Optimal Operation Strategy of Virtual Power Plant Considering Real-Time Dispatch Uncertainty of Distributed Energy Resource Aggregation

Abstract

Virtual power plant (VPP) technologies continue to develop to embrace various types of distributed energy resources (DERs) that have inherent real-time uncertainty. To prevent side effects on the power system owing to the uncertainty, the VPP should manage its internal resources' uncertainty as a whole. This paper proposes an optimal operation strategy for a VPP participating in day-ahead and real-time energy market so that a distributed energy resource aggregation (DERA) can cope with real-time fluctuation due to uncertainties while achieving its maximum profit. The proposed approach has bidding models of the DERAs including microgrid, electric vehicle aggregation, and demand response aggregation, as well as the VPP. The VPP determines internal prices applied to the DERA by evaluating its real-time responses to the day-ahead schedule and updating proposed pricing function parameters, and the DERA adjusts its energy reserves. By repeating this coordination process, the VPP can establish an optimal operation strategy to manage real-time uncertainty on the DERA's own. The effectiveness of the proposed strategy is verified by identifying a capability of the DERA to cope with real-time fluctuation through scenario-based simulations. The result shows that the VPP can reduce 1.6% of cost while the internal price applied to the DERA is close to the maximum.

Network‐constrained unit commitment‐based virtual power plant model in the day‐ahead market according to energy management strategy

“…Ref. [15] introduced a strategy for the optimal operation of a VPP that is involved in day-ahead and real-time energy markets. The aim was to maximize the profit gained by the DER aggregator by dealing with the oscillations caused by uncertain parameters.…”

Review on the Optimal Configuration of Distributed Energy Storage

“…This application mode has been applied in important load users, especially in medicine, advanced electronic manufacturing, data centers, chemical fiber production and other industries, with obvious benefits [75]. In order to minimize load loss during a power outage and guarantee production, life safety and economic property, the joint operation method of mobile energy storage, distributed energy storage resource aggregation technology and post-disaster recovery strategy using mobile energy storage have become new mainstream research directions [76,77]. This also reflects the demand of the current new power system to improve the resilience of the distribution network, that is, the ability to prevent and recover from extreme events.…”

Distributed Energy Scheduling Problem Based on Learning Algorithm

“…(1) Decision tree algorithm Decision tree belongs to supervised learning, which is a kind of classification algorithm, which presents the final classification result in the form of tree flow chart. Decision trees explain the response results of categorical variables [6]. Essentially, a decision tree is to classify a series of regular data.…”