Re-engineering smart grid data post-deployment


The first part of unlocking that full potential is defining what the utility wants its smart grid to look like. Smart grids are all unique and should be built with each utility’s drivers in mind. Some utilities will focus development on a smart grid to reduce outages, others to integrate green technologies, to offer pricing incentives for demand response, while another utility might seek to engage customers digitally, amongst a host of other reasons.

While smart grid designs will vary from utility to utility, there are core similarities between them. Virtually all smart grid projects have a foundation rooted in advanced metering infrastructure (AMI) technology which enables two-way communication from a customer’s meter to the utility. Unfortunately, a common truth is that much of the value of past AMI projects has not been realised. There are several reasons for why that may be:

1. The AMI system was designed to improve meter reading only. When engineered properly, the AMI system should help a utility better manage their assets across the distribution system. There is tremendous value in being able to monitor power delivery on an hourly or sub-hourly basis, coupled with outage information (momentary or sustained), voltage data and power quality metrics.

a. On a positive note, many projects today are undertaken with the goal of asset management and employ data analytics as the drivers.

2. Data is not presented in a way that is actionable for users. A meter data management system (MDMS) can deliver AMI information in tandem with data from other systems that can better assist utility employees in diagnosing problems and making informed decisions. The key is presenting the information in a way that is easy to understand.

3. Only one group or a limited number of groups in the organisation participated in the design activities. It is critical to have all the groups at the utility involved so that they:

a. Are active participants in the project and vested in its success and;
b. Have designed a framework that provides value throughout the entire organisation.

Post-deployment, there are opportunities to re-engineer the data and pull forward additional benefits.

The first step is to conduct a ‘discovery’ to understand what data each group has available to them and what information is desired but not available and/or easily visible.

After gaps in data mapping are discovered, the utility needs to identify how to provide desired data to its employees. In some instances, it can be as simple as updating permissions and providing access to dashboards and reports. In others, the missing fields may require re-configuration and re-engineering of systems integrations. Data re-engineering can be cumbersome and costly, but is almost always better than the alternative of not realising the value of the data.

The information below provides insight into the process:

1. Identify data users and what they need.
a. Is the data available from existing systems?
b. Does the system of record have analytics tools that can be used by non-experts?
c. Can the data be provided through a third-party analytics tool?

2. Evaluate priorities.
a. Estimate the value (economic and soft) of providing data to each requestor.
b. What are they going to do with the data?
c. Does it add value to the organisation in any way?
d. Depending on the situation, intangible benefits may have greater value than hard-dollar benefits.

3. Make choices.
a. The highest tangible benefits should be addressed first.
b. High intangible areas follow right behind.

4. Accept reality.
a. Sometimes it will be necessary to provide data to support intangible benefits before limited tangible benefits.

Whether the design is completed initially or as part of a re-engineering effort, smart grid implementations involve complex data management and analysis systems to maximize benefits and enable more functionality. Reasonably, a utility can expect that 10-15% of project costs will go towards system integrations efforts. Unfortunately, if initial efforts were not adequate, corrective actions on the backend will incur additional costs.

If a full-fledged data re-engineering effort is required, it is highly recommended to leave the design open and flexible. Use of industry recognised standards such as Multi-Speak can provide more predictable integration frameworks. There are certainly other considerations such as International Electrotechnical Committee’s (IEC) Common Information Model (CIM), specifically IEC 61968 related to system integrations. It is important to assess which standards are best suited to the utility in terms of meeting their requirements and adhering to resource limitations.


Nicole Griffin, Manager, UtiliWorks Consulting, LLC.

Stephen Nees, Senior Associate, UtiliWorks Consulting, LLC