Discovery Green is betting that the next problem in renewable energy will be deciding what to do with electricity that changes by the hour. The company has launched EnergyOS, a proprietary platform designed to manage the uncertainty created when wind and solar generation, customer demand and electricity prices move independently of one another. Its proposition is that renewable power becomes more commercially useful when those variables are managed together rather than locked into static procurement arrangements.
EnergyOS starts by forecasting three things at the same time: how much renewable electricity different assets are likely to produce, how much electricity each customer is likely to consume and what market prices could look like. But instead of treating one forecast as the answer, the system runs multiple scenarios to understand the range of possible outcomes. That matters because a business buying renewable electricity is exposed to more than whether a wind farm or solar plant generates power. A mismatch between generation and consumption can leave the customer buying additional electricity from the market or carrying surplus power at a different value.
The platform then brings the customer’s circumstances into the calculation. A mine, factory or commercial operation does not have unlimited freedom to change when it consumes electricity, and different businesses have different tolerance for price volatility and different emissions targets. EnergyOS therefore uses those constraints to determine how renewable supply should be allocated. Discovery Green says its allocation model can then move electricity across its customer portfolio, using the combined diversity of generation assets and loads to reduce the financial impact of individual fluctuations.
The important part is what happens after the initial allocation. EnergyOS is supported by what Discovery Green calls EnergyAI, a feedback system that compares forecasts with what actually happened — generation, customer consumption and market prices — and uses that information to adjust future decisions. In effect, the platform is designed to learn the behaviour of the energy portfolio rather than assuming that yesterday’s forecast will remain accurate tomorrow. The larger the portfolio becomes, the more opportunities there are to offset one customer’s shortfall with another customer’s different consumption pattern or another generating asset’s output.
Read Also: Why Egypt Is Linking Its Wind Boom to a Turbine Manufacturing Industry
Discovery Green says internal simulations produced 105% more savings, 88% lower financial-outcome volatility and roughly 90% emissions reduction compared with traditional renewable procurement, but these are modelled results rather than audited realised performance. The commercial test will come as EnergyOS is deployed across new and existing customers, particularly large electricity users such as mines. That is where the proposition becomes more interesting: renewable energy is moving from a simple question of adding generation capacity to a question of managing an increasingly complex electricity portfolio. EnergyOS is essentially an attempt to build the software layer that sits between renewable generation and the businesses trying to use it.
