Leaks and Audits: Stanching Water Waste

Like nearly every nation on the planet, the US is facing a water crisis. Old, rusted cast iron piping in cities and towns in the Northeast are at the end of their useful lives, fights and lawsuits over water rights in the desert Southwest are heating up, increasing numbers of pipe bursts around the nation are impacting often thousands of Americans each day, and network inefficiencies and more frequent droughts are resulting in water rationing in hard-hit areas.

With access to one of the world’s most precious commodities increasingly at risk even in developed nations like the US, it’s no wonder more municipalities and government officials worldwide are looking at ways to stem water waste beyond just the major pipe breaches. Technological advances are making their jobs easier, with a variety of options available today that assist water utilities in the management of water distribution, even when it is not fiscally possible to change out old pipes for new.

Aging Water Distribution Infrastructure

A recent American Water Works Association report showed the US water infrastructure, and its more than 1 million miles of underground pipes, has aged to the point where the pipes currently need, or will soon need, replacement. That’s no big surprise, given that most of the water infrastructure in the US was designed and built 50 or more years ago. (And in many areas in the Northeast, for instance, cast iron pipes still in use today may have been laid as many as 100 years ago.) The average lifespan for water pipes is about 75 years. Add to that rapid demographic changes and population and economic growth, which affect the usage of pipes, and one can easily see why failures are frequent.

Headline-generating pipe failures in major cities speak to the decaying infrastructure. However, just as insidious are the less visible events such as smaller leaks, faulty valves, and other network inefficiencies that increase the amount of non-revenue water (NRW).

NRW and infrastructure inefficiencies are not unique to the US. In fact, the high price of water in many other countries has forced these issues to boil to the surface more quickly than they have in America. That is because in the US there are few incentives for making needed changes to the distribution infrastructure to lessen the pain. With some of the lowest water prices in the world and fewer federal-level regulations around water loss than many other nations, US utilities may have been spared the worries of their foreign counterparts relevant to water loss.

Still, more and more municipalities are recognizing that water loss costs them money, and are looking to cut those losses. Moreover, greater logistical and operational demands are being placed on water utilities in terms of supply, distribution and waste. A 2011 Frost & Sullivan research report titled “Addressing Water Utility Inefficiencies through Smart Water Grids,” notes water utilities will need to tackle initial problems concerning non-revenue water, leakage, and quality.

Dr. Mark LeChevallier, director of innovation & environmental stewardship at American Water, recently posted on the Smart Water Networks (SWAN) Forum that the list of challenges for water utilities is long and includes water pricing issues and infrastructure. “Our future investment can’t continue to come from rate increases…. We need to focus on operational efficiencies so we can continue to improve without raising rates.” He added that pipeline replacement, more efficient operations, and tighter management of leakage and water loss were all major concerns.

Addressing the Issues

One way some utilities begin to address water loss is by dividing their distribution regions into smaller zones, sometimes referred to as district metered areas (DMAs). But physically disconnecting one part of the network from another is not always cost-effective or feasible. And this is where emerging water monitoring technologies can help. In fact, with the newest technology, utilities can create “virtual” divisions of the network with metering equipment.

Such virtual DMAs may even already exist in the form of defined regions such as pressure zones. A pressure zone, with its limited number of inlets and exits and with pressure sensors at each of these points, can serve as a DMA for network monitoring purposes.

The argument for distribution regions can be made more compelling when tied to public health–improving hydraulic modeling of distribution systems, understanding residence time in a real-time sense, and identifying backflow are all means of increasing and controlling water quality. And, as a means of providing continuity of revenue, reducing non-revenue water and the costs associated with eliminating excess production is also important.

Better Water Monitoring

Typical modern-day tools for effectively and efficiently managing a water network include flow meters, pressure valves, and quality sensors. In most cases, these technologies are used to more effectively understand and manage the water flow. Data transmitted from these network elements are used to ensure that water flows continuously, smoothly, and efficiently to consumers. And when used for smart network monitoring, these technologies enable faster (and more pinpointed) leak detection.

With the right types of sensors and meters in place, utilities can mine data for additional benefits, like water pipeline pressure optimization, identifying faulty meters, and pointing out other inefficiencies within the pipe network. Smart water networks gather, process, and analyze the data, then feeds back to the utility alerts and detailed information on events such as water leaks. And they do it far earlier than current practices typically can.

For example, Yarra Valley Water utility in Melbourne, Australia, deployed a smart network monitoring solution in the last year. The utility viewed network monitoring as the next step in its arsenal for reducing NRW, and began monitoring a small region of the network. The results were so positive that within a few months they expanded coverage to the entire network. Their results included being able to discover several large leaks and bursts up to two weeks earlier than their previous systems allowed for, they were able to narrow down the search area where a leak or burst occurred by as much as 66% within a zone, and they were more quickly able to detect meter faults and malfunctions.

In other cities, like London, creating smart water networks has been happening for more than 20 years. Thames Water, the water utility overseeing 20,000 miles of water mains for London and environs, uses DMAs, advanced sensors, and a multitude of data systems to combat water loss and major disruptions–with improvement year over year as mandated by the stringent British water regulator, the Ofwat.

But regulation is not the only driver for smart grids in London; the city is gearing up for the 2012 summer Olympics, a huge undertaking for the city and even moreso for its Victorian-age pipes. Thames Water has about 4000 flow meters and pressure gauges in place, which will help it deal with the huge influx of Olympians and tourists flocking to the games–all of who will require ample water pressure for their needs. But increasing pipe pressure can increase the likelihood of bursts, so Thames Water will rely on the data it collects from its smart water network to alert it to danger zones.

With more than 50,000 water utilities in the US serving a population of 310 million, clearly there are ample opportunities for better water monitoring on this side of the pond, too.

Too Much Data?

With the numbers of water utilities worldwide taking a data-driven approach to managing and controlling their networks, the next hurdle is overcoming the rapid influx of data. In addition to flow and pressure meters being deployed, telemetry and supervisory control and data acquisition (SCADA) mean more data can be collected more often.

Raw SCADA readings generally are used for ongoing operational needs, but they also contain the information that can reduce water loss. However, much of the information is obscured by “noisy” readings, network operations activities, faulty meters, variable consumption, and many other complications.

So the challenge for the data-rich utility is that ever-growing quantities of data have to be processed, analyzed, and correlated. The requirements from data analysts are increasing, and “shortcuts” such as night-flow analysis or fixed-bound alerts can cause numerous alerts, false alarms, and data-gaps due to meter faults, transmission errors, GIS mismatches, and more.

Water Network Monitoring helps overcome these challenges by using existing data, even sparse and noisy data, and analyzing it to automatically create actionable alerts. This real-time analysis of network monitoring is a new field in which sophisticated mathematical and statistical algorithms process the data collected by monitoring systems through sensors. By design, smart water grids identify immediate problems and allow utilities to focus on strategic challenges and gain foresight into emerging issues

The collected data is synthesized and used to detect and warn of network problems and inefficiencies, compare day-to-day differences in flows to pinpoint problems down to the meter, and more.

Subtle differences are identified in the data through a process of statistical analysis, rather than looking just at fixed absolute or percentage-change bounds, allowing for the detection of water loss, energy inefficiencies, faulty meters, transmission problems, and more. The network data is analyzed as a whole–data from all sensors in addition to external information such as weather, holidays, time-of-day, and more. Repeating patterns and ongoing long-term changes can be discovered in the data so small leaks can be detected and treated before they become large bursts. This approach means immediate cost-saving alerts about water loss and anomalies that would otherwise go undetected.

The goal of water network management is ensuring a high-quality water supply to consumers while, at the same time, optimizing utility operations and energy consumption, and minimizing non-revenue water and leaks.

Software As a Service

In some water network management deployments today, utilities take advantage of software as a service (SaaS) solutions, where data is gathered, analyzed, and stored in the “cloud.” For utilities, the benefits to a SaaS model are simple: there is no financial outlay for hardware or software licensing. And once a utility has sensors in place and data to be analyzed, the SaaS model allows for making the most of the data with little or no risk. A SaaS solution allows a utility to start small–monitoring only a limited section of the network and expanding the monitoring as the utility sees fit, as in the case of Yarra Valley.

By utilizing a SaaS solution, the system is accessible anywhere, at any time, by anyone with the proper credentials. Depending upon how the system is set up, data from the utility’s sensors gets transferred automatically to the SaaS system for analysis–with no intervention required from the utility. The data is cleaned and analyzed with the results displayed on a Web interface (or dashboard), which the utility can quickly act upon any issues that have arisen.

The best part of a SaaS approach is that it requires no network changes, hardware additions, or other field work. The system simply makes better use of the existing data within the network and makes it more actionable.

Making Network Monitoring Practical

In today’s economic climate in the US, though, as practical and smart as online water network monitoring sounds, tight budgets might thwart the examination of water efficiency goals.

But is it really out of reach? For each 100 miles of main pipes in a network, the utility should have approximately 10 meters measuring flow and/or pressure. That means for a city the size of Chicago, as many as 800 meters might be ideal. But even with just 200 in place in a large city, online network monitoring can be utilized and meaningful data extracted. As benefits and savings are realized, the network can be expanded. In smaller cities with fewer than 500,000 people, 100 meters would likely suffice. Meter prices start in the low thousands and can range as high as $100,000 each, but most utilities can expect to spend a few thousand dollars per meter. Improvements in technologies are even making it easier and less costly to install meters. In some cases, municipalities are installing them through manholes.

Consumer and Other Cost Benefits

Graham Symmonds, CTO & SVP-Regulatory Affairs & Compliance, GWRI–Global Water Resources in the US, noted in the SWAN Forum that efficient monitoring benefits water utility customers.

“The average consumer knows neither how much water they consume, nor the impact that small changes in behavior can have on their own costs,” he says. “We know that water prices are going up as a result of increasing scarcity, decreasing quality, degrading infrastructure, and increasing operational costs. Once the cost of water crosses the “˜care point’ for consumers, they will be demanding more information from their utilities.”

As consumers increasingly demand more data, Symmonds says, they want it customized for themselves. ” “˜How much water did I use today, and how did that compare to yesterday?; Is my consumption statistically different from a week ago?; How do I fare compared to my street, my neighborhood, my city?; How much water should I use?; Based on weather data and evapotranspiration calculations, how much should I have used outside?’–information from the Smart Grid answers these questions,” says Symmonds.

Forward communities like Sonoma County and Glendale, in California, and DC Water, in Washington DC, are among those regions leading the charge for improved water usage in the US. All three have sophisticated monitoring systems in place today and have realized clear benefits, including benefits for consumers. (See “Glendale Water & Power” in Additional Content.)

 And at Aquas de Antofagasta, which operates a concession for the distribution of water in Chile’s Antofagasta Region–supplying water to the five largest cities in the region–network monitoring has made repair crews more efficient. They have been able to save twice as much water from the same number of repairs by more accurately pinpointing trouble spots.

They also experienced a noticeable improvement in quality of service to their customers with regard to pressure and continuous supply.

A Less Wasteful Future

There is no doubt that smart water grids with network monitoring can only serve to help utilities save money, energy, and water. The benefits are not only to the utilities, but to the consumer who will enjoy the more efficient operations of the water utility, not to mention not suffering from paralyzing water main breaks.

The challenge in the US is making the smart grid accessible for communities of all sizes. With more than 50,000 community water systems in the US–and more than half of those serving fewer than 100,000 people–it’s imperative that smart grid functionality be reasonably priced. SaaS models, easier implementations of sensor technologies, and return on investment will all help smart water grids establish a firm foothold in the US.