Using temporal causal models to isolate failures in Power System protection devices

Power transmission systems are instrumented with a network of fast-acting protection devices that detect anomalies and arrest the fault propagation, thereby isolating the faulty components to protect the remaining system. However, often a local protection action leads to cascading effects to other regions resulting in blackouts. Also, inherent faults in protection devices such as relays and breakers could lead to failed or misguided protection action. A fault model associated with such a system needs to capture the dynamic behavior of the system as well as protection scheme in both nominal and faulty modes of operation. With an effort to capture the temporal behavior and the fault propagation of the system and its autonomous protection mechanism, we propose a new modeling formalism - the Temporal Causal Diagram (TCD). In this paper, we will describe the TCD modeling formalism and apply it to create fault-models that capture the fault propagation and state evolution of the Power Systems and their autonomous protection units. Further, we showcase simulation models derived from the TCD models and use these to simulate single and multi-fault scenarios in power transmission systems.