Analysts in the utility industry have much work to do these days. With the deluge of data that has accompanied the smart grid evolution, analysts are leveraging the billions of rows of data and working to transform it into actionable insight. Analytics that utilizes data from the smart meters, coupled with data from a smart distribution system, has empoweredand will continue to empowerutilities to answer operational questions and solve problems with greater speed and accuracy.
At SMUD, there are three categories of analytics that we are pursuing in our evolving smart grid world, including consumer analytics, operational analytics and enterprise analytics. Business intelligence and enterprise analytics include revenue protection and theft detectionor energy trading with a live look at the energy grid. Examples from the operational analytics include fault detection and outage management, which are typically employed in the grid planning and operations department.
At SMUD, as part of our Smart Sacramento initiative, we deployed a situational awareness and visual intelligence software solution that correlates, analyzes and visualizes data in smart grid, distribution and outage systems to improve the decision making across our grid planning and operations department. The real-time data provides SMUD with the ability to collaborate as one team to respond rapidly to emergency situations and outages, and more readily understand the real-time impact of weather, fires and emergencies on our daily operations.
Consumer analytics in the utility realm is another area of analytics that SMUD is making great strides in. Consumer analytics includes the broad realm of behavior analytics, demand response and time-of-use pricing, DG/EV/microgrid analytics, and market research. For SMUD, analytics in this area combine meter data, demographic data and utility program data to inform program design, rate design and marketing that improves the customer experience and utility initiatives.
An example of customer analytics at SMUD includes a segmentation study our project team conducted in 2013 with the load profile data collected from the smart meters. The customers that were included in the analysis were either participants in our time variant pricing pilot, SmartPricing Options (SPO), or were in the study control group. The goal of the analysis was to determine if there were demographic similarities among customers with similar load shapes when on the rates. What could we learn from different customers load shapes when on time variant rates? Were there patterns we could identify?
We conducted a cluster analysis on the load data to normalized load shapes for customers participating in the SPO pilot. We then attached the results from a demographic survey to each cluster of load shapes. From each of the rate groupstime-of-use (TOU), critical peak pricing (CPP), and TOU/CPPwe began identifying distinct groups based on load shapes and similar demographics. Within each rate treatment, we were able to identify load shapes that followed very distinct usage patterns, such as dramatic load reductions during peak hours or load shifting. When looking at the demographic attributes of each cluster group, we were able to categorize large similarities for each cluster. For example, for customers on the CPP rate, the customer cluster with the largest peak time reductions was comprised of 81 percent single family homes, 93 percent English speaking, higher income households, with 80 percent of the households reporting some college education or higher.
The segmentation study allowed us to understand the characteristics of our pilot customers in relation to their load shapes. This allows us to identify opportunities to reach customers in new ways, such as providing educational materials in other languages or information that specifically addresses multifamily households. In addition, we are able to identify the characteristics and similarities of participants that are engaged with SMUD and participating in programs and pilots. We can leverage this information to tailor our programs and educational materials to better serve our customers and provide them with tools and information to better manage their energy use.
A long-term strategy of conducting customer segmentation and customer analytics that combines load profile data with demographic and end-use information can help SMUD design time-based rates that empower the customers on an individual level, rather than providing mass pricing plans that fit the average customer. Through this type of analysis, we are able to identify similarities among customers based on their responses to price signals. Eventually, as the data we collect becomes more dynamic and instantaneous, we can use this information to understand which groups of customers are more capable of producing large or small load reductions based on demographic and housing characteristics, which can help SMUD implement real time pricing programs that benefit the utility and customer. In essence, customer analytics that combine individual load profiles, time variant pricing and demographic information help SMUD to create pricing and energy resource management that is responsible to SMUDs customer-owners.
Engaging the customer has never been as importantor as possiblethan in the smart grid era, and tapping into the oftentimes disparate data sources to integrate that information into actionable information for decision making is one of the larger challenges for utilities. Analytics requires utilities to break down organizational silos that hinder data sharing, integrate information systems, and improve data platforms for handling unstructured data. All of these tasks require utilities to rethink their data management architecture and capabilities so that we can become better equipped to provide solutions and tools to our customers that empower them to manage their energy use.
View all SMART GRID Bulletins click here
Enter your email-id to subscribe to theSMARTGRID Bulletins
04 September 2017
05 September 2017
16 September 2017