Our blogs Blogs home
Energy in Transition

Energy in Transition


Dynamic Parameter Estimation: Bridging the gap between simulation and real world power system response

Regional transmission planning authorities including utilities and/or RTOs/ISOs currently rely on three different types of planning models to assess the short- and long-term transmission needs of the system from a reliability standpoint:

  • Steady State Models
  • Short Circuit Models
  • Stability Models

Of the aforementioned models, the validation of the steady state and short circuit models for various power system equipment is relatively straightforward. Additionally, while the steady state models are useful in providing indications of potential transmission system concern, the key insight into the response of a highly non-linear and dynamic power system network is obtained from dynamic simulations. Dynamic modeling and studies which reveal the dynamic characteristic of a power system play more critical roles in providing transmission system planners with key insight into the system performance under varying severities of system disturbance. However, the dynamic simulation results are heavily dependent on the dynamic parameters of the system components such as generators, exciters, governors, loads and any Flexible AC Transmission System (FACTS) devices in the power system network.

Most dynamic parameters for power system equipment, barring dynamic load models, are derived from manufacturer specification and/or field tests performed prior to the commissioning of the equipment in the transmission system. However, in numerous instances, differences are observed in the dynamic parameters of the power system equipment utilized in the model vis-à-vis that reflecting actual system equipment performance. There could be numerous explanations for the same a few of which are mentioned below:

  • The parameters derived from field tests and/or manufacturer specification do not match those deployed in the field due to special operational conditions
  • The parameters may have drifted over a long period of operation
  • The parameters may have been changed during practical operation of the equipment
  • Other factors

It is common to see mismatches in the dynamic parameters associated with various power system equipment when comparing modeling databases and actual system equipment in the field. However, the accuracy of the dynamic model and associated data set, in terms of being reflective of the actual system response, is a function of the following:

  • The number of dynamic parameters exhibiting such discrepancies
  • The magnitude of the parameter variation between that depicted in the model and those employed in the field
  • The significance of the parameter in terms of affecting the response of key power system quantities associated with the power system equipment

In large power system networks, widespread inaccuracy of the power system equipment dynamic parameters can lead to drastic differences in the post-disturbance response of key system quantities when comparing model and actual system response. Figure 1 depicts a comparative assessment of the actual vs model system voltage and active power response associated with the August 10, 1996 WSCC blackout event. While the model simulation results depict acceptable post-disturbance performance, the actual system shows poorly damped oscillations of the north-south swing mode associated with the Pacific AC Intertie (PACI) which eventually lead to a system blackout. WECC has since made significant strides in the development of specific generator model validation testing guidelines to ensure accurate dynamic data for generation facilities in network simulation models.

Figure 1: Pacific AC Intertie Voltage and Power Swings, Actual Vs Simulation Responses, Western Systems Coordinating Council August 10, 1996 Event

Figure 1: Pacific AC Intertie Voltage and Power Swings, Actual Vs Simulation Responses, Western Systems Coordinating Council August 10, 1996 Event

To that effect, the accurate calibration of dynamic models and associated data sets to reflect actual system response has been identified as a key ingredient for transmission planning from a reliability and security standpoint. The North American Electric Reliability Corporation (NERC) outlined a White Paper on Power System Model Validation in May 2010. The white paper identifies a “top-down approach to model verification; comparisons with measured data indicate the quality of the overall model”.

DNV GL’s Power System Planning (PSP) group has been working on the development of a PMU-based dynamic parameter estimation tool with some of the salient features of the tool outlined below:

  • Ability to utilize more than one PMU at “boundary bus location” and the data thereof to calibrate more than one dynamic device simultaneously
  • Use of system reduction technique to eliminate dependency on the need for power flow model for instant snapshot associated with dynamic event
  • Ability to make the dynamic model calibration process generic in order to calibrate models associated with any of the following dynamic devices:
    • Conventional Generator Models
    • Wind Generator Models
    • Dynamic Load Models
    • Dynamic VAR Device Models

Figure 2 depicts the process overview associated with DNV GL PSP’s proposed PMU-based dynamic parameter estimation approach.

Figure 2: DNV GL’s PMU-based Dynamic Parameter Estimation Approach, Process Overview

Figure 2a: DNV GL’s PMU-based Dynamic Parameter Estimation Approach, Process Overview

Figure 2b: Hybrid Dynamic Simulation Approach Overview

Figure 2b: Hybrid Dynamic Simulation Approach Overview

As evident from the discussion presented above, well-validated and calibrated dynamic models are key to the future of transmission planning and more importantly a key ingredient in the industry’s transition from preventive to predictive system planning. As part of its thought leadership effort in the energy and power industry, DNV GL’s Energy Advisory Division continues to test and refine the dynamic parameter estimation tool including collaborating with electric utilities to test large scale deployment/testing on large practical power system networks.

0 Comments Add your comment

Reply with your comment

Your email address will not be published. Required fields are marked *