Canadian firm BluWave-ai has expanded its Smart Grid Optimizer line with the launch of its Autopilot software-as-a-service (SaaS) to help data centre operators manage power demands in concert with the dynamics of local grid load.

Goldman Sachs Research estimates the electricity demand from data centres worldwide will increase 50% by 2027 and up to 165% by 2030 from 2023 levels. This rapid increase is straining power grids globally. This is resulting in data centres and other large loads being forced to wait until utilities and grid operators can prepare and ensure sufficient generation and delivery capacity.

BluWave-ai Data Center Autopilot will allow for operation with much lower grid connection build-out.

Using patent-pending technology, the Data Center Autopilot moves tasks that can be scheduled to operate at times where grid congestion and energy costs are lower, smoothing out spiky AI workloads while maintaining quality of service and uptime.

“The data centre industry is struggling to balance immense growth with the imperative for sustainability and grid reliability,” said Thomas Triplet, vice president of engineering at BluWave-ai. “Our experience and proven technology for optimising the grid using electric vehicles and renewable energy storage has given us a unique position to solve this problem. The BluWave-ai Data Center Autopilot leverages our patented approach to translate what’s going on in the grid into an actionable signal for data centre application workloads. This means data centres and their associated applications can finally act as a sophisticated, flexible partner to the grid, with up to 35% reduction of the peak electricity utilisation. This effectively enables data centre owners to achieve significant cost savings while lowering their carbon footprint.”

The Autopilot (www.bluwave-ai.com/products-data-center-autopilot) lets application owners who do not have time sensitive applications subscribe to the service and opt in to energy use reduction during grid constrained times and receive a direct revenue stream from BluWave-ai reducing their overall cost of operations. BluWave-ai aggregates multiple applications within a data centre or across data centres in an electricity system operator service area.

The platform provides joint optimisation for all connected assets – on-site batteries, data centre management system and applications – to perform peak demand reduction and energy arbitrage. This delivers a demonstrable RoI with net reduction in total electricity expenditure and measurable improvements in power usage efficiency using existing hardware assets.

By preferentially scheduling workloads to run during periods of high renewable energy availability, it actively helps operators achieve aggressive carbon usage effectiveness targets and meet net-zero sustainability mandates, which are quantified by the platform for ESG reporting purposes.

The system uses a proprietary, predictive signal, BluScore (www.bluwave-ai.com/bluscore), which communicates the real-time cost, carbon intensity and availability of grid power. This signal is consumed by the BluWave-ai AI Optimizer to throttle, pause or shift non-critical workloads, containers or individual applications automatically. This precision ensures quality of service for mission-critical functions is maintained while enabling participation with highly granular demand response and other electricity market services.

The system operates on the BluWave-ai Infrastructure Cube (www.bluwave-ai.com/technology-ev-everywhere-infrastructure-cube), already deployed across Canada to connect seamlessly with utilities, system operators and with BluWave-ai’s other Smart Grid Optimizer products. The platform standardises and handles dirty data from the grid providing more accurate grid predictions for optimal control in concert with other electricity consumers and producers, supporting the power grid. It also coordinates with other distributed energy resources on the grid such as any available on-site battery energy storage at data centres, providing opportunities for participation in demand response and energy arbitrage, further improving net operating costs.

“With the large volume in grid interconnection requested and ramp in loads from existing data centre operators, distribution grids are constrained given the increasing electrification coming from general grid loads and EVs,” said Devashish Paul, CEO of BluWave-ai, “Our Data Center Autopilot will allow non-time-sensitive data centre loads to self-throttle as much as 100 per cent during grid constrained periods. This capability is operational across Canada with EVs and we’re bringing the capability to data centre applications. For a 100MW group of non-time-sensitive applications such as AI model training or other batch processing compute intensive applications, our simulations for a weekly load suggest the potential for a drop from 100MW peak loads to an average of less than 65MW peak”

Founded in Canada in 2017, BluWave-ai (www.bluwave-ai.com) is building an AI company for global energy transition, headquartered in Ottawa. Backed by global investors, the company has raised over $16m, inclusive of a $9.5m series A round.