Time Domain Cascading Analysis Framework with Integrated Substation Configurations and Protection Systems

Abstract

Blackouts represent the most severe consequence of cascading outages, where a significant portion of the power system loses service, often within a short time frame. These large-scale outages can result from an uncontrolled propagation of disturbances, aggravated by inadequate protection coordination, insufficient system reserves, or delayed operator response. By studying the conditions that lead to blackouts through cascading outage models, system operators and planners can identify weak points, improve protective schemes, and develop more resilient operational strategies to prevent such catastrophic failures. Most of the available cascading outage tools focus mainly on steady-state models of power systems. With increasing complexity of power networks due to renewable integration, complex controls and protection schemes, the dynamic modeling and time domain simulation of power systems became a necessity for accurate cascading outage analysis. In a substation, circuit breakers, isolators and bus bars are used to establish the electrical connectivity of the power equipment with seven standard station configurations based on the reliability requirements. The protection and control devices (e.g. relays, bay control units) in substations use multitude of input measurements obtained through different Current Transformers (CT) and Voltage Transformers (VT). Depending on the station configuration and reliability requirements substations employ different types of CT-VT arrangements. It is very important to consider the CT and VT arrangements for accurate protection system modeling, especially for the substation configuration specific protections, such as breaker failure protection and teed protection as they play a significant role in cascade propagation. Representation of different substation configurations require Node-Breaker (NB) models. However, the existing power system stability assessment tools such as PSSE, PSLF, and DSA Tools use Bus-Branch (BB) models for representing the power system network. Moreover, separate protection modeling tools such as CAPE need to be integrated to represent detailed protection systems, CT-VT arrangements associated with the substation configurations. These tools are not practically suitable for time domain cascading analysis of large systems. This thesis develops a transient stability simulation based cascading analysis framework with integrated substation configurations, CT-VT arrangements and protection systems. The applicability of Sparse Tableau (ST) has been investigated for transient stability simulations (TSS) of large power systems including NB models. Since the standard test systems data for TSS is typically available in the BB data format, algorithms are developed to convert the BB data into NB data supporting seven widely used station configurations. However, as the system size increases, ST order increases considerably. The concept of Multi-Area Thevenin Equivalents (MATE), divides large network into decoupled partitions, allowing parallel simulation of each partition. However, MATE was developed for BB Models. So, the MATE concept has been extended to NB models in this thesis. A combination of spatial parallelism through the MATE method and task-level parallelism is employed to maximize the computational speedup of the proposed NB-TSS. The concept of Signal Matrices (SM) is introduced to model different types of CT and VT arrangements. Algorithms are developed for automated deployment of protective relays, along with their corresponding CTs, VTs, and associated CBs. A space-parallel OpenMP-C++ based Time Domain - Cascading Analysis Tool with Substation configurations (TD-CATS) is developed to simulate Relay Failure (RF), CB Failure (BF) and Breaker Failure Protection (BFP) events which lead to higher order contingencies depending on the station configuration. A cascade assessment module consisting of component outage tracking, load shed tracking, relay supervision and stability tracking is developed to study the cascade propagation mechanism. The voltage and frequency at all buses are used as static stability indices. Transient stability index (TSI) and rate of machine acceleration (ROMA) are used to assess the stability of the system. These indices are employed to define the stopping criteria in TD-CATS. To quantify the risk associated with cascading events, the failure probabilities of protective relays and CBs combined with the magnitude of load shed are utilized. The TD-CATS capabilities in simulating cascading events are demonstrated on New England, 10 generator 39-bus and Polish 327 generator 2383-bus test systems. The impact of station configurations, CT-VT arrangements and the order of connectivity of power system elements within a substation on the propagation of cascade and associated risk has been studied. These studies reveal the inefficiencies of existing TSS tools for high order cascade events analysis.