IEC 61850 and the revolution in substation communications

In this CPD-accredited article, produced in association with Schneider Electric, Network looks at the latest developments in peer-to-peer communications for power networks.

IEC 61850 and the revolution in substation communications
This CPD Article is provided in association with Schneider Electric. In order to complete the CPD, read the article and then answer the questions in the CPD questionnaire. Closing Date: 8th September 2017. CPD credits: 1 hour

The days of substations filled with copper cable are coming to an end. Smart grids need smart communications, and peer-to-peer communication within and between substations is set to quickly become business as usual for UK power networks. Facilitating this transition is IEC 61850, a standard launched in 2004 and updated in 2011, which provides a new approach to automation and communication in substations.

IEC 61850 is a single protocol for substation automation that allows devices from different vendors to talk to one another directly over an Ethernet connection that operates in a similar way to a local area network (LAN). It is divided into several parts dealing with issues such as data, communications and compliance. Taken together, the standard enables distributed automation and significant information exchange – vital for the rollout of smart grids.

The architecture of IEC 61850 is based on the principle of “abstract” data, whereby data groupings and naming conventions are abstracted from underlying protocols, meaning the language is the same regardless of which manufacturer has made the device. The creation of such a vendor-neutral standard is a major step forward for interoperability, as well as offering speed, security and cost benefits.

“With this standard and similar peer-to-peer communications standards, you can set up a LAN within either the control room or the whole substation,” says Randolph Brazier, head of innovation and development at the Energy Networks Associations. “You essentially have a network like you would have in an office, and that allows devices within that substation to talk directly to each other or to multiple devices.”

The benefits of IEC 61850 are multiple:

  • Because communications happen over an Ethernet system, they use fibre optic cables, which are not susceptible to magnetic interference. Therefore, interference in the substation, or faults, don’t affect the communications.
  • Traditionally when two devices needed to communicate within the substation, they would use copper cables running between them. With the new standard, there may be just one or two point-to-point copper cables, connected in to one or two Ethernet rings. The saving on copper cables represents a significant cost saving – which can run into hundreds of thousands of pounds for a single substation, and cut labour costs for installation.
  • The need for copper cable is further reduced by the ability to convert signals, such as those for a measurement of voltage or current, from analogue to digital.
  • The standard enables devices to talk to each other that previously were unable to, because it is consistent across all devices in the station. Whereas previously some manufacturers’ devices could only talk to other devices from the same manufacturer. With this common protocol, any devices can talk to each other.
  • The standard enables communications within the substation to be future-proofed. When one device fails or comes to the end of its natural life, rather than reinventing the whole wheel with the communications system in that substation, that device can simply be pulled out and replaced with a new one in that speaks the same language.
  • The protocol has been extended to include windfarms and other forms of distributed generation.
  • The standard offers improved security of communications, because signals are prioritised and are repeated. Previously, when one device was meant to, say, send a message to another device to isolate a line, that message could sometimes get lost, through interference or just a fault. With this protocol, the device repeats the signal so it can’t be lost.
  • Communication is physically faster than using traditional methods, which again helps network security.

IEC 61850 is currently being rolled out by networks across the UK, and its use for wide-area networks is being trialled. “Using peer-to-peer communications within the actual control room itself is quite common now, whereas using it outside in the substation yard is not so common,” says Brazier. “We’re on the cusp of doing that – the networks are all trialling it. They realise it’s definitely the way of the future.” 

IEC 61850 Solutions

As a minimum a solution is likely to integrate several intelligent electronic devices (IEDs), such as protection devices, measurement centres, bay controllers, etc. communicating with the rest of the system via one or more protocols.

Today modern systems can be constructed through an Ethernet network linking components through an appropriate architecture. The Ethernet network might be within a substation, typically for a transmission or distribution application, or it may interconnect dispersed substations such as are commonly found in industrial or infrastructure applications.

Deploying a client-server communication exchange can avoid any central point of failure and facilitates a tailored redundancy. For large sites and when many voltage levels are to be considered, multiple-ring Ethernet architectures may be considered to allow an appropriate network segmentation.

Physical communications between components are based on two main technology families (Ethernet and serial links) in order to cope with various application performance needs and re-use of existing devices and integration of third-party equipment. Solutions that support typical serial protocols (T103, DNP3 and MODBUS, for example) in order to interface existing devices and can be fully integrated into an IEC 61850 system through a gateway device.

State-of-the-art communication technology is based on client-server and peer-to-peer links. This relies on fast Ethernet networks and offers new perspectives in terms of distributed functions, performances and flexibility. It enables innovative automation schemes and flexible addition of new applications. Ideally these are based on standard Ethernet communication infrastructures. Switched Ethernet is the preferred choice because it manages collisions over the network. The possible latency time of the network itself becomes negligible with regards to the equipment response time, so the system determinism is primarily dependent on the choice of the devices and its architecture.

Correct selection of Ethernet switches, robust devices designed for the substation environment will enable various topologies (ring or star or mixed) with the possibility of a redundant communication path for increased availability.

Ethernet IEC 61850

IEC 61850 is the standard for communication in power systems based on Ethernet and provides fast peer-to-peer and self-descriptive communication between IEDs, up to 100Mbps. It includes general aspects (project management, substation functions, and more), detailed data model, configuration language and conformance tests.

The IEC 61850 communication can be fully redundant. Time synchronisation is part of this standard and 1ms resolution can be achieved for sequence-of-events analysis (for example).

Software modules are available which offer IEC61850 client and server services, meaning that an application can act as a client and/or server.

IEC61850 client and server applications may be separate software modules or merged in one.

Peer-to-peer communications

With new technologies, protection schemes are becoming more complex. Inputs and outputs are traditionally hardwired between different IEDs.

During the engineering process, a small change in the protection scheme logic could require a substantial amount of effort to implement, especially during later stages of the engineering process. This change could require updating engineering drawings, making modifications to, or adding new wiring between devices. These changes can be time consuming and labour intensive, requiring different trades to accomplish what could be considered a trivial change to some digital logic.

GOOSE

IEC 61850 generic object-oriented substation event (GOOSE) messages are high-speed messages optimised to be multicast over an Ethernet network. These messages allow for digital representations of hardwire I/O to be published on the network and received by the subscribers within a given timeframe. The major benefit of using GOOSE messaging for I/O is that adding new logic variables and virtual inputs and outputs can be simplified, requiring the engineer to modify only the device configurations. The potential now exists for much more complex, distributed protection schemes.

Removal of physical I/O limitations of protection relays allows for many virtual input and output signals – internal to a relay and not typically used in today’s schemes – to be shared between multiple devices. In theory it could be several hundreds, in reality there are limitations to be considered.

Ethernet architectures

An Ethernet network can be over optical fibre and/or copper twisted pair and fully redundant Ethernet solutions can be engineered.

Ethernet network redundancy can be realised through ring architecture, redundant star architecture or mixed architectures.

PRP (parallel redundant protocol)

PRP principle is based on the duplication of the Ethernet network infrastructure. Any device is doubly attached to the two LANs. The traffic is duplicated to the two networks. Frames are sent at the same time on the two networks and the first arrival is used the second is discarded. This mechanism protects applications from data loss – in case of any failure, there is always one available link to transmit data. The network unavailability time is by construction 0ms.

PRP relies on the parallel operation of two local area networks (LANs). It allows a mixture of both redundant and non-redundant equipment on the same network.

The operating principles of PRP are:

  • Two completely separated Ethernet networks (LANs) are operated in parallel.
  • Each doubly attached node with PRP (DANP) has an interface to each LAN.
  • A DANP source sends a frame simultaneously on both LANs.
  • A DANP destination receives both frames (in normal operation) and discards the duplicate.
  • A singly attached destination receives only one frame.

If a LAN fails, a DANP destination operates with the frames from the other LAN.

When applied, those principles amount to this: if an IED fails, only its protection function is lost. The other IEDs continue to communicate over the redundant LAN to protect and control equipment. And if a switch fails, communication redundancy is lost, but not the protection and control functions. That’s because messages sent on the healthy network are not disturbed.

Further, PRP allows single and double network attachment devices (non-redundant and redundant devices) to be mixed on the same LAN, thus allowing laptops and workstations to be connected to the network with standard Ethernet adapters (non-redundant devices).

And because PRP uses a double star topology, the communication bandwidth on each star made with a single link is not different than for a non-redundant network. PRP can thereby accommodate more IEDs than HSR (high availability seamless redundancy protocol).

The PRP protocol is also plug and play. In other words, no special network engineering is required. However, additional switches are required to build the double star network infrastructure, making PRP deployment costs higher than HSR.

HSR (high-availability seamless redundancy protocol)

Like PRP, HSR applies the principle of parallel operation to a ring interconnecting IEDs with two links. HSR relies on ring topology, sending data not over two networks but in both directions of the single ring.

Unlike PRP, however, HSR does not allow single and double network attachment devices (non-redundant and redundant devices) to be mixed on the same LAN. Because of that, laptops and workstations must be connected to the network via a dedicated redundancy device called a redundant box (RedBox).

The principles of HSR are:

  • One Ethernet network ring connects each doubly attached node with HSR (DANH).
  • A DANH source sends a frame simultaneously on both ports and blocks the sent messages if received.
  • In normal operation, all DANHs receive frames from both ring connections and instantaneously forward them.
  • A DANH destination receives both frames in normal operation. It uses the first and discards the duplicate.
  • If a link fails, a DANH destination operates with the frames from the other healthy path.
  • A singly attached destination receives only one frame via the redundant box (RedBox) it is connected to.

In other words, if an IED fails, its protection function and the communication redundancy are lost, but the other IEDs continue to communicate, protect, and control the electrical equipment through a single communication network.

Because HSR forwards all messages from all IEDs, the communication bandwidth is proportional to the number of IEDs. Each device needs to send not only its own messages but also pass along all messages coming from all other IEDs. Therefore, HSR is limited to about 16 communicating devices.

HSR, like PRP, is also plug and play. And because it employs single ring topology, additional switches are not needed. That makes deployment less expensive than PRP.

PRP and HSR solutions are based on IEC 62439-3.

 


Comments

Login on register to comment

Login Register


    Related content


    Related supplier content