Communication system for Smart Grids EE-576

M.ENGG. ELECTRICAL ENGINEERING PROGRAMME, FALL SEMESTER 2022

INSTRUCTOR: DR. SUNDUS ALI, ASSISTANT PROFESSOR, TELECOMMUNICATION ENGINEERING, NED UET

Week 6 TOPIC:
Networking Fundamentals for Smart Grid Communication-
a. Layers 4
b. SCADA Systems

Email: sundus@neduet.edu.pk, sundus@cloud.neduet.edu.pk

Office Location: Telecommunication Engineering Building Faculty Offices
 2
IP: Connectionless Service

 IP does not have any procedures for guaranteeing packet delivery, packet error detection or
correction (except for error detection of the packet header only, so that the packet is discarded
if the received header is in error), or acknowledgment procedures for reliable packet delivery.
 Lower layers(such as Ethernet or Frame Relay) may provide these functions.
 The upper layers(L4) may implement features for reliable data delivery between the
corresponding endpoint entities as needed.
 IP has no procedures for establishing a connection between a pair of endpoints prior to the
transfer of packets between them.
 Every packet is independently and individually delivered to the destination.
 A packet may get lost in the network (e.g., if the frame at L2 layer carrying the packet is
discarded because of an error). In some cases, there is also the possibility of a packet being
delivered again (a duplicate packet) at the destination. Thus, the service provided by the IP layer
is a connectionless service.
 3
IP: Quality of Service

 Quality of service (QoS) in IP networks refers to the IP network providing preferential treatment to data from
certain priority applications over data from other applications. The most common example in the NSP IP
networks and many enterprise IP networks is preferential treatment given to VoIP traffic over traffic of all other
applications carried by the network.
 Such preferential treatment is necessary, since voice quality is degraded if the VoIP packets are delayed in
the network, experience large jitter (i.e., large variation in the delivery time of the successive packets of the
VoIP packet stream), and/or suffer significant packet loss.
 Often the next on the priority list after VoIP are the data traffic for video, critical business applications traffic,
and the noncritical data traffic (called “best effort”), in that order.
 The need for QoS arises since, at times, network links and/or the routers may be congested with heavy traffic
volume, with limited capacity of links and/or limited router buffer storage capacity to store packets. This leads
to delays since the packets may have to remain in the router buffers awaiting their turn for transmission.
Buffer congestion may lead to packet loss, since incoming packets may be discarded if buffer space is
unavailable; hence the need for QoS.
 4
IP: Quality of Service

 IP includes a TOS byte in its header. The source endpoint (or the application therein) may insert
an appropriate code in the TOS byte to identify the application that generated that packet.
 Using the content of the TOS byte, the router may then provide preferential treatment for that
packet, forwarding it ahead of packets with TOS byte values denoting lower priority
applications. This procedure ensures that high-priority packets do not get dropped from the
buffer before packets of lower priorities.
 There are several standards which address QoS. Networking product vendors provide different
QoS implementations including vendor-proprietary implementations. Because of the extreme
delay and priority requirements for many applications such as teleprotection, synchrophasors,
and SCADA, QoS support is very important in Smart Grid networks.
5
 6
TCP and UDP

 There are two transport layer protocol standards defined over IP – Transmission Control Protocol (TCP) and
the User Datagram Protocol (UDP)
 However, some applications may not use either of these or any other L4 protocol
 As indicated earlier, IP delivers individual packets from the source to the destination based on the
destination IP address. The connection-oriented TCP provides reliable delivery of L4 layer messages from the
source L4 entity, called the source port, to the destination port.
 TCP provides acknowledgment of each packet sent by the source port. Lost packets are detected and
retransmitted. In addition to supporting acknowledgment and retransmission procedures, sequence
numbers in the L4 header are used to reassemble the L4 message from the received packets.
 TCP also provides for a 2-byte checksum that detects errors in the entire IP packet. Thus, the error detection
is supported end-to-end at the L4 layer. TCP is considered a heavyweight protocol, since it requires router
and link resources for support of its reliability features.
 TCP’s header is at least 20 bytes long and includes the designation and source port numbers, sequence
number and acknowledgment, checksum, and other parameters.
 7
TCP and UDP

 UDP is a connectionless and lightweight protocol. It adds little additional overhead to packet
(called datagram in this context) delivery over that of IP. The delivery of a datagram is not
guaranteed – it may be lost in the network.
 There are applications that may use UDP because of its very low processing overhead and/or
because delivery guarantee is either not required or not appropriate due to time sensitivity of
data.
 For example, there is no value in retransmitting a VoIP packet after its loss, as the retransmitted
packet would reach the destination too late to be useful. The UDP header is 8 bytes long and
includes the destination port number and the length of the datagram.
 Note that UDP is defined between endpoints. The intermediate routers do not participate in
UDP protocol.
8
9
 10

SCADA
 11
SCADA

 Supervisory Control and Data Acquisition systems have been in use for the last few decades to
proactively monitor and control the power from the utility DCC using Remote Terminal Units
(RTUs) deployed at the transmission and distribution substations.
 A substation may be a transmission substation or a distribution substation.
 Monitoring (or data acquisition) refers to the measurement and:
 Reporting of voltages, currents, power (W), and reactive power (VAR) and reporting on the
status of different systems in the substation such as the circuit breakers and switches.
 Control refers to the control of substation operations such as tripping of circuit breakers and
adjusting taps on voltage regulators.
12
 13
SCADA Evolution with IEC 61850 Set of Standards

 IEC 61850 is a 10-part comprehensive set of standards [61850-01-10] encompassing

communication networks and systems in substations.
 Many utilities across the world have begun or are planning to deploy substation devices
Intelligent Electronic Devices (IEDs) and substation communication networks based on these
standards.
 The standard formally defines an IED as any device incorporating one or more processors
capable of receiving or sending data/control functions from or to an external source.
 Examples of IEDs include electronic multifunction meters, digital relays, and controllers
 Further, a configuration language is defined for SCADA operations and maintenance
functions.
 This allows for the use of standards-based tools from multiple application providers to support
SCADA operations and maintenance functions. With IEC 61850, all substation devices are IED
based.
 14
SCADA Evolution with IEC 61850 Set of Standards

 Each IED supports one or more functions including switchgear, CT, VT, bay controller, and
relay.
 Connectivity to conventional switchgear and instrumentation software may be maintained for
a period of time by connecting them to IEDs that also simultaneously support such
conventional equipment.
 One or more process busses are deployed to support local interconnections of IEDs in several
parts of the substation yard.
15
 16
SCADA Evolution with IEC 61850 Set of Standards

 A process bus is an Ethernet LAN, generally with a single Ethernet switch. Each IED communicates
with other IEDs connected to its process bus. For example, a local multifunction meter can send
measurements to relays connected to the process bus.
 The number of process busses depends on the number and location of the switchgear and
instrumentation transformers in the substation. Note that primary substation equipment is deployed
at many different points of the power circuits in the substation, necessitating the use of multiple
process busses.
 The substation’s station bus provides connections between IEDs and other systems in the substations.
Thus, process busses connect into the station bus. The station bus is also an Ethernet LAN with one or
more Ethernet switches.
 The local HMI and engineering station communicate with the IEDs over their connections to the
station bus.
 The station bus provides LAN connections for other substation devices including IEDs that may
provide functions other than the ones identified here. The station bus connects to the SCADA
master controller in the DCC through the IP router connected to the station LAN.
 17
SCADA Evolution with IEC 61850 Set of
Standards
 As a target standard architecture, RTU functions for aggregating data or control functions from the IED
are not defined. Each IED communicates directly with the SCADA master over protocols like DNP3
(carried over the IP connection between the substation router and the SCADA master).
 For IEDs (and legacy RTUs, if present) with only serial network interfaces, the serial connection should be
tunneled using IP. An important aspect of IEC 61850 is the definition of the Generic Substation Events
(GSEs).
 GSEs provide a fast (within 4 ms) and reliable mechanism for generating event notifications within a
substation.
 Generic Object-Oriented Substation Events (GOOSE) defines a standardized mechanism for reporting
event status and associated information.
 One class of events supported by GOOSE is protection events. GOOSE messages generated as a result of
protection events are multicast over the Ethernet.
 VLANs are used within the substation Ethernet network. The use of VLANs provides traffic separation (for
security) and priority handling of messages.
 18
Distributed Network Protocol
DNP3 over IP
 DNP3 is becoming the most prevalent SCADA protocol for communication between an IED or RTU in
a substation and the SCADA master controller in the DCC.
 DNP3 was adopted as an IEEE 1815 standard.
 DNP3 standard is replacing many traditional proprietary and other SCADA protocols.
 DNP3 was developed by the DNP3 user group, an organization with members representing utilities
and SCADA product vendors.
 DNP3 is still evolving. Despite the word “distributed” in DNP3, SCADA systems for both the transmission
and distribution systems use DNP3. DNP3 is also used in SCADA systems deployed in industries other
than the electric power industry, such as in water or gas supply.
 DNP3 provides for periodic polling of the IED/RTU by the SCADA master control for measurement data
and status information on substation devices including switchgear and relays. The measured
quantities can include both analog and digital data such as voltage, current, power, reactive power
(VAR), and frequency.
 Status information is usually binary – such as active/inactive or circuit closed/circuit tripped. In
addition to the measurements and status, the IED/RTU may also provide measurements on
accumulated data such as energy (kWh).
 19
DNP3 over IP

 DNP3 protocol has its own application, transport, and data link layers.
 Note that the DNP3 data link layer is end-to-end. DNP3’s application layer breaks DNP3 messages
into fragments. Its transport layer breaks fragments into packets and adds 1 byte of transport
header. The packet size including the headers is 250 bytes.
 The data link layer adds its header (10 bytes) to the packet to form a frame. Additionally, at the data
link layer, 2 bytes of CRC are added for every 16 bytes of payload. Thus, for the maximum size
packet of 250 bytes, the frame size is 292 bytes including all the overheads.
 DNP3 is defined over serial physical layer connections. Popular standards like RS232 and RS485
provide streaming of bytes from the source to destination. Many different connection technologies
are used to support serial connections, including voice-grade data lines; leased data lines; MPLS
pseudowires; serial link emulation through Frame Relay, Ethernet, or IP services; and narrowband
wireless data connections such as over 900 or 220 MHz spectrum.
 With the advent of IP for SCADA communication, DNP3 has added a “Data Connection
management” layer. This layer allows for running DNP3 over a TCP/IP or UDP/IP connection
20
 21
DNP3 over IP

 DNP3’s full set of procedures and all its layers together form an application layer riding over
TCP or UDP. The IP layer at the substation will be defined over the Ethernet layer
(implemented by the station bus) between the IED/RTU and the IP router at the substation.
Generally, the systems at the DCC including the one incorporating the SCADA master will
be on the DCC Ethernet LAN, so that the IP layer at the DCC will be on top of the
corresponding Ethernet layer.
 For consistency with IEC 61850 standards, DNP3 will need to support the object models
defined in the IEC 61850 standards. In the interim, it will be necessary to map these IEC
61850 object models to DNP3 as utilities migrate to the 61850-based substation devices and
functions.