Electric utilities need to enhance the levels of automation in their distribution grids. The levels of complexity within the smart grid are demanding new ways of managing the grid. With the changes to the distribution grid driven by injection of distributed generation, energy storage, electric vehicle charging, and the ageing of core electrical infrastructure, the effects will be profound. These issues are compounded by an ageing workforce that is nearing retirement. As the workforce departs into retirement, the collective knowledge of how the distribution grid works could walk out the door with them.
The result is a dramatic assortment of impacts coming together all in a very short time span. Utilities need to respond to these challenges and the smart strategy is to introduce advanced intelligent automation into the distribution grid. Computing systems can control the grid and analytical and knowledge management systems can gather that collective intelligence.
Using automation to control and manage the distribution grid will improve service by minimizing service disruptions. Points of grid outage can be identified instantly. The affected customers can be isolated and the total number of customers impacted can be greatly reduced. Service can be restored to customers adjacent to the problem section faster than before when using the old methods or when manually restored. The directly affected customers can experience shorter outages too once the restorations are implemented by automation systems. Not all problems can be resolved with automation; of course truck rolls are still necessary in many cases. But, if problems can be identified quicker, if they can be more tightly isolated, and if some can be remotely resolved, then fewer customers need to be affected for shorter durations.
What is the best approach to distribution grid automation? Does one solution fit all Utilities? While there are several main strategies to automation of the distribution grid, different solutions fit different operators. So, selecting the best approach is critical to long term success and minimizing outages.
This article will consider the emerging world of advanced grid control and applications, in terms of the impact on the supporting network architecture that underpins this new automation which leads to the need to change from a primarily centralized control to a largely distributed control.
We will consider the knowledge of recent industry thinking about advanced distribution automation, destabilizing trends and the need for increased grid stabilization measures. It will also discuss the impacts of variable energy resources (VER) and distributed energy resources (DER) on grid power balance, and issues of scale for previously stand-alone distributed automation applications.
It will illustrate how present utility communications are both islanded (i.e. multiple non-converged networks) and are structured to support a centralized control model only. This is problematic going forward since the distribution grid will be destabilized as a result of the new technical challenges from ageing infrastructure, renewable energy integration, electric vehicles, as well as the remedial actions for energy storage, all the while being compounded by the mass departure of the workforce and their veteran expertise as the baby boomers’ retire.
IPv6 will be fundamental to the next generation smart grid networks. Superior security will be essential too. These next generation networks will be converged solutions that blend both IT and OT into a shared multi-services communications architecture.
Communication networks are of prime importance in all solutions and in all scenarios for this next generation distributed automation solution.
Smart Grids are Becoming Federated
When, in the 1990s, deregulation and massive expansion became the defining characteristics of the power industry, it quickly became apparent that automation was going to hold the key to increased productivity and increased cost efficiencies. But in the early 90s, there was virtual anarchy in the SCADA automation industry, with a myriad of different vendors throwing their weight behind an equally large number of different, and often proprietary, communication standards.
With hindsight, we can all see that this worked to no-one’s benefit except that of the vendors themselves, but at the time most industries simply paid their money and made their choice. The power industry, however, was perhaps a little more forward thinking. It recognised the cost and complexity problems associated with building automation systems around intelligent electronics devices (IEDs) from many different vendors, and set about drawing up standards which would allow it to reap the benefits of automation free from the constraints of individual vendors (Moss, 2006).
With so many communication protocols costing transmission grid companies time and money, the International Electrotechnical Commission (IEC) set a standard with an aim to streamline resources by speaking everyone’s language.
Since protection and control were automated in substations, the large number of communication protocols involved has undermined the joint operations of devices from different manufacturers. The introduction of IEC 61850, a single communication standard for the world, allows substation automation systems to be streamlined by permitting the interoperability of devices from different manufacturers without the use of protocol converters. It does so by considering data models of the applications, services for transferring these data, and real communication protocols (Hoga & Wong, 2005).
Six months after the IEC 61850 communication standard for substation automation was released, Siemens installed one of the world’s first substations with complete IEC 61850-based control. The equipment was installed in November 2004 in a 16-kV substation in Switzerland owned by Aare-Tessin AG, an independent energy group operating throughout Europe. Siemens installed its SICAM PAS (Power Automation System) control system in accordance with the new standard throughout the substation. The new communication standard ensures the required interoperability in the substation. All connected protection devices now “speak” a common language, regardless of manufacturer, which smooths the exchange of data with one another.
In addition to making the substation fit for the future by ensuring its communication capability in line with IEC 61850, Siemens also replaced the more than 40-year-old power switchgear at the substation with its new metal-clad, compartmentalized medium-voltage switchgear system (Siemens commissions substation control system based on IEC 61850, 2005).
Further requirements were the abilities to remotely control and monitor this new power equipment with a suitable communication network. Data switching and routing gear was used but lacked the necessary robustness to operate inside a substation. A lot of development has transpired since these early deployments in 2004. For example, today vendors such as Siemens (RuggedCom), Cisco and Alcatel-Lucent amongst many other companies have developed data switches and routers designed specifically for this task. These data communication devices are hardened with new standards and are ruggedized for the environment. In the context of the power utility substation, “ruggedized” means that the devices comply with specifications IEEE 1613 and IEC-61850-3, which certify that products meet stringent environmental, surge, and electromagnetic interference (EMI) requirements for utility substation environments. Ruggedized products tolerate a broad range of temperatures, surge, fast transients, radio frequency interference, and electrostatic discharges (Cisco, 2012a).
The introduction of IEC 61850 as the de facto substation automation standard for the communication between devices in a substation and the related system requirements has made it possible and justifiable to integrate station Intelligent Electronic Devices (IEDs) on a high-speed peer-to-peer communication network (Ethernet) through standardization. IEC 61850 presents a set of standard object-oriented data models to describe the processes to be implemented and controlled in a substation and a set of service models for the interactions between devices in a substation and the transfer of all sorts of IED data (Power delivery; research from Victoria University provides new data about power delivery, 2009).
The basis and the method of standardising substation communications in IEC 61850 are new. The nearest relatives of IEC 61850 are IEC 60870-5 and UCA 2.0 (Utility Communications Architecture version 2.0). IEC 61850 adopts, from UCA 2.0, principles such as object-oriented approach, TCP/IP and Ethernet. It built on IEC’s experience in digital communication in substations. Furthermore, IEC 61850 takes into account the requirements of substation automation worldwide (Schubert & Wong, 2004).
According to Francisc Zavoda, Hydro-Québec Research Institute (2011), as specified in their advanced distribution automation definition: distribution equipment is monitored, coordinated and operated in a real-time mode from remote locations. These activities are possible because of different communications links, which allow information to flow both ways between the remote system control centre and equipment controllers. Meters and major distribution equipment controllers belonging to different advanced distribution automation systems can be used as elements of an integrated power quality monitoring system. This symbiosis between advanced distribution automation applications and the power-quality monitoring activity is one of the advantages offered by smart grids. These intelligent electronic devices (IEDs) include many available features:
- Three-phase metering
- Measurement of voltage, current, demand, energy, power factor and frequency
- Harmonics, voltage and current plus total harmonic distortion
- Voltage flicker (so far only on meters)
- Symmetrical sequences and waveform capture
- Communications interface and protocols
- Multi-port (serial, infrared, Ethernet, modem)
- Multi-protocol access (DNP 3.0, MODBUS)
The current trend in the industry is that IEC 61850 communications, applicable to equipment in substations, will be extended in the future to include distribution equipment on the feeders and beyond. Many Utility distribution departments have planned the utility’s vision of a smart grid on its road map for 2016 and beyond. According to these technological roadmaps, major distribution equipment controllers will be replaced by standardized IEDs, which will comply with IEC 61850 and comprise plug-and-play devices.
The accuracy of the data acquisition process is an important factor, critically affecting the efficiency and reliability of advanced distribution automation systems and, furthermore, the efficiency and reliability of the distribution network. Remote control and surveillance of distribution equipment together with data acquisition are important aspects of the automation process. Combining the surveillance of distribution equipment with power-quality monitoring is an inspired and sound decision. There are several advantages of using power distribution equipment controllers and intelligent meters as elements of an integrated power-quality monitoring system:
- IEDs are already connected to the network, either on the medium-voltage side or low-voltage side,
- Devices are in constant evolution,
- Communications links for data transfer are available (IEDs remotely controlled and meters belonging to the advanced metering infrastructure).
To date, intelligent meters have been developed more rapidly than controllers, but evidence suggests the evolution in meters will be replicated and enhanced in the next generation of controllers and feeder devices such as reclosers, cap banks, segmentation switches, tie switches, power line monitors, transformer smart meters, and more.
Autonomy at the substation is the next step to evolve a truly smart grid. A shift will occur that will see the point of control be relocated from the network operations centre out to the major substations. As the substations take on more control over their own domain, there is still a need to be coordinated and orchestrated with a central entity that keeps the big picture in mind and makes sure that these stand alone control centres are not operated in conflict to each other. However, they will push the intelligence and control towards the edge of the distribution grid and away from the centre at the NOC. Automation is driven by software and logic controllers that maintain a constant vigilance over their serving areas. These controllers stand ever-ready to respond or act in the best way to protect the grid and maintain a stable service to the end customers.
Latency is the enemy of these real-time distributed controllers. They must operate within extremely tight time frames in order to execute changes and reconfigurations swiftly. Some estimates are provided for consideration regarding latency needs for smart grid devices within the substation and connected to the feeders in the distribution grid.
So, what is a federated network?
A federated network is one composed of multiple tiers. Each tier has specific functions to perform in the end to end network fabric. However, upon demand, these network tiers can come together or break apart upon demand.
An analogous example would be the United States government model. It is similarly composed of tiers in the form of the federal, state, and municipal layers. Each level of government has its own functionality. However, upon demand, these levels of government can join together to solve problems, such as emergency response to storms.
So, a federated network is one that is able to make and break upon demand and can act in harmony or in isolation of the other network tiers as required. Authority and functionality can be grouped, or segmented. It is also managed as a centralized network and as a distributed network simultaneously, this is what makes it federated. The intelligences is at the edge, but it is centrally orchestrated.
In the smart grid, we desire a federated network as a means to manage the many domains that will be established. As well, as the grid intelligence is pushed out of the data centre towards the network’s edge, we will be creating a multitude of clusters that form to deliver power to serving areas formed around the grid feeder and substation architecture. The smart grid is expected to be large and complex, so the classic flat model for networks would quickly become unmanageable. We must segment the networks into clusters that map to the power grid.
Federated networks are dynamic. They evolve over time driven by the functionality of the applications riding over them. While they are geographically diverse, they share attributes that still demand a centralized control system in order to orchestrate between these diverse segments.
Management and operation of federated environments over administrative domains and multiple networks require specific mechanisms. A central federation control unit communicates a virtual private network (VPN) set-up request to gateways to provide requested links, which assures connectivity to all domain resources. Thus, the technology enables setting up secure overlay networks over unsecured network links. Building a federation requires certain entities and control mechanisms that separate the data plane from the control plane, so a software defined network solution will play a key role in these networks. The architecture relies on a centralized approach where functionalities are delivered by centrally administered tools and entities.
Federation in networking systems means users can send messages from one system to another. Federated machine to machine (M2M) networks permit communication across different M2M clients and platforms. They maintain an open directory allowing the networks to message their users. A few federated networks also work on an interoperability basis, where software from two or more vendors shares data between different platforms.
A central business entity provides the federation control unit and service composition tools where services and components are orchestrated on demand. Interconnected domains are configured and managed in a top-down approach, while domains publish services and capabilities in a bottom-up approach.
Other critical and vital aspects of the network fabric are becoming federated too. For example, Firewalls, Intrusion Detection Systems (IDS), Intrusion Protection Systems (IPS), Authentication, Authorization and Accounting (AAA) security, diagnostic systems, and more.
A federated network leverage the multimodal and intermodal concepts described in a previous LinkedIn Pulse article. Smart Grid: Multimodal Networks
The network exists to service the applications that it underpins to connect. A perfect network is transparent to the applications and permits them to perform and function perfectly. without influence or affect of the network upon the applications.
While networks are important, it is the applications that are truly important. Utilities desire smart grids to improve AC power delivery to their customers. The government, via NERC CIP regulations expect that these networks provide a trusted and stable platform for the safe and stable delivery of the nation’s electrical grids.
Michael Martin has more than 35 years of experience in broadband networks, optical fibre, wireless and digital communications technologies. He is a Senior Executive Consultant with IBM’s Global Center of Excellence for Energy and Utilities. He was previously a founding partner and President of MICAN Communications and earlier was President of Comlink Systems Limited and Ensat Broadcast Services, Inc., both divisions of Cygnal Technologies Corporation. He holds three Masters level degrees, in business (MBA), communication (MA), and education (MEd). As well, he has diplomas and certifications in business, computer programming, internetworking, project management, media, photography, and communication technology.