A Fuzzy Logic Approach to Evaluate Transformer Fleet Risk From Geomagnetic Storms
This paper presents a simple approach to evaluate risk to transformer units from geomagnetic storms. A simple fuzzy logic based approach was used to develop the model which is capable of categorizing transformers in a fleet into three risk categories, i.e., high, medium and low. This model may be used as a first screening step to evaluate a fleet of transformers without conducting time consuming simulations and studies. Critical factors that affect geomagnetically induced current (GIC) flow are used as input parameters to the model. These factors include location-specific, equipment-specific and geomagnetic storm event-specific factors. Location-specific factors include geomagnetic latitude of the location, earth conductivity structure at location and distance of location from coast. Equipment-specific factors include transformer rating and age. The storm event-specific factor of geoelectric field strength was used which is an indication of the return period of the geomagnetic disturbance event. The paper describes the implementation of this model to evaluate fleet risk for a 1 in 100 year event and the Carrington event (largest recorded geomagnetic storm in history). The fuzzy logic membership functions for the inputs are described in detail and the performance of the fuzzy logic model is evaluated.