The Energy industry is going through major disruptions, in the way energy is generated, managed, and consumed. Energy markets are being deregulated, electricity is powering vehicles, and entire communities are turning to microgrids to power their homes. Sustainable, clean energy production has arrived, but only when the concept of wasted energy is addressed, will a sustainable energy ecosystem be attained. That’s the last-mile problem Israeli AI startup Grid4C is solving, with an optimization engine that uses AI and Machine Learning to forecast consumer demand at an extremely granular level, and optimize generation and distributed energy assets at the edge of the energy grid. In simpler terms: beyond accurate forecasting for the next generation of energy consumption.
Grid4C was founded in 2013 by AI expert Dr. Noa Ruschin-Rimini, an alumni of IDF’s elite military unit 8200, and PhD in Artificial Intelligence and Machine Learning, who specializes in predictive analytics and anomaly detection. She gained enterprise managerial experience at Oracle, IBM and other startups, before she founded Grid4C to commercialize her deep-technology expertise.
The startup provides a plug-and-play engine that measures and forecasts energy consumption for distributed generation assets and predicts behaviors at the meter level, for the entire energy value chain, and for all customer classes, from industrial production to residential consumption. Throw data into the engine, like interval usage data from smart meters, and the algorithms build models that predict demand and renewables power generation, to optimize the delicate and increasingly complex balance of supply and demand. In fact, the start-up is already generating billions of predictions every day for millions of smart meters for global energy providers.
“AI is delivering incredible value for global energy providers by leveraging the data they already have” explained Ruschin-Rimini. “The Energy industry has the most IoT data available, in the form of millions of smart meters deployed all over the world… billions of device-reads” added Ruschin-Rimini. Such a wealth of data is genuinely tough for a startup to get a grip on. Grid4C is already working today on four continents.
According to Mackinnon Lawrence and Jan Vrins of Navigant Research, “The prolific rise of renewables and distributed energy resources, behind-the-meter smart devices, digital infrastructure, advanced controls and analytics, and changing customer demands are ushering in a new era of highly networked power.” These new realities require granular forecasting techniques, which predict behaviors for every consumer, identify individual loads and appliance patterns, forecast granular load profiles at each premise, and optimize and coordinate renewable and distributed energy resources to maximize grid efficiencies.
A notably fascinating partner for them are currently smart meter vendors, like global leader Landis+Gyr. “Smart meter vendors are reselling our AI solutions and embedding our algorithms into their smart meters, bringing insights onto the device and to the consumer home in real time, on the edge of the grid” explained Ruschin-Rimini. “Many of the industry’s most advanced use cases can now be solved with greater intelligence at the edge of the grid, and in real-time” she adds.
The startup employs its AI-powered energy analytics to predict consumption patterns from consumer data derived from interval usage data from smart meters. They aggregate these forecasts up to assets on the distribution grid, to predict potential load issues and outages for transformers and other grid assets. After years of work and smart meter data from millions of homes on multiple continents, they have developed disaggregation capabilities that identify individual appliance loads within every premise, without the need to install equipment or sensors inside the home or business. Additionally, they have developed classifier algorithms, which can detect faults and inefficiencies for those appliances (i.e. HVAC, water heaters, pool pumps, refrigerators) as well as the costs and usage patterns, even predicting equipment malfunctions before they occur. It can profile residents’ energy consumption patterns across the myriad devices at home, saving them money, generating comparative efficiencies, and delivering actionable insights.
Perhaps the most promising innovation of all, is the combination of smart meter data with AI-powered analytics and machine learning in real-time at the grid edge, which is solving the most challenging use cases in the industry, with automated decision making closest to every consumer. Future smart meters will be able to identify individual loads and appliances within each premise, predict, detect, and diagnose faults and inefficiencies for home appliances and alert customers to anomalies before they break, provide early warnings of gas or water leaks, and even optimize charging for electric vehicles, based on customer patterns, energy pricing, and needs of the electric distribution system.
The startup has patented their AI algorithmic framework for identifying abnormalities in energy usage of households. And they’ve recently raised a series funding round of $5 million, bringing the total funding to $6.5 million and they’re operating in the US, Israel, Australia and Asia. They’ve completed many installations and currently service over 16.5 million consumers across their client installation base, such as Direct Energy, China Light & Power, Exelon Corporation, and Southern Company. They monetize their AI solutions by charging a monthly license fee based on the number of smart meters it delivers insights for.
The future is shimmering green with Grid4C’s AI based solution at work. It’s not a question of if, but of when the energy industry will reach full sustainability. Until then, we can appreciate how AI is solving the industry’s most challenging use cases, thanks to Dr. Rushin-Rimini and her team’s hard work.