Offering a New Use for Autoencoders in the Utility Industry

By Adam Nathan, Jason Lu, Jonathan Hahn, Kevin Wald and Rich Lack

Each day, utility companies use smart meters to collect an enormous amount of data on energy usage. In this white paper, you will learn how to cluster customers based on patterns in their electricity usage, as well as be presented with potential service enhancements utility companies may be able to make based on results.

Learn more about our work in AI research and development.

Abstract

Smart meters used by electric utilities collect an enormous amount of data on energy usage each day. Leveraging this data, we can better understand how individual consumers use their electricity by clustering customers according to their pattern of electricity usage. Clustering, however does not perform well on high-dimensional data such as meter readings. To overcome this, a type of deep neural network called an autoencoder can reduce the dimensionality of the meter reading data to a representative set of relevant patterns. Clustering then can be performed on the lower dimensional, representative data to group together consumers with similar patterns of usage. This clustering is a fully unsupervised technique. In general, this process of reducing dimensionality through auto-encoding and subsequent clustering may be used to characterize a variety of high-dimensional data sources other than electric meter readings.

Login or register to view the rest of the white paper.

  • Obscuring and Analyzing Sensitive Information with Generative Adversarial Networks

    In this white paper, we explore the feasibility of generating representative data for two types of data: binary input from medical records and real-valued sensor data from industrial mining trucks.
  • Image Classification of Race Cars

    In this paper we explore the use of artificial intelligence to perform an image sorting task to gain a competitive edge in a NASCAR race.
  • Recommendation Engine: ATC Connect

    A hybrid recommendation engine was built for recommending relevant articles to the users of the WWT ATC Connect mobile application.
  • Deep Learning with NVIDIA DGX-1

    We have highlighted some practical considerations for the Deep Learning practitioner relevant to neural network training on the NVIDIA DGX-1. Benchmarking experiments showed that GPU performance is related to three dimensions.
  • Machine Learning for Shovel Tooth Failure Detection

    Current computer vision-based methods for identifying broken teeth on mining shovels suffer from a prohibitively high false positive rate (FPR). We describe a two-stage methodology for the detection of broken teeth that reduces the FPR.
  • Infographic: Data-driven Insights Create Richer Experiences

    WWT helps retailers unsilo data sources for insights that improve the brand experience, cement customer loyalty and boost ROI. See how in this infographic.
  • FlexPod Outperforms JBOD for Hadoop at Scale

    This white paper explores the FlexPod Select architecture for Hadoop versus JBOD (“Just a Bunch of Disks”) architecture from both a technical and financial perspective.
  • Solution Guide: Big Data

    In this practical guide, we take an in-depth look at the challenges, use cases, architectural implications, technologies and strategies that must be considered for a successful big data implementation.
  • Executive Guide: Big Data and Hadoop

    This is an executive-level book that introduces you to big data, describing what it consists of and what’s driving its momentum, including methodologies to harvest knowledge from structured and unstructured data.
  • Brochure: Big Data Infrastructure

    This brief will help you navigate the crowded technology landscape of new software and hardware products and determine what is "noise" and what may be relevant to you.
  • How Utilities Can Derive Value From Sensor Analytics

    Learn how utilities are tapping their data resources to directly impact business models, operations, processes and revenue through big data analytics.
  • Brochure: Big Data Practice

    Our service and solution offerings allow customers to engage at any point on the big data roadmap and are tailored to meet organizational needs.