Professional Documents
Culture Documents
EPON Communication Network Topology Optimization Based On Electrical Nodes' Importance in Active Distribution Networks
EPON Communication Network Topology Optimization Based On Electrical Nodes' Importance in Active Distribution Networks
EPON Communication Network Topology Optimization Based On Electrical Nodes' Importance in Active Distribution Networks
EPON Communication Network Topology Optimization Based on Electrical Nodes’ Importance in
Active Distribution Networks
Yingqiao He1,2, Li Ma2, Lingfeng Wang2, Junjie Tang1, and Rao Rao1,2
1The State Key Laboratory of Power Transmission Equipment & System Security and New Technology, College of Electrical Engineering᧨
Chongqing University, 400044, China
2
Department of Electrical Engineering and Computer Science, University of Wisconsin-Milwaukee, Milwaukee, WI 53211, USA
Email: l.f.wang@ieee.org
Abstract— A large amount of DGs’ (distributed generations) hand, ICT (Information and Communications Technology)
accessing to the distribution network is a characteristic of the system’s random failures may lead to adverse consequences
present and future power distribution systems. According to the such as operational condition deterioration and widespread
actual operation state of power system, distribution network blackouts. The active distribution network can be regarded as
actively manages the massive distributed generations and a typical Cyber Physical System (CPS).
adaptively adjusts the network to satisfy the requirements of After the access of a large number of distributed energy
economical and safe power supply both in the physical network
and cyber network. This paper aims to propose a feasible
and intelligent user terminals, the distribution network takes
the responsibility of not only reliable electricity supply to
planning optimization model for the communication network
loads but also effective consumption of new energy. Thus,
based on the evaluation of all electrical nodes’ importance in the
physical distribution network. This solution not only meets the
this paper proposes a new importance index for electrical
current technical requirements, but also considers the possible nodes on the physical side. Focusing on the topology of cyber
side based on the node importance of physical side, this paper
expansion of communication network in the future. Based on
provides a communication network optimization model
the existing single-link network scheme for all the
considering reliability and economy simultaneously.
communication nodes, a certain portion of the important nodes
in the distribution network is chosen to be upgraded to dual-
link connection. The optimization result can provide a more DISTRIBUTION NETWORK IN POWER SYSTEM
economical solution for distribution network operators to build A. Distribution automation system structure
the corresponding communication network.
Distribution network locates at the end of power system.
Due to its direct connection to users, the quality and ability
Index Terms—cyber-physical system; active distribution
of power supply are greatly influenced by it. This paper pays
network; EPON; node importance; topology optimization;
complex network. more attention on 10kV distribution network. For the
convenience of management and maintenance, distribution
INTRODUCTION
automation system usually adopts hierarchical structure.
The emergence of new elements in power grids such as
distributed generators, energy storage devices, and flexible The communication channel is a communication
loads that have the ability to control itself according to the network that connects the main distribution station and the
users’ demands poses new challenges to the operation and distribution substation, and transmits automation
planning of distribution networks. The distribution system information between them. In particular, the communication
will transfer from the traditional power distribution network channel acts as the nervous system of the distribution
to current power exchange system, which takes the automation system, and is responsible for the uploading and
responsibility of power production, storage, transmission and dispatching of various monitoring information and control
distribution. These new characteristics bring profound commands. It is an important basic part for realizing
influences on the pattern of power supply in the distribution distribution automation, which leads to the discussion of the
network. Under the circumstances, the communication key issues in this paper—the communication network
network planning is more complex and important, and topology in the distribution automation system will be
requires planners’ systematic considerations. Planners should elaborated in the next chapter.
Obviously, in this heterogeneous network, if the
ensure the reliability of the distribution network and
corresponding distribution communication network on the importance of the node with a high degree of importance is
attacked firstly, and this will cause the connectivity
premise of guaranteeing a secure and stable power system
considering the load growth and power planning scheme performance and reliability of network to drop sharply, and
during the planning period [1-4]. even cause the corresponding cyber network cause serious
Structurally, CPS (Cyber Physical System) can be divided problem. Therefore, it is extremely important and necessary
into two major parts: cyber subsystem and physical to optimize the network for the CPS of the distribution
subsystem, which interact with and interdepend on each other network at the planning process to identify the nodes or
[5]. In CPS, the cyber subsystem represents the information regions with high importance [10].
world including discrete computing processes, logical B. The indicator of node importance in distribution
network
communication processes and feedback control processes; the
The distribution network is a complex structure
physical subsystem represents processes, objects or
components in the physical world [6]. On the one hand, timely consisting of lines, equipment and multiple loads. In order to
extract key information of distribution network, it needs to
and accurate transmission and decision-making are the basis
for enabling a variety of active control functions; on the other be simplified, which means that a simple topology composed
1
of nodes and lines is used to describe a complex distribution COMMUNICATION NETWORK OPTIMIZATION
network. The distribution network could be modelled as a C. Introduction of distribution communication network
graph ܩൌ ሺܸǡ ܧሻ, where ܸ is a set of nodes and ܧis a set of and EPON technology
m edges. Due to the wide variety of distribution network points,
Each node of the distribution network may be connected complex topological structure and poor operating conditions,
to distributed generations, energy storage devices, and loads planning of topology has become the key and difficult
with different types and importance [7]. This paper not only problem in the construction of distribution network
considers these factors in the new indicators, but also communication system [8]. Passive optical network (PON) is
an optical communication network composed of many
considers the contribution of this node to the entire
passive optical components using pure optical transmission
distribution network. The node importance indicator is medium, and has excellent transmission characteristics of
expressed as follows: optical fiber communication. From the perspectives of
I node (i ) = α ∗ Finput (i ) + β ∗ Fourtput (i ) + δ ∗ Δ E (1) transmission speed and transmission distance, PON has
incomparable virtues, and it has gradually become the
where, ߙ, , is the correction factor, and α + β + δ = 1 . mainstream technology in several countries. Now many
operators have started to increase the investment in the
Finput (i ) = Gi / G max ; ܩ୧ is the amount of distributed
construction of passive optical network. EPON/GPON have
generation connected to node i; ܩ୫ୟ୶ is the largest become mature [11]. EPON (Ethernet passive optical
distributed generation among all the nodes; network) adopts PON technology in physical layer, using
Foutput (i ) = Li / Lmax ; ܨ௨௧௨௧ ሺ݅ሻ represents the total amount Ethernet protocols in link layer and PON topology structure
to realize Ethernet access. This paper uses EPON, which
of load, which connected to node i; ܨ௨௧௨௧ ሺ݅ሻ reflects the
reduces the cost of network construction and operation,
load of node i; ܮ୧ is the load of node i; ܮ୫ୟ୶ is the largest load facilitates network construction and expansion, and provides
E (Y ) − E (Y − 1) more users with flexible access options. EPON is a single-
among all the node; ΔE (i ) = , E (Y ) is
E (Y ) fiber and bidirectional Optical access network from point to
multiple points. It consists of Optical Line Terminal (OLT)
original global efficiency, E (Y − 1) is the global efficiency
installed in Distribution station, Optical Network Unit
when one node is removed, ΔE (i ) is the efficiency change (ONU) installed in Distribution Terminal and Optical
after removing node i. Distribution Network (ODN) in Distribution Network. ODN
can connect one OLT and multiple ONUs, which consist of
2 ˄dij - d˅ vij -1 main optical cables, splitters and branch optical cables, thus
E (Y ) = ∗ ¦ i∈{G}, j∈{ D} exp˄ ∗ ˅ (2)
n ∗ (n − 1) d vij ,max providing bidirectional transmission of optical signals. The
structure of EPON is shown in figure 2. In this paper, the
where, node i belongs to generator set {G} and node j location of OLT is fixed. The planning is focusing on the
belongs to load set {D}, and other nodes belong to set {C}. topology of ODN.
n is the number of nodes in power grid, ݀ҧ is the average ODN
ONU1
End
OLT ONU2
2
ேೀಿೆ
E. Model of communication network σୀଵ ܤ ͻ اͲܯȀ (9)
1᧥ Objective function The calculation formulation of total flow for all users
Min C=F᧤ܰைௌ ,, ᧥ܦ (3) connected to the PON is as below:
௧
The total cost C of an EPON communication network σே ೀಿೆ ேೀಿೆ
ୀଵ ܤ ൌ σୀଵ σ୫ୀଵ ߮
(10)
can be expressed as:
CASE STUDY
C =ܰை் *ܥை் +ܰைே *ܥைே +ܰைௌ *ܥைௌ +᧤ܥ כ ܦଵ ܥ כଶ ᧥
(4) F. Case study of distribution network
where, ܰை் ǡ ܰைே ,ܰைௌ is the number of OLT, ONU and 1᧥The test system
POS, respectively. ܥை் is the cost of one OLT; ܥைே is the According to ܫௗ ሺ݅ሻ , the important nodes in
cost of one ONU,ܥைௌ is the cost of one POS;ܰை் is the distribution network with various distributed generations
amount of OLT. ܦis the total length of optical fiber setting will be found. This section selects the IEEE 33-bus system
through the existing cable channel; is the total length of as the experimental object. The structure of this system is
optical fiber setting through the newly building cable trench; shown in figure 4. In addition, a time-span data is used in the
ܥଵ is the cost of the unit length optical fiber, ܥଶ is the unit evaluation process, which is the average of typical scenes.
cost of optical fiber and the installation cost of unit length These nodes with blue lines represent the node is connect to
optical fiber on the new cable trench. the distributed generations in the distribution network.
2᧥Constraints
Constraint 1: reliability
The reliability of the communication network is that the
communication path could connect from ONU to OLT
without interruption. The path of the communication network
is a collection of nodes and links, and the availability of one
path represents that all nodes and links on this path are
available. If the two nodes have higher redundancy, it Fig.4. The topology of distribution network
indicates that the connectivity of these two nodes is stronger
and the reliability of their communication is higher. If an 2᧥Simulation results and analysis
ONU has more than one path to OLT, the ONU is more
reliable from the perspective of topology. We should make
some important ONUs have higher reliability. The reliability
of path from ݅௧ ONU to OLT:
Ri = rONU ∗ (1 − ∏ t =1 (1 − ∏ s =1 rt , s ))
T S
(5)
3
As shown in figure 6, the original communication 23 24 25 26 27 28 29 30 31 32 33
those locations has been existed. The orange lines present the 19 20 21 22
33 to node 13), which means the optical fiber cross this new
cable trench. This result is based on the proposed 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 34
one OLTs. 19 20 21 22
23 24 25 26 27 28 29 30 31 32 33
Fig.11.The communication planning scheme in ratio=0.6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 34 23 24 25 26 27 28 29 30 31 32 33
19 20 21 22
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 34
19 20 21 22
23 24 25 26 27 28 29 30 31 32 33
23 24 25 26 27 28 29 30 31 32 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21 22
The Ratio of
0.2 0.3 0.4 0.5 0.6 0.7
important node
23 24 25 26 27 28 29 30 31 32 33
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
19 20 21 22
In the communication network, the most important
factor is the connectivity between the ONU and OLT. In this
Fig.8.The communication planning scheme in ratio=0.3
figure, the value of the reliability is based on all edges in
communication network. In figure 13, it is clear that the
reliability of whole communication system has been
significantly improved after optimizing the topology of
communication network using the proposed model in this
paper. At the very beginning, the reliability of original
4
network is 0.9623, when setting ratio as 0.7, the average improve the reliability of communication network by
reliability of whole network is over 0.98, which is a great providing some important nodes a standby link to another
improvement. OLT, which will increase the investment to some extent. In
a word, planning optimization should take a balance between
the reliability of communication network depends on the
topology connection and the cost of constructing new fibers
and channels. The sensitivity analysis of reliability and cost
show that choosing a propriate ratio of important nodes to be
upgraded is reasonable and economical.
ACKNOWLEDGMENT
This work was supported in part by the National Science
Foundation Industry/University Cooperative Research Center
on Grid-connected Advanced Power Electronic Systems
Fig.13. The change of reliability after optimization (GRAPES) under Award GR-18-06.