One of the most daunting tasks associated with modeling water utilities is handling the demand distribution for the water utility. It is necessary to develop a baseline model where an average demand for all users is allocated to appropriate junction nodes in the model. The modeler may choose to use an average monthly, yearly or some other representative average. Distributing this demand to appropriate nodes, updating this information and accounting for the demand type (residential, industrial, etc.) all can require a great amount of work. The Pipe2000 GUI offers a very useful option for handling water utility demands. There is a meter element which allows users to place one or more meter connections at any location in any pipeline and associate these meters with the user and demand type. In addition there is a pipe attribute for the number of residential meter connections for each pipe. With these capabilities, once the user provides the meter data the demand distributions are handled by the software. .
We envisioned folks using the meter element to define larger users and the residential connection feature for residential customers. Many of our users are doing this with great success. A few users have defined metered connections for all customers. Notably, the usage can be updated automatically using billing records resulting in less tedious updating of water utility demands. Also note that a meter connection adds no overhead to the model (it doesn’t count as a node and doesn’t split a pipe into two or more sections). A great advantage of this method is that you can have many meter IDs with user information and demand types associated with this element. Using meters to handle demand just makes sense and billing records using a matching meter ID allows easy and automatic updating of demand data. Using meters is just a better way to handle water utility demands
A good example of the benefits of incorporating meters and meter data into your water utility model is the City of Huntsville, TX model. They have incorporated their meter data into their model and can quickly update their demand data. Mark Moore, their Senior Designer and Principal Modeler, tells me that their model predictions are so close to field data (within 1-2 psi) that Engineers and Contractors that work in their City will use the model predictions in lieu of a hydrant test. Moore feels the key to the model accuracy is the use of meters to handle the demands and their distribution.
Finally, the use of meters provides an extremely quick and accurate method for distributing residential demands (usually around 80% of total demand) avoiding the time consuming task of assigning residential demands. I will elaborate on this great Pipe2000 feature and how it can save you hours of work in my next blog.