Infrastructure based on supernodes and software agents for the implementation of energy markets in demand-response programs

Abstract

The most successful peer-to-peer networks are based on the concept of supernode, which is an operating point of the network that provides services and advanced functionalities to other nodes. Inspired by this idea, this paper proposes to integrate nodes that provide intelligent advanced services in the future architecture of the electrical grid. Besides facilitating the access to data services such as demand estimations and weather forecasts, these nodes are especially meant to hold virtual environments in which software agents, after being contracted, negotiate on behalf of users in energy markets. This architecture is designed to be compatible with the Energy Interoperation OASIS standard. The capabilities and feasibility of the proposal is demonstrated through realistic experiments based on OpenADR programs, in which users exchange energy by using parallel auction markets. In addition, in order to have the roles of buyer and seller in demand-response programs, thus allowing the creation of markets, a conceptual model based on negative loads and critical loads is provided. The experiments have proven that the proposed architecture facilitates the implementation of advanced distributed management systems in order that smart metering infrastructures, in contrast with traditional agent-based solutions, are released to perform negotiation tasks and access data services, while users gain both autonomy and decision-making capacity.