Grid4C’s algorithms can detect, diagnose, and in many cases predict malfunctions, inefficiencies and behavioral anomalies for many home appliances, without the installation of any hardware and just by analyzing smart meter data. These issues are identified for major home appliances, including HVAC systems, water heaters, pool pumps, the home envelope, and refrigerators.
By analyzing smart meter usage data from each premise, Grid4C’s algorithms can break down energy usage for customers for each appliance in a non-intrusive manner, and alert customers when anomalies are detected for unusually high appliance usage or costs.
The machine learning engine monitors and learns the usage patterns of each customer in various conditions (weather conditions, holidays, day of week, time of day etc.). Based on these usage pattern profiles, the actual usage behavior of the customer is monitored in order to automatically detect usage deviations.
Grid4C’s algorithms build predictive models for every customer that predict how customers will behave and how much energy each of their appliances will consume for the next 31 days, normalized for weather. Customers can view forecasted daily costs and electricity usage for the forthcoming week. The models also generate daily and weekly forecasted electricity usage in one hour intervals.
By combining customer usage data with smart thermostat data, we can help customers optimize thermostat settings, simulate how any temperature setting will impact a customers energy bill, and predict and detect faulty home appliances to deliver real customer value.
Utilities can understand appliance ownership for every premise, and proactively provide personalized notifications that certain appliances are expected to have high usage before a high price tier, e.g. pool pumps that are running during high price periods.
Grid4C has developed innovative Chatbots, which are automated interactions through a chat window that use machine learning to ask customers 2-3 questions to diagnose or troubleshoot an issue when there are multiple possible explanations.
Load forecasting at the meter/transformer level provides accurate load forecasts at the meter and sub-hour levels. Bottom-up aggregations and top-down disaggregations enable the forecast at any level of granularity (appliance/ transformer/feeder/ industry/territory level, etc.). Advanced visualization tools enable the end user to self-explore the data, run simulations and what-if analysis and more.
Based on the ability to forecast solar power generation and the electric consumption at the premise on a sub-hourly level, Grid4C’s DER Optimization solution forecasts and optimizes Distributed Energy Resources, which is a critical need for utilities. Together with the ability to automatically disaggregate smart meter reads to the appliance level usage in sub-hour granularity (including the usage of EVs, pool pumps, HVACs and more), this solution helps utilities with short and long term planning needs that improve reliability, lower costs, and optimize grid assets.
It also simulates the impact of Demand Response events, to target the optimal customers for a control event and proactively provides notifications that certain appliances are expected to operate in advance of DR events.
An adaptive automated engine provides accurate solar power forecasting based on proprietary machine learning algorithms, can detect inefficiencies with solar panels, and maximize profitability of solar usage against the predictive model for each home for customers with solar.
A solution that takes into account customer parameters, consumption behavior profiles and patterns, appliance ownership (provided by the disaggregation algorithms), together with spatial data, in order to build bottom-up customer segmentations to personalize pricing and marketing offerings.
A solution that takes into account customer parameters, consumption behavior profiles and patterns, appliance ownership (provided by the disaggregation algorithms), together with spatial data, in order to build bottom-up customer segmentations to design new rate plans for customers.
The Electric Vehicle solution detects EV ownership and disaggregates the EV load at a sub-hourly level using AMI data. It also creates clusters of customers based on EV load patterns generated from the bottom up, and predicts EV usage at the meter level, aggregated to the grid asset level, to help stabilize the grid.
A solution that analyzes customer parameters, consumption behavior profiles and patterns, responses to previous offerings, and spatial data, in order to predict each customer's probability to accept any energy efficiency and marketing offering and provide customer root-cause analysis.