Distribution Automation / Substation Automation, Smart Grid

Performing Brain Surgery on Seattle City Light’s Distribution System

Issue 6 and Volume 18.

Performing Brain Surgery

BY MICHAEL PESIN, SEATTLE CITY LIGHT

The past 30 to 40 years have seen Seattle develop into a major technology hub with a host of companies such as Microsoft Corp. and Amazon.com Inc. setting up shop in the Pacific Northwest. It was about that long ago when a fleet of Seattle City Light’s (SCL’s) substation remote telemetry units (RTUs) went into service for an area that’s grown to some 1 million residents spread across 130 square miles.

To better serve this dynamic, growing community, the city’s utility embarked on a project to modernize its systems including integration of information technology (IT) with operation technology (OT) that would bring the system in line with current and future demands. A “Smart Grid Road Map” was developed in 2009 and become part of SCL’s strategic plan. It detailed the necessary technology investments that included substation automation and distribution automation as some of the core systems. Adding more intelligence to the distribution system would be the foundation of SCL’s grid modernization program. The challenge was that the existing RTUs at all SCL substations were of a vintage that could not support Seattle’s smart grid requirements. The solution to enabling required functionality was to build new substation control systems that would rely on intelligent electronic devices (IEDs); however, Seattle was facing some challenges with this approach.

Seattle’s distribution substations were built significantly larger than those at typical North American utilities. A typical substation has anywhere between 20 and 30 feeders. With a significant majority of substation control and protection devices’ being electromechanical, it could take several years to migrate fully to a new IED-based control system at just one location.

SCL neither could afford to replace all its RTUs immediately nor to delay its smart grid plan.

The option to upgrade existing RTUs-basically performing brain surgery-was the best short-term option for modernizing the local grid.

Substation-Server Architecture

A substation-server architecture is just like a distributed client-server application and presents a better supervisory control and data acquisition (SCADA) model to configure a substation with less complexity and more accessibility for engineers and technicians.

It consists of local IEDs connected to a substation server that is interfaced with an energy management system (EMS) and other enterprise IT systems.

The new system provides improved visibility, command and control and data retrieval for a smarter grid that delivers improved operational results.

SCL’s legacy SCADA architecture employed Moore Systems MPS-9000S RTUs that were installed in the early ’80s.

RTUs were hardwired to electromechanical devices and connected with the EMS over serial modems.

Existing hardware offered no interfaces to IEDs or open communication protocols beyond the existing L&G 8979 currently used (see Figure 1).

Figure 1

There are no RTU spare parts, repair or technical services available, and the age of these RTUs affected reliability.

Internal analysis indicated a complete “forklift” RTU replacement was out of the question because of the significant financial and operational challenges, particularly labor costs and downtime.

SCL began studying possible work-arounds that could provide the least-intrusive methods for eliminating the weak link from its system and getting maximum functionality from existing equipment.

In researching alternatives to full replacement, SCL learned about an RTU upgrade program offered by Efacec ACS.

The Norcross, Ga.,–based company had been revamping substation RTUs from vendors since 2001 and had upgraded more than 100 similar RTUs at several utilities.

The attraction to the upgrade approach was that SCL wouldn’t need to disturb existing I/O terminations or replace enclosures.

Best, a single RTU upgrade usually could be done in less than a day-a 90 percent savings over replacement in labor alone.

Using this approach, SCL significantly simplifies transition to the desired modern substation control architecture.

With an upgraded RTU capable of multiprotocol communications, utilities gradually can move SCADA functionality from hardwired I/Os to new IEDs as they get deployed.

At some point, all SCADA functions will be transferred to IEDs that communicate with a substation server, and all RTUs can be decommissioned.

The number of IEDs that still need to be deployed means it might be years before it can happen, though.

Nearly off the Shelf

The object of an RTU upgrade is to leave the local I/O terminations and original field I/O wiring in place.

To do this, the original MPS-9000S electronic card files are swapped for Efacec’s Connex logic modules with a nearly identically sized 19-inch rack-mounted chassis and expansion card files for analog and digital input I/O.

The original MPS-9000S control relay modules are interfaced directly through special adapters and cables to provide standard Efacec ACS isolated relay driver logic to the original relays.

A new power supply is included and replaces the original unreliable MPS power supplies.

The high-speed serial and Ethernet gateways with open protocols for multiple master and IEDs also are added.

Every RTU upgrade unit is customized for each substation to accommodate the necessary I/O interfaces and termination points.

An included software configuration tool duplicates the legacy RTU database in its original protocol envelope, leaving the master database and displays largely untouched.

After pre-installation training, SLC performed the upgrade in a day, thanks to having its labor substantially reduced or eliminated, including wiring, testing and drawing revisions, as well as master station database and display revisions. Figure 2 shows a before and after upgrade image of the hardware configuration.

Figure 2

Installation was fairly painless. Each substation stays online during the cutover, although there is no telemetry from the node until the full retrofit is complete.

Among a few surprises, SCL’s current EMS, only one of its kind in service in the continental U.S., used a slightly different variation of the standard L&G 8979 communications protocol.

Efacec ACS accommodated modifications of the protocol to handle the variation.

The RTU upgrade included the necessary functionality to translate the data so it could communicate with the existing EMS, but it also offers serial and IP DNP3.0 protocol for an easy interface to a new EMS planned for the next year.

A beachhead for grid modernization

As stated, the RTU upgrade is a medium-term solution, but the results from the first two substations have been very good. By moving forward with an RTU retrofit, SCL is on a path to replace unsupported electronics with modern, reliable and fully supported hardware and software.

Adding modern, open protocols, high-capacity, high-speed master and slave gateways (serial, TCP/IP or both), Windows PC-based configuration and diagnostic tools and the latest communications digital interface mediums, SLC can begin a transition to grid modernization in earnest that’s also cost- effective (see Figure 3).

Figure 3

Upgrade projects such as this one have shown significant savings at other utilities, averaging 59 percent less in total project costs compared with traditional RTU replacements; 97 percent less engineering time; and a total field crew labor average reduction of 92 percent.

With only two completed projects, SLC doesn’t have enough data available yet to compare statistics, but the time and cost savings have been pronounced.

By upgrading RTUs, however, SCL can eliminate the weak link in its SCADA system and continue its substation automation project at its own pace without the threat of sudden system failure.

Michael Pesin is the chief technology advisor and smart grid architect for Seattle City Light and is involved in many industry organizations. He has more than 25 years of experience in the electric utility industry and is a nationally recognized expert in smart grid and utility automation technologies. Reach him at [email protected].


Move or Improve? A Guide to Retrofitting RTUs

Under the best circumstances, a substation remote telemetry unit (RTU) has a useful life of 20 years, although technical and spare parts support might be only half that span. Many utilities with legacy systems find that their original vendors no longer support the RTU or might be out of business. Aging RTUs have limited functionality and are unable to integrate with intelligent electronic devices (IEDs) such as digital relays or handle high-speed data communications or 16/32-bit IED data values necessary for distribution automation.

Completely replacing an old RTU can be time-consuming and expensive, leaving the alternative of retrofitting modern logic hardware and software to an existing system. It’s important to consider a few questions before an upgrade:

1. Are the existing wiring termination assemblies of the RTU active or inactive? If they are inactive, the RTU is usually a good upgrade candidate.
2. If the termination assemblies are active, are they reliable or repairable?
3. Are the existing control interposing relays reliable or repairable?
4. Is the existing cabinet or rack reusable?
5. Can the original or outside vendor provide an upgrade kit?

The decision to maintain the original I/O termination boards on a legacy RTU depends on whether the present wiring terminations are reliable. This makes up the bulk of the cost savings with an upgrade and should be considered when answering the first, second and fifth questions. If the answer was “no” to all of the above, the only choice is a full replacement. Conversely, a negative answer to one or two of these questions will affect the total financial gains realized from a retrofit upgrade but does not necessarily negate all the advantages like replacing unreliable I/O modules with more capable, form-fit modules.

A complete replacement requires more than just budgeting for a new RTU. There is additional budget needed for project management, substation wiring diagram revisions, field installation and wiring, labor for retesting to the end devices, possible changes to the master database and substation downtime.

With a relatively simple upgrade, many of these costs can be reduced or eliminated.

More PowerGrid International Issue Articles
PowerGrid International Articles Archives
View Power Generation Articles on PennEnergy.com