Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Novel Dynamic Peer-to-Peer Clustering Algorithm and Its Application to Aggregate Energy Storage Systems

Runfan Zhang, Branislav Hredzak

TL;DR

The paper presents a fully distributed online clustering algorithm for multi-agent energy systems, addressing limitations of centralized/offline clustering in microgrids. It uses pre-selected feature states and distributed average estimates to form dynamic groups and employs auxiliary states to support control tasks, all through a neighbor-to-neighbor communication network. Two microgrid applications illustrate its value: reducing power losses by clustering batteries with similar loads and capacities, and creating a virtual energy storage by clustering based on price and capacity. The method offers robust, scalable clustering that adapts in real time, with potential applicability to a wide range of distributed control problems beyond energy storage.

Abstract

The proposed distributed dynamic clustering algorithm enables to group agents based on their pre-selected feature states. The clusters are determined by comparing the distance of the agents' current feature states with average estimates of the states in all clusters. The algorithm also provides average estimates of pre-selected auxiliary states that can be utilized for control purposes. Two example applications of the algorithm are introduced. In the first application, the algorithm is applied to a microgrid with distributed batteries that are controlled to achieve a common state of charge within a group. However, a random selection of the batteries' groups results in additional power losses during operation. The algorithm reduces the power losses by clustering the batteries based on the selected feature states: local loads and battery capacities, while the state of charges and output voltages are selected as auxiliary states for control purposes. In the second application, the algorithm is used to form a virtual energy storage from batteries distributed in a microgrid.

A Novel Dynamic Peer-to-Peer Clustering Algorithm and Its Application to Aggregate Energy Storage Systems

TL;DR

The paper presents a fully distributed online clustering algorithm for multi-agent energy systems, addressing limitations of centralized/offline clustering in microgrids. It uses pre-selected feature states and distributed average estimates to form dynamic groups and employs auxiliary states to support control tasks, all through a neighbor-to-neighbor communication network. Two microgrid applications illustrate its value: reducing power losses by clustering batteries with similar loads and capacities, and creating a virtual energy storage by clustering based on price and capacity. The method offers robust, scalable clustering that adapts in real time, with potential applicability to a wide range of distributed control problems beyond energy storage.

Abstract

The proposed distributed dynamic clustering algorithm enables to group agents based on their pre-selected feature states. The clusters are determined by comparing the distance of the agents' current feature states with average estimates of the states in all clusters. The algorithm also provides average estimates of pre-selected auxiliary states that can be utilized for control purposes. Two example applications of the algorithm are introduced. In the first application, the algorithm is applied to a microgrid with distributed batteries that are controlled to achieve a common state of charge within a group. However, a random selection of the batteries' groups results in additional power losses during operation. The algorithm reduces the power losses by clustering the batteries based on the selected feature states: local loads and battery capacities, while the state of charges and output voltages are selected as auxiliary states for control purposes. In the second application, the algorithm is used to form a virtual energy storage from batteries distributed in a microgrid.

Paper Structure

This paper contains 11 sections, 4 equations, 3 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. Distributed Clustering Algorithm
  3. Distributed Communication Network
  4. Dynamic Clustering Algorithm
  5. Utilization of Clusters
  6. Applications of Algorithm to Cluster Batteries in Microgrid
  7. Power Loss Reduction
  8. Local Loads
  9. Battery Capacity
  10. Virtual Energy Storage
  11. Conclusion

Figures (3)

  • Figure 1: Distributed clustering algorithm for the $i$-th agent. The double line arrows represent distributed communication.
  • Figure 2: Illustration of the proposed clustering algorithm. Nine agent nodes are clustered into three groups. The black lines represent the distributed communication network between agent nodes. The colored dash lines indicate virtual communications within a cluster. One color represents one cluster.
  • Figure 3: The average estimations of the auxiliary state $z_i$ for the $i$-th agent. The double line arrows represent distributed communication.