Data analytics solutions, including disaggregation and energy modeling, provide detailed information and real-time feedback that can inform energy providers' home energy efficiency and demand response programs. Data sources include interval energy consumption data from meter data management systems and devices in the home such as thermostats and major appliances.
The effectiveness of many energy programs is tied to applying this type of technology. More consumers see the benefits of smart technologies, including the cost and energy savings from smart energy solutions. According to Parks Associates consumer research, 11-13% of US internet households own a smart thermostat, and nearly 50% of smart thermostat owners report saving more money than expected from their device.
People want to save energy and money, but they don't know how to find and use energy information. Consumer attitudes are varied, but most need a push or more help to take action. Consumers want better insights on their energy usage, from bills or otherwise.
Disaggregation technology and identifying load
Disaggregation technology is evaluating the energy usage of a residence or building and discerning which appliances or systems are using energy and at what time. Analysis of this data can determine how efficiently any device is used and how well it performs. Supervised and unsupervised disaggregation algorithms classify the appliances based on their load signatures. Data sources can include the utility meter data management system, hardware that polls meter data more frequently (7-second interval), or a third-party device that samples energy consumption at a high rate (kHz sample rate).
Sampling the data at higher data rates, such as a kHz sample rate, allows disaggregation analytics engines to break out even more devices and more refined load identification. Collecting the real and reactive power components also provides more variables to precisely identify loads.
Similar technology can be applied to water and gas meters. Water analytics is a less mature application than energy disaggregation, but new flow meters prove their value in data analytics to determine water leaks and waste. Disaggregation of water data may be used in the future to measure and manage water use more precisely by various tasks. The flow rate and pattern of running the dishwasher are different than the washing machine, the lawn sprinklers, or the shower. Knowing what portion of your water bill goes toward watering the grass would allow consumers to see the impact of changing from watering once every few days to once a week. The information can also alert consumers that they have a small leak that needs investigation or that they are or are not in compliance with watering restrictions.
Predicting and detecting hazards
In addition to using energy disaggregation for usage monitoring, smart meter data can also be analyzed to predict and detect appliance faults that create inefficiencies or hazards. Data-science-as-a-service can enable utilities and retail energy providers to offer AI-powered energy services including fault detection and energy disaggregation. Algorithms can detect, diagnose and predict inefficient electricity usage and behavioral anomalies for major home appliances so electricity providers can provide value-added service to their customers.
Predictive operational analytics help energy providers coordinate distributed energy resources (DERs) with meter, sub-meter, and asset forecasting.
The relatively low price of electricity and decreasing consumption due to gains in efficiency have subdued demand for home energy management systems and services. Government forecasts of energy expenditures anticipate an incremental decrease over the next fifteen years, bottoming out at an average household annual expenditure of $1,932 in 2030 before starting to rise. (Energy.gov. “Appliance and Equipment Standards Program”)
Parks Associates surveys have found that smart thermostats are prized most for their convenience and only secondarily for their energy savings. Marketing of AI-enabled energy solutions must balance messaging around convenience, comfort, and control in the process of saving energy. Efforts by energy providers to drive behavioral change to reach energy efficiency goals have struggled to capture consumer imagination.
Consumers want better insights on their energy usage, from bills or otherwise. Utilities and smart energy platforms, and app-makers are partnering to deliver clear, accurate energy-consumption information to their customers. Connecting this data to the energy bill and cost savings will maximize impact.
Energy analytics will get more interesting to consumers as distributed energy programs offer more incentives with the added value of contributing toward energy independence and even selling energy back to the grid. AI technologies will play a critical role in the new architecture for the smart energy grid that will take years to develop.
In the meantime, leveraging AI to capture and keep consumer participation in demand response programs and encourage load-shifting behavior will be wise investments.
