Power Generation Technology ›› 2019, Vol. 40 ›› Issue (2): 115-121.DOI: 10.12096/j.2096-4528.pgt.18243

• Energy Internet • Previous Articles     Next Articles

ADMM Algorithm for Distributed Computation of Block Chain System and Application in Active Distribution Network

Dongdong ZHOU1(),Yibo LI2(),Anqi YANG3,Tao SHENG4   

  1. 1 School of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, Jiangsu Province, China
    2 State Grid Zhejiang Maintenance Branch Company, Ningbo 315020, Zhejiang Province, China
    3 State Grid Haian Power Supply Company, Nantong 226681, Jiangsu Province, China
    4 State Grid Jinhu Power Supply Company, Huaian 211600, Jiangsu Province, China
  • Received:2018-11-28 Published:2019-04-30 Online:2019-05-09
  • Supported by:
    Jiangsu Postgraduate Practical Innovation Plan(SJCX18_0578);Nanjing Institute of Technology's Innovation Fund(TZ20180032)

Abstract:

As a novel computer technology, block chain has great potential in the future active distribution network due to its advantages. Therefore, how to integrate the centralized optimization task of distribution network into the distributed computing module of block chain system is an urgent problem to be solved. In this paper, the distributed model of dynamic economic dispatch optimization of distribution network was taken as an example. With the help of convex relaxation theory of second-order cone programming (SOCP), the traditional coupling model of distribution network was transformed into a decentralized electrical model by referring to CIM in IEC61970 standard. A distributed optimal power flow solution method based on alternating direction method of multipliers (ADMM) algorithm running on block chain was proposed, which can solve many problems faced by traditional centralized optimization in future active distribution network. The proposed method does not need global coordination or hierarchical partitioning of distribution network, but only a small amount of communication between adjacent nodes of block chain was used to obtain the global optimal solution of the model through parallel computation.

Key words: distribution network, optimal power flow, distributed optimization, relaxation technique