?We wanted to understand how peak demands vary by specific consumer groups. We began to evaluate various automated meter reading devices and had the good fortune to find Teldata. Their AC-4 Telemetry Interface Units (TIUs) armed us with what we needed to launch a ?fine-tuned? flow investigation.?
-Frank Lau, Project Engineer for EPCOR Water Services
In an effort to delay an expansion of its water system, Canada?s EPCOR Water Services has turned to technology to learn something it had never before known about its biggest customers ? when and how they use water.
Thanks to a new automated meter reading (AMR) system, the municipally owned and independently operated water provider for the city of Edmonton can now capture usage data in 15-minute increments 24 hours a day. The data will be used to optimize the city?s entire water system, including its rate structure.
In 1993 EPCOR (formerly city of Edmonton Water Utility) was faced with a new municipal edict prohibiting rate increases for a period of five years. With usage on the rise but rates frozen, the utility shelved plans for expansion and began developing strategies to optimize the capacity and operation of its existing system.
As with many other utilities, the biggest challenge for EPCOR in understanding the capacity of the water system ? and how best to optimize it ? was knowing where to start. At first, the utility initiated a water conservation program designed to curb wasteful demand. The program featured, among other things, public education around odd and even lawn-watering days.
?This program allowed EPCOR to retain a reserve of about 10 percent of capacity,? said Frank Lau, Project Engineer for EPCOR Water Services, ?but we were going to need additional efficiencies in water use to adequately serve our growing customer base before expansion of facilities is justified.?
Better Data Needed
To analyze demand, the utility traditionally relied upon large-scale meters, costing $30,000 to $40,000 each, installed on transmission pipelines. However valuable, the data provided by these meters did not enable EPCOR to look at individual customers or analyze customer classes.
?We wanted to understand how peak demands vary by specific consumer groups,? Lau said. ?We began to evaluate various automated meter reading devices and had the good fortune to find Teldata. Their AC-4 Telemetry Interface Units (TIUs) armed us with what we needed to launch a ?fine-tuned? flow investigation.?
In 1997, EPCOR deployed the Teldata TIUs (also called data loggers by EPCOR) to the utility?s top 50 commercial and industrial consumers, who represent 5 percent of the utility?s overall customer annual consumption. Typically, these were locations with annual water consumption greater than 100,000 cubic meters (approximately 26.4 million U.S. gallons). An electrical contractor was retained, along with a meter installer, to install and reseal meters, and test the operation of all components. All of the customers had, or were supplied with, new encoder-type meters. Of the 50 major customers, six required new meters and 21 required slight modification to existing meters to change from direct read to encoder type; the balance required no modifications.
In some cases, the water meters had to be installed in unfavorable locations ? far from power sources, for instance. In other cases, the meter chambers were exceedingly confined.
?Installation of the Teldata data logger was simple and straightforward,? Lau said, ?even in those tight spots.?
With the AC-4, data collection requires just three wires between the meter and the TIU. Two other wires connect the TIU to a source of AC power. Following an aggressive schedule, installation of most of the units for EPCOR?s major customers was completed within 10 days.
Some Surprises
Once all of the TIUs were connected and powered up, they were set to record meter readings at 15-minute intervals. Using the information collected with the AC-4s between 1997 and 1999, EPCOR began to see some interesting results ? and not always what they expected.
?With customers like Molson Canada, we anticipated consumption patterns that might lend themselves to change,? Lau said. ?We thought batch draws of water at Molson?s Edmonton Brewery could be rescheduled to off-peak hours. What we ended up seeing was consumption patterns that did not lend themselves to change. Molson consumed an even flow of water throughout the day.?
While Molson?s consumption patterns weren?t conducive to change, examination of other consumers? usage patterns revealed a completely different scenario.
?One building products company?s data displayed classic batch demand,? Lau said. ?Examination of more than one day shows a pattern of high demand periods every three days. This is the type of consumption pattern that may allow for modification. Since flow is only required for one-hour intervals twice a day, it may be possible to arrange those demand periods to miss the typical high demand periods normally encountered in the city. Such modification of their demand patterns to accommodate city demands could possibly result in preferential rates for non-peak drawing of water.?
Other customers, such as the West Edmonton Mall with its vast indoor water park, show a consistent demand pattern similar to Molson. The days are all similar and the consumption rate is typical: less usage early in the morning with increases later in the afternoon, tapering off in the evening.
?This type of usage leaves little room for change,? says Lau. ?But, thanks to their stable (rates of) use, customers such as Molson or West Edmonton Mall can continue to draw water without taxing system capacity.?
Full AMR Initiative
When the TIUs first were installed all of the meter reading was done manually ? partly for budget reasons and partly because accessing phone lines was not always possible within the project?s tight timeframe. Data from the units was downloaded on-site into a laptop computer, which was connected to a serial port on each data logger.
Once downloaded, the new flow information proved invaluable. The labor costs, however, were enormous. Once a month for two years, EPCOR personnel visited each of the 50 sites with a laptop.
?It required 4-5 days to collect all 50 readings,? Lau said. ?One person spent three months per year for collection.?
According to Lau, ?Taking the readings was the easy part. The bulk of our employees time was spent gaining access to the data loggers!?
A prime example was the expansive West Edmonton Mall.
?Their facility stretches the equivalent of 24 city blocks,? Lau said. ?With a facility that large, imagine the time it takes to reach security, page maintenance, walk that labyrinth and finally extract data from the data logger. It often took 40 minutes before a laptop was ever opened.?
In 1999, EPCOR decided to review consumption patterns for its multi-residential customer group (apartment buildings containing some commercial uses and apartment buildings over four stories tall) and to institute a more efficient method of gathering flow data. The utility began to evaluate communication options that support AMR data transfer. These options included land lines, radio frequency, cellular digital packet data (CDPD) and cellular.
?It all boiled down to economics,? Lau said. ?40 percent of our customers were willing to share their existing phone lines. They did so, because it doesn?t take any time to collect data ? less than one minute in the middle of the night. For others, we paid to have separate land lines installed.?
The costs, he notes, have been well justified compared with the time and expense of manual data collection.
The vendor?s meter communications software, Teldata Manager ? designed specifically to support the TIU technology ? allowed EPCOR to collect meter data at pre-set times, then store and manipulate it for analysis. The programmable dial-inbound feature of the TIU, which can detect whether the customer?s phone line is free or busy, meant the data could be sent to EPCOR at times that were most convenient to the customer.
Data Proves Valuable
This proven AMR technology allows EPCOR to restructure customer categories, identifying those able to modify consumption patterns. ?Water supply planning relies on it,? Lau said.
?In addition, AMR enhances customer service. We can now share enhanced consumption data with our customers. Bar charts showing detailed usage patterns help customers diagnose trouble spots.
?This cant be done without submetering,? he said. ?Submetering can be done easily with Teldata?s telephony-based design. The phone infrastructure already exists.?
What began as a 50-piece, one-time initiative is now becoming a long-term objective for the company. As of February 2000, EPCOR had 105 units installed in the field. The long-term objective is to build water-demand profiles for as many customer types as possible by installing more units and/or re-deploying existing units.
?In the end, preferential rate is possible,? Lau said, ?as is further deferral of treatment plant, storage facility and transmission main expansion.?
Lau concedes that all the AMR technology in the world can?t stop demand from eventually exceeding capacity. ?In that case,? Lau said, ?planning for treatment facility expansion will have better support with more realistic peak demand patterns.?