Distributed generation of energy coming from various vendors, even private homes, is a big challenge for tomorrows power management systems that, unlike today, will not dispatch energy centrally or under central control. On the contrary, the production, distribution and management of energy will be treated and optimized in a distributed manner using local data. Even today, parts of the power system are highly nonlinear with fast changing dynamics. It is hard to predict disturbances and undertake countermeasures on time. In existing approaches electricity is distributed to the final users according to its expected estimated demand. Such non-dynamic approaches, are difficult to evolve and can not accommodate rapid changes in the system. By having a cross-layer and open information flow among the different actors involved we can make better and more timely predictions, and inject new dynamics in the system that will lead to better energy management and achieve better energy savings. The NOBEL project is building an energy brokerage system with which individual energy prosumers can communicate their energy needs directly to both large-scale and small-scale energy producers, thereby making energy use more efficient.