Support vector based encoding of distributed energy resources' feasible load spaces

Abstract

The sets of feasible load schedules that distributed energy resources are able to operate, jointly define the search space within many virtual power plant optimization tasks. If a centralized approach is considered, a central, single scheduling unit needs to know for each energy resource what schedules comply with all given constraints, because only these are operable and might be taken into account for optimization. As many constraints depend on state or time, sets of currently operable alternatives have repeatedly to be communicated to the scheduler in order to avoid central modeling of each single resource.We here present a support vector based approach for learning a highly efficient geometric representation of the space of feasible alternatives for operable schedules. This description is communicated to the scheduler and the encoded information implicitly contains all constraints and therefore makes their modeling dispensable at scheduler side.

Decentralized Coalition Formation with Agent-based Combinatorial Heuristics

“…In our test setting, each energy unit is associated with a controlling agent responsible for conducting local decisions and communication with other agents. For simulating distributed energy resources we used a model for co-generation plants that has already served in several studies and projects for evaluation (Bremer et al, 2010;Bremer and Sonnenschein, 2013a;Neugebauer et al, 2015;Hinrichs et al, 2013a;Nieße and Sonnenschein, 2015). This model comprises a micro CHP with 4.7 kW of rated electrical power (12.6 kW thermal power) bundled with a thermal buffer store.…”

Encoding distributed search spaces for virtual power plants

“…This description will be based on topological properties. For illustration, we will here discuss the scope geometrically analogue to [7]. Figure 1 shows a simple 2-dimensional example.…”

“…In general it can be stated that the remaining part can usually consist of a set of (separated) regions. In addition, it has been shown in [7] that the scope of action is not always a convex polytope, so that -in general -concavity has to be taken into account, too.…”

“…In this paper, we discuss an approach which aims at learning the boundary that separates valid and invalid (operable and not operable) schedules [7] and present additional simulation results.…”

“…In addition, the scheduling unit would have a need for detailed knowledge of the operator's preferences on allowed and (un-)desirable alternatives. This approach has several drawbacks [7]:…”