AMI-Based Load Research
By Reed Karaim | RE Magazine
Before Kaua‘i Island Utility Cooperative(KIUC) installed its advanced metering infrastructure (AMI), conducting load research was a time-consuming and costly prospect, involving hundreds of man-hours to install special meters and hundreds of thousands of dollars in expenses.
“The major cost of the load research is actually getting the data,” says Mike Yamane, KIUC’s chief of operations.
In fact, the co-op, which operates on the Hawaiian island of Kaua‘i, had never done a full load research project, relying instead on “proxy” data from other Hawaiian utilities.
So in 2008, when Hawaii’s Public Utilities Commission recommended the co-op conduct load research to ensure fair cost allocation among its customer classes, KIUC, which already had begun making the case for an AMI change-over, decided the time was right.
The co-op won approval from its board in 2009 and had already begun installation of a Landis+Gyr Gridstream® AMI system when word went out in 2010 that NRECA was looking for participants for the $68 million Department of Energy-sponsored Smart Grid Demonstration Project (SGDP). KIUC jumped on board, spearheading the project’s analysis of the usefulness of AMI when conducting load research.
The co-op followed a series of steps toassemble AMI-based class load profiles:
- Develop stratified sample designs and sample selections
- Capture hourly interval sample customer loads
- Validate interval data
- Create reports showing class loads and load details
- Support rate design studies and other analysis with load information
“For our typical load research study, you put in special meters that could take more data,” Yamane says. Sample size is usually about 400 meters per rate class, regardless of the utility’s size, at a cost of up to $1,000 per customer. “With AMI, we can just extract the data from 400 meters out of the 30,000-some meters we have. We can choose a sample set that makes sense and just grab that data.”
KIUC reported that AMI-enabled load research returned multiple benefits including:
- Elimination of cost for swapping billing meters with special load-sampling meters
- Study lead time reduced from months to days or weeks
- Reduced likelihood that customers will elect not to participate
- Extremely clean load data
One challenge the AMI system presents is the sheer volume of data it makes available.
“Part of the load research study is more analysis after the data is collected, so it will still cost something to manipulate that data,” Yamane says. “But it makes everything easier.”
A report on the project concludes: “KIUC has demonstrated the effectiveness of using AMI for the load research program. The information provided can be of such detail as to fully satisfy regulatory requirements and monitor the impacts of alternative rate structures and consumer response.”