By Dave Herchko, Sensus Metering Systems
“Data, data everywhere but not a place to store it” is the mantra being sounded more and more frequently in today’s utility market. The following is a conversation at “Utility ABC” between “Joe,” director of metering, and “Sam,” director of IT. (All names have been changed to protect the guilty.)
Joe: Sam, I just received a call from Larry, our vice president of operations, indicating we have a new tariff that requires collecting daily interval data for our 1,500 large commercial and industrial customers.
Sam: OK Joe, but where do we store all this additional data?
Joe: We just have to figure it out. Oh, by the way, we also will be installing a new “advanced metering infrastructure” (AMI) from vendor XYZ for our 8,000 small commercial and industrial customers. The requirement is to collect hourly data for a demand-side management tariff that we are offering the customers.
Sam (sarcastically): I guess I will need to buy a “disk stretcher”!!
Joe: Just one more thing: We will be installing an AMI system from vendor MNO for our medium and large class of residential customers, a population of 1.2 million endpoints. We’ll be collecting daily data for a load management tariff.
Sam: *#%@^$! Where do we store all this data? More importantly, how will we ever interface these different systems?!
Joe: Did I mention that we need to provide daily web-based access to all this validated/edited/estimated (VEE) data for our customer service reps and account managers?
Sam: (a thud, then banging sounds)
Joe: Sam? SAM!
So, what is the solution to Sam and Joe’s dilemma?
“- Software that manages large volumes of meter reading data;
“- A system that can interface multiple systems with the utility’s billing and support systems;
“- Software that can support VEE requirements for different classes of data; and
“- Internet-based access to the billing and settlement quality data
This article discusses the solutions required to close the gap between a utility’s existing IT infrastructure and the systems and tools provided by advanced metering infrastructure (AMI) systems.
Identifying the Need
As price points of AMI systems and devices become economically feasible, several utility companies are in some stage of an AMI implementation for some or all of their metering points. These AMI solutions are more advanced than the traditional AMR data collection that used some form of walk-by/drive-by meter RF solution. AMI incorporates a true fixed network (FN) system, where all meter reading data is collected via some form of automatic electronic infrastructure.
As a result, these utilities now receive large volumes of meter reading data that interfaces with revenue-cycle based systems. This raises the need for a system that can store the increased cache of metering data. AMI systems provide the ability for granular meter reading data from daily reads all the way down to intervals of 15 minutes or less. The issue becomes: Where will the utility store the several million points of valuable meter data?
A gap exists since most utility IT systems are built around monthly revenue reading/billing cycles and are not prepared to store daily, hourly or even more granular meter reading data. While the AMI support systems provide the ability to collect data and mange networks, they are not designed to bridge this all-important gap.
Meter Data Management Systems
The solution is known by many different names: meter data management (MDM) system, meter data warehouse, meter data mart or meter enterprise system. In short, an MDM system is a software tool that provides today’s utility with a solution for the gap created as AMI systems begin to be installed by providing daily meter reading data for connected customers. An MDM system should offer the capability to bridge the gap between the utility’s existing IT infrastructure and the different AMI systems the utility may deploy now and in the future. A quality MDM system provides the following functionality:
“- A scalable system that can support large or small quantities of meter data, anywhere from less than 50,000 to more than 5 million data points.
“- Industry standard data warehouse design, either:
“- Enterprise solution for rapid delivery of meter data information for reporting and data access; or
“- Star schema design (Star Schema is a relational schema whose design represents a multidimensional data model. The star schema consists of one or more fact tables and one or more dimension tables that are related through foreign keys.)
“- True Open Database Connectivity (ODBC) with a high degree of open architecture, consisting of either:
“- Relational databases (Oracle, Sybase, SQL-Server, Sybase, DB2); or
“- Off-the-shelf enterprise middleware systems-web server and charts development.
“- Operating platform independent: Windows, Unix or Linux.
“- Energy (gas/water/electric) and interval (C&I/residential) independence.
“- Web-enabled languages to make for easy support of web presentation or applications that need minimum additional investment.
“- Energy/interval independent validation/editing/estimation (VEE) engine, such as:
“- Pre-Validation before data warehouse update;
“- Industry standard validation, editing and estimation functionality;
“- VEE functionality targeted to meter readings only; or
“- Versioning and auditing of any edited/estimated data.
“- Data structure that supports both regulated and deregulated markets. Major structure keys include the utility, supplier, generator, customer, premise, account and meter.
“- Support for aggregation of meters at all major keys and defined “Totalization Groups.”
“- Extensive graphical, tabular and Excel export of the metered and totalized data from the interval to system level. Report types include:
“- Total consumption;
“- KVA analysis;
“- Comparisons; and
“- Detailed interval.
MDM System Selection Decision
As utilities enter into the decision process for an MDM, there are two possible scenarios that can be taken based on the approach required to support AMI: the traditional MDM or the multi-functional.
Traditional MDM (Data Warehouse)
The traditionalist would be more inclined to look at MDM systems that have minimal transaction functionality built into the data warehouse. He or she would rely heavily on the AMI systems for managing the data collection/AMI infrastructure. In this approach, the utility would use the data warehouse in its purest form: enterprise management of large quantities of billing/settlement quality data.
However, even in this approach, a necessary transaction function is a validation/editing/estimation (VEE) engine. As AMI systems have varying degrees of VEE capability, having one central set of VEE rules that can be applied to all the collected meter data is a fundamental requirement. The VEE engine should also segment VEE based on meter type (residential, commercial and industrial). Having the VEE function attached to the warehouse also requires versioning of the data so that the values on any interval may be stored over time and be easily accessible for auditing purposes.
Another advantage to this approach is the MDM cost. MDM systems with multiple transactional functions traditionally cost more for the license than those with minimal functional capability. With the traditional approach, the utility can work with the MDM supplier to develop specific functions and interfaces that meet the utility’s specific requirements.
This system features multiple support transaction functions that are bolted to the data warehouse, including (but not limited to) asset management, demand-side management capability, a billing engine and other meter-related functions. The net effect of these functions on the enterprise capability and performance of the data warehouse may have an impact.
More importantly, the functions that are part of the data warehouse system in almost all cases require modifications to meet the utility’s requirements. In some cases, the cost of the modifications could possibly surpass the original MDM license cost. In this case, more is not better.
Knowledge is Power
“Data, data everywhere” will surely be the mantra for those utilities that are moving into the AMI arena. Remember that the internal rate of return (IRR) for these systems is becoming more attractive as the endpoint costs continue to decrease. From the utility’s perspective, having all classes of meters on AMI provides a distinct advantage in many areas: customer switch on/off, energy theft, power outage notification, customer account management, customer service, load planning, load forecasting, customer retention and much more.
Ultimately, as a utility approaches the AMI solution for meter data collection, the right MDM will be as important as the AMI solution(s) selected. At that point, the mantra will probably change from “Data, data everywhere” to “Knowledge of data is power.”à¯£à¯£
Dave is vice president of AMR products and services with Sensus Metering Systems and is on the AMRA Board of Trustees. Dave has more than 20 years of experience in the AMR and utility metering markets and is a member of the AMRA and AWWA organizations.