Representation of a network made up of hundreds of thousands of pipes requires rigorous data management as well as a fast hydraulic solution. InfoWorks is geared for this, with features such as in-built quality control and the multi-user platform, which prevents compromising or duplicating the data.
Using larger models can bring considerable benefits to a network operator. A comprehensive model provides a more accurate understanding of the system so that operational and investment decisions are based on the best possible information. The resulting model is versatile and can then be used to simulate all kinds of scenarios, such as planned network expansions, optimization of pumping or the procedures for managing system failures. Adding more detail also aids calibration, as a simplified system can never fully represent reality.
Ever larger and more complex models are being built both for water and sewer networks, though the driving forces might differ.
In InfoWorks WS, a key impetus is a move towards all-mains water supply models, which could take account of even the smallest pipes supplying individual properties. Full modeling of a large network can involve hundreds of thousands of pipes. Examples on Wallingford Software’s website include a 400,000-pipe model of the entire water distribution network of Shanghai’s main supply company and a model of all 250,000 pipes in Miami-Dade’s network in the USA, as well as models in major cities such as Sydney (Australia), Antwerp (Belgium) and Belgrade (Serbia).
In contrast, it is uncommon for sewer models to include all the lateral connections. They tend to concentrate on the primary sewers and would generally exclude anything below 150mm (6”) or so. Nevertheless, sewer models are also increasing in size and often have tens of thousands of pipes in order to encompass all the carriers including rivers and streams that carry flows in the urban environment. “Sewer models in InfoWorks CS are getting bigger for this reason,” says InfoWorks CS Sales Manager Andrew Walker.
Major recent InfoWorks CS models include large networks in Berlin in Germany, Utrecht in The Netherlands, San Francisco in the USA as well as an integrated river and sewer system in Japan and a flood-prone urban catchment in north west London, England.
Access to more data
Recent years have seen a huge growth in the availability of data that is held in asset data registers and geographical information systems (GIS). The increased adoption of these systems is the key driver for the expanding use of large models, explains Wallingford Software Inc Senior Vice President Dr. Saša Tomiæ. InfoWorks can handle huge amounts of modeling data and has been designed to integrate with an organization’s asset and GIS systems to draw in data. This ease of access provides a further impetus for a comprehensive, true-to-life hydraulic model that capitalizes on the wealth of information available.
It is important for modelers to know that they have access to a good, quality-controlled register of the organization’s assets, says Wallingford Software Director of Technical Services Tony Andrews. This provides current information that can be used by an organization in many ways, including model building.
Wallingford Software’s InfoNet asset and data management system offers all the functionality of generic GIS and asset management solutions, with the added advantage of being designed with the specific needs of water and wastewater network operators in mind. It is ideal for storing large amounts of data and is being used successfully for the management of networks of over a million pumps, pipes, manholes and water pipes. Details of the assets to be modeled can then be input into InfoWorks.
Although GIS systems are not designed primarily for asset data, a lot of people do use them in that way, adds Wallingford Software Director of Product Management David Fortune. InfoWorks therefore incorporates extensive tools to overcome shortcomings in the use of GIS as an asset register.
Faster processors and plummeting costs of data storage and memory have also aided the move towards bigger models.
InfoWorks was developed from the outset to carry out hydraulic simulation and modeling of enormous networks. It has extensive capabilities in aspects such as hydraulics, mathematics, processing, performance and results presentation. It continues to be developed with twice-yearly updates that reflect clients’ changing needs and capitalize on the latest developments in technology. InfoWorks CS was the first modeling system in the sector to support dual-core processors. It is optimised for dual and quad processors, and even supports dual-quad processors. Running the software on a dual or quad core machine brings significant speed improvements of typically 40-50%, or up to 70% for dual-quad processors, says Mr Walker.
“A single hydraulic network could be 100,000 or more pipes, and InfoWorks can simulate this without any need to split it up,” says Mr Fortune. This is in contrast to some other products, where claims that they can handle large models simply mean that they can store the underlying data. “There may be reasons for splitting up a model for ease of management, but with InfoWorks there is no need to do so for hydraulic reasons,” he says.
One way of building ’mega models‘ is a ’top-down‘ approach, explains Dr. Tomiæ. A regional utility might cover 20 or more distinct urban areas but require them to be included in a single model. InfoWorks gives the flexibility to run individual parts of the model for routine management, or the overall model to address global issues such as budgeting.
The alternative ’bottom-up‘ approach involves modeling and managing separate models which can subsequently be merged into a regional model. InfoWorks provides tools so that the integration can be fully automated. The GeoExplorer within InfoWorks also includes a ’network of networks‘ for joining models at the network level without any need to see the detail of individual pipes.
InfoWorks is packed with tools that aid the modeling of networks, and presents the results in formats that are easily interpreted. There have also been significant improvements in the display of the network through InfoWorks’ GeoPlan, which ensures that it is easy and quick to work on all aspects of the model, no matter how many elements there are.
Rigorous quality control assumes particular importance as model size increases. Major decisions will be made on the basis of the results. “However fast it runs, however good the results look - it’s all for nothing if the results can’t be trusted,” says Mr Fortune. “The likelihood also increases of mistakes being hidden in these big models,” he says.
Wallingford Software’s products have been developed with the appropriate quality assurance tools built in as part of their structure - not added as an afterthought, he says. “People’s memories alone are not enough to keep a check on the provenance of the data,” he says. “We make it easy for people to build quality into whatever they do.”
One of the aspects of quality control is to ensure that the underlying data is current. Every parameter that goes into making up the models or asset data registers has its own flag - a simple but vital tool that can be used to indicate where the information has come from.
There are three distinct jobs in hydraulic modeling, he believes: building an asset register; building and maintaining hydraulic models; and using them. “I think that in the future we’ll begin to see these three aspects being treated separately, rather than one person doing all three,” he says.
“There is no way a single person can do all the work needed for a hydraulic model of 100,000 pipes - it is a multi-user activity,” he says. InfoWorks is already geared for multiple users, making it ideal for use on major projects. “True multi-user software is essential and that is what we provide.”
There is full transparency, with each person knowing what the others have done and a full audit trail for every item of input data. This emphasis on quality control and an audit trail can also provide benefits for anyone needing to prove the reliability of the results, for instance in showing that the organization had done everything possible to prevent a pollution incident. “Models can be such a powerful force in this,” says Mr Fortune.