By David Schoenwald
Lightly damped electromechanical oscillations are a source of concern in the Western Interconnection (WI). There are two primary motivations to increase damping of inter-area oscillations. First, if damping is insufficient, oscillations may lead to system-wide tripping events, and in turn to a series of cascading outages. The 1996 system break-up across the west coast of North America can be in part attributed to undamped oscillations. Avoiding these large-scale power outages provides a significant financial incentive in damping inter-area oscillations. Second, power transfer through long transmission corridors in western North America is often constrained due to stability concerns and limited by poorly damped electromechanical oscillations. Thus, additional damping may increase the power transfer capacity. Recent development in reliable real-time wide-area measurement systems (WAMS) based on phasor measurement units (PMUs) has enabled the potential for large-scale damping control approaches, in order to stabilize critical oscillation modes. A recent research project has focused on the development of a relevant prototype feedback modulation controller for the Pacific DC Intertie (PDCI). The damping controller utilizes real-time wide-area PMU signals to form a power command to modulate the real power on the PDCI. Recent results demonstrate desirable performance and improved modal damping, consistent with previous modeling and simulation studies.
A collaborative effort between Sandia National Laboratories (SNL), Montana Tech University (MTU), and Bonneville Power Administration (BPA), dating back to 2013, was launched to design, develop, and demonstrate an active damping control system (DCON) on the PDCI. However, the original idea to modulate PDCI power flow to damp inter-area oscillations was first designed and tested in 1975. The original design utilized the real power flow on the California-Oregon Intertie (COI) as the feedback signal. Even though this method provided damping to low frequency modes of oscillation, further analysis determined that the local AC power flow feedback signal, had a transfer function zero, which limited the gain of the controller and worsened oscillations at higher frequencies. The modern controller deployed is able to avoid this problem because it incorporates GPS time synchronized WAMS to improve damping. This data is available due to the recent deployment of PMUs throughout the WI, which provide fast, reliable, system-wide measurements.
Currently, the primary approach to mitigate grid oscillations and avoid blackouts in the WI is to operate well below transmission capacity, which is not economical. The DCON uses measurement data, acquired in real time, from PMUs, recently installed throughout the WI. This measurement data serves as a feedback signal to inform the controller as to how much power to add (or subtract) to the power flow on the PDCI. This carefully controlled “injection” of power to the PDCI is the action that damps oscillations in the grid. A supervisory system, integrated into the controller, ensures a “do no harm” policy for the grid in which damping is never worsened. Through the improvement of damping these inter-area oscillations, the DCON has the potential to allow increased power transfers in the WI.
The control law is a proportional action on the frequency difference between the northern and southern areas of the WI. The power command, which changes the flow on the PDCI by up to 125 MW, is sent to the PDCI controls at the northern terminal of the PDCI (BPA’s Celilo Converter Station in The Dalles, Oregon). The DCON was carefully designed such that it can effectively interact with inter-area oscillations between 0.2 and 1.0 Hz. Oscillations above 1.0 Hz are significantly attenuated and oscillations below 0.2 Hz are effectively attenuated through the use of frequency difference as a feedback signal. This control strategy provides damping to the primary north-south oscillatory modes in the WI without interacting with speed governor actions.
Closed-loop tests initially conducted in September 2016 and repeated in May 2017 consistently showed that damping of the primary north-south mode improved by 4.5 percentage points over the open-loop tests, when performed in response to a Chief Joseph brake insertion (1400 MW brake pulse for 0.5 sec). Additional tests performed in response to square wave pulses and forced oscillations provided by a probing signal generator, also show improved damping in closed-loop vs. open-loop control operation. In all the tests conducted, there was no instance of the damping controller worsening any peripheral modes or otherwise causing harm to the grid.
Due to the existence of satellite synchronized clocks at all PMU sites and at the controller site, time delays in the DCON could be precisely measured and characterized. Excessive delay could not be tolerated by this control system, so extensive studies were carried out to verify that the delays were well within tolerances. These studies concluded that the effective round-trip time delay never exceeded 113 msec during these tests. This is well within tolerances, which are in the range of 150 – 200 msec. The variance of the delays was also studied and shown to be well-behaved.
New design features and significant additional testing will be necessary before the DCON can be deployed at an operational status on the grid. Specific design features to be studied include the estimation of open-loop gain and phase margins of the grid as a means to monitor damping effectiveness and margins of stability. Another feature under consideration is the incorporation of a deadzone in the modulated power command to reduce excessive actuation effort. Finally, the issue of DCON operation near saturation limits on the power command during near-capacity PDCI power flow will be investigated.
The DCON is the first successful wide-area grid demonstration of real-time feedback control using PMUs in North America. This is a game-changer, enabling the use of widely-distributed networked energy resources that have the potential to transform our existing power grid into the future smart grid. Benefits that the DCON is capable of delivering, once operational, include: (1) Additional reliability to the grid from improved damping of electromechanical oscillations. (2) Additional contingency management of the grid under stressed system conditions. (3) Higher power limits in specific transmission corridors. (4) Reduction and/or postponement in new transmission capacity expansion.
The author gratefully acknowledges the support provided by the BPA Office of Technology Innovation, Dr. Terry Oliver, CTO, the DOE Office of Electricity (DOE/OE) Transmission Reliability Program, Mr. Phil Overholt, Manager, and the DOE/OE Energy Storage Program, Dr. Imre Gyuk, Manager.
David Schoenwald, IEEE Senior Member, is a principal member of the technical staff in the Electric Power Systems Research Department at Sandia National Laboratories. In his current work, he focuses on control system design for damping inter-area power system oscillations, mitigation of network-induced issues in control systems employing real-time measurement feedback, and development of performance standards for grid-scale energy storage systems. Dr. Schoenwald is currently the Technical Chair for the 2017 Electrical Energy Storage Applications & Technologies (EESAT) Conference. Previously, he was an associate editor for IEEE Transactions on Control Systems Technology (2000-2007), IEEE Control Systems Magazine (1996-2000), IEEE Control Systems Society Conference Editorial Board (1993-2016), and he served on the Board of Governors of the IEEE Control Systems Society (2003-2006). He is a Senior Member of the IEEE Power & Energy Society, IEEE Control Systems Society, and the IEEE Robotics & Automation Society. Dr. Schoenwald received his Ph.D. in electrical engineering from Ohio State University.