Try to square the circle. Could one train an artificial intelligence that understands dynamic systems, by providing data from the largest machine humankind has ever built: the interconnected power grid? If we could, what would it do? What would we do with it?

After 150 years of electrification, 1 billion people still have no access to modern energy. The immediate problem that can be solved within this first dynamic system (the power system) is obvious: adaptive integration of non-dispatchable renewable energy. Provide usable energy at nearly zero marginal cost: < free > electricity. But how do we get there?

"If we knew what it was we were doing, it would not be called research, would it?" - Albert Einstein.

Train to understand. We formulate hypotheses, small and big, and aim to be less wrong over time. We are experimenting using an energy data oracle for smart contracts we are developing as part of our products: Energy Solidarity Token and Electraseed Fund.

Data-driven insights and optimization. AdptEVE learns based on high resolution local energy data, which should allow us to obtain valuable insights on Digital Energy Asset conditions and to optimize their performance.

Support the development of AdptEVE: enter an Energy Data Challenge, join our Tensor Computing for IoT Community.

Envisioned Benefits

  • Reduce capital costs of combined energy assets
  • Achieve continuous efficiency gains during operation
  • Tailor energy products & services to the user