solution 1. Pressure violations are shown in the right side columns. Minimum and maximum violations are in separate columns. PIPE CONSTRAINT view (Solution 1): This view contains each pipe with velocity or headloss violations for solution 1. These violations are shown in the right side columns.
Maximum velocity or hydraulic slope constraints were not specified for this example and therefore this view is empty.
Run another optimization with a penalty cost value of 100,000 and a maximum number of allowable trials of 100,000.
Because of the stochastic nature of GA optimization, you may encounter optimal solutions that differ from the local solution shown below.
The Pipe Modeling database table can be updated with the new pipe diameters and roughness coefficients by exporting the Designer optimization results.
Exporting Data to the Information DB Table is useful for tracking changes or making a log of recommendations. The Modeling DB Table should be updated with caution; using an alternate scenario (Pipe Set) is the best option.
The second example involves modifying an existing project called “NewYork”. This project should be located in the same location as the previous example. The “NewYork” project modified in this tutorial illustrates how Designer calculates parallel pipe sizes to meet the design criteria at the minimum cost. The model schematic is shown below and considers the expansion of an existing system through paralleling existing pipes. The model contains one reservoir, 19 junction nodes, and twenty-one large diameter pipes (tunnels).
The objective is to determine which existing pipes to parallel and what size should the new parallel pipes be to supply the projected increase in demands while meeting specified pressure criteria. The new pipe diameters will be selected from 15 commercially available sizes.
For this problem, 16 decision variables including the “do nothing” option make up a solution space of 1621 or 1.93 x 1025 possible pipe combinations.
Start InfoWater Pro and Open the project NewYork.aprx.
In this example, 21 pipe design groups will be created, one group for each pipe. A new database field must be added to the existing Pipe Information database table and populated with the appropriate group IDs.
The Pipe Group Definition table is now populated.
Define the desired minimum and maximum limits for the pipe diameters with their associated cost code and applicable actions for each of the twenty-one design groups. Enter the following in the Pipe Group Definition table for each pipe group (1-21). Minimum Diameter: 36 in
Maximum Diameter: 204 in
Cost Code: 1
Applicable Actions: PARALLEL
Right-clicking anywhere in the table allows you to access additional commands, including copy and paste.
The Designer-Pipe Group dialog box now appears as follows:
Specify the Design Action for the specified pipe groups using the ACTIONS view. A total of 15 pipe sizes and associated cost data will be inputted since only one Action (PARALLEL: Parallel Pipe) is specified for this example. Use the table below as a guide when entering this data.
Select the GROUPS view and select the Action NEW for the Pipe Groups, as illustrated in the picture below for the first Pipe Group:
For this example, you will specify a minimum pressure of 155 ft and a maximum pressure of 250 ft to be maintained for all the junctions in the system. However, the constraint for two junctions will differ.
No maximum velocities or hydraulic gradient requirements for pipes will be specified for this example.
Select the Options view and specify the following GA optimization options:
Advanced GA Options should only be modified when you completely understand how Genetic Algorithms work.
All required information for executing a design run is available to the Designer. Click
Run
(or
Run - >
Start) to start the optimization. Click
YES when prompted for Confirmation. The Simulation view appears and the statistics are updated with each trial. The figure below depicts the Simulation view after the optimization process is completed:
A design solution at a cost of approximately $97.75M was reached after 50,002 trials.
Since GA optimization is a stochastic search process, you may encounter local optimal solutions that differ from the global solution shown above. In case such a local solution is reached, simply rerun the design simulation to further explore the solution space to narrow the search towards the global solution.
The design results can be reviewed in the Report section of the Designer. The Report consists of five views: Summary, Pipe Group, Individual Pipe, Junction Constraint, and Pipe Constraint. Note that each view contains separate views for Pipe Group, Individual Pipe, Junction Constraint, and Pipe Constraint.
SUMMARY view: Summarizes each of the three best solutions out of 50002. Three solutions were retained because the Allowed Maximum was set to 3. Note that the lowest cost solution has a greater penalty cost than the second lowest cost solution. This view is useful for looking at the total costs.
PIPE GROUP view (Solution 1): This view details the pipe group results for solution 1. The determined diameters are shown, and the total length of pipe in each pipe group is shown. The total cost for each pipe group is included as well.
INDIVIDUAL PIPE view (Solution 1): This view is very similar to the PIPE GROUP view, except that individual pipes are shown instead of pipe groups. This view can be used to locate pipes that may be determined to be too costly; these pipes can be placed in an alternate or new pipe group for a subsequent Designer run. JUNCTION CONSTRAINT view (Solution 1): This view contains each junction with pressure range constraints for solution 1. Pressure violations are shown in the right side columns. Minimum and maximum violations are in separate columns. PIPE CONSTRAINT view (Solution 1): This view contains each pipe with velocity or headloss violations for solution 1. These violations are shown in the right side columns.
Maximum velocity or hydraulic slope constraints were not specified for this example and therefore this view is empty.
The Pipe Modeling database table can be updated with the new pipe diameters and roughness coefficients by exporting the Designer optimization results.
Exporting Data to the Information DB Table is useful for tracking changes or making a log of recommendations. The Modeling DB Table should be updated with caution; using an alternate scenario (Pipe Set) is the best option.
Congratulations! You have now completed the InfoWater Pro Designer Quick-Start tutorial.