Grid Operations

Email This Author Comment On Article About The Author More Articles By This Author

Substation automation has slowly evolved from the classical remote terminal unit, to the current network based terminal unit. However the inception of the Smart Grid is a catalyst towards a new evolutionary step in distributed automation.

The first generation RTU was limited in functionality, size and deployment. These units were installed in almost all transmission substations but were found in much fewer distribution substations. The current generation RTU is a network terminal unit, NTU, which has evolved to take advantage of many new technologies in handling the burst of IED data as well as meeting the information demands of the Enterprise. Whereas the RTU was a slow speed serial device, the NTU is a high speed network device. Whereas the RTU used a slow polling scheme the NTU interrupts the host with unsolicited reporting. Whereas the RTU used proprietary protocols, the NTU employs industry standard protocols. Whereas the RTU delivered discreet I/O points numbered in the hundreds for even the large units, the NTU is self configuring from an impressive library of multiple IED products, retrieving hundreds of I/O points from a single IED; the total point count of a modern NTU can reach 100,000 points.

The endeavor of the Smart Grid is to support promising technologies that include smart metering systems, demand response systems, distributed generation management systems, electrical storage management systems and distribution automation systems. Implementation is precipitating a third evolutionary change in order to support applicable functionality in the current network based substation automation platform: It’s adding Intelligence to the network terminal unit.

Utilities have a diversity of plans, phases and business cases prompting their version of the Smart Grid designed to achieve different but complimentary objective functions. Some are focused on ‘self healing’ networks to increase reliability while others are minimizing losses and voltage drop to reduce costs, while others are focused on data collection or combinations of each.

In the past many feeder automation functions were deployed as ‘pilot’ projects. However, utilities discovered that as deployment of these technologies expanded beyond the ‘pilot’ phase, the cost of maintenance and supporting infrastructure overwhelmed the practicality of expansion. Furthermore, each system bore the cost of its dedicated infrastructure for processing and communications.

The Intelligent Terminal Unit

The optimum solution is an intelligent substation automation platform that supports local and remote access with a common graphical user interface. The Intelligent Terminal Unit (ITU) supports command and control capabilities at the substation processing the data locally for local automation. This improves performance, distributes computing, reduces communications and maximizes reliability. The ITU directly addresses Smart Grid Objectives with advanced network optimization applications and feeder network automation. Fault Detection Isolation and Restoration (FDIR) is a Smart Grid ‘Self Healing’ feature. FDIR has the ability to automatically detect feeder faults then perform automatic isolation followed by restoration of service to un-faulted feeder sections of both upstream and downstream. The FDIR based self healing network reconfigures and restores service within seconds of un-faulted sections that are supporting customer loads. This and other network automation applications which were the domain of only the largest distribution control centers may now be implemented at any time by the intelligent substation platform. Other applications include:

• Fault detection isolation and restoration to restore service

• Integrated Volt / VAr control to reduce losses, improve power factor and minimize voltage drops.

• Feeder load transfer and balancing to transfer loads between feeders.

• Transformer and feeder phase balancing: to balance each phase of the main transformer and feeder.

• Real-time IEEE 1366 Performance Indices calculation and event reporting and archiving.

Legacy independent substation and feeder based systems perform self healing or power quality functions. But each brings with it a disjointed point of maintenance. Self healing solutions that employ a pre-planned solution or switching scenarios or at best use ‘rule based’ switching solution, cannot handle large complex network scenarios which may initiate from any, or an abnormal, system topology. They are reliant on an assumed network topology, which may not be a valid assumption during normal periods, let alone during a time of crisis. The old paradigm of adding and supporting a dedicated system for each function cannot practically support the Smart Grid’s integrated objectives.

The ITU supports any Smart Grid application using a common substation database and model. The advantage of a common model driven approach is that the complexity or state of the network is not an issue. The applications can adapt; they can expand to any degree of complexity or size. Since the substation database is very large, due to IEDs, the Smart Grid network applications must be able to access any data value using a common retrieval mechanism to avoid independently collecting or manipulating the data. The data points and network model are central to the substation not to the application.

Smart Grid substation and feeder automation schemes must be incrementally deployable while meeting the following requirements, in order to maximize their adaptability, and their solution robustness:

• The application’s proposed solution must be adaptive to the current real-time network topology regardless of whether the network topology is in a normal or abnormal state; it must not require a predefined system topology. Otherwise maintenance is high and it will only work once during an event.

• The self healing capability must dynamically derive complex restoration solutions involving multiple feeders using where necessary automatic load transfer schemes to achieve optimum restoration.

• The intelligent substation platform must be expandable without limitation to the size of the network island being automated.

• The model driven switching solution must be dynamic to seek the optimum restorative state ensuring no loss of customers during the restoration nor incurring network violations. If complete solutions are not possible without violation, the algorithm must be able to reduce the un-faulted network size to maximize the area of restoration. Predefined scripted responses are not acceptable in order to minimize maintenance and maximize successful restoration scenarios.

• The feeder automation solution must function with existing switches of various manufacturers. It must not be switch manufacture dependent.

• Post fault event reconstruction with SOE alarm recording is required for auditing purposes.

• The system must be able to perform its function without supervision (unmanned). However it must be able to report time tagged alarms, application messages and events to a centralized control center or user.

• With supervision the system must be capable of switching to advisor mode, presenting the downstream switch plan for approval and execution.

• A restore back to normal switching plan must be generated for advisory mode implementation.

• The system must support NERC CIP security.

In summary, as the grid automates, the information gathering and decision making must first of all be distributed to the substation or feeder. The control center must supervise the strategic management of the grid while the distributed intelligence manages the tactic operation of the feeders. Both sites must operate from a unified real-time model of the network as an integrated solution comprising the Smart Grid.