1:45 - 2:45
October 25, 2018
Grand Cypress A
301: Prediction of Cold Load Pickup using AMI and SCADA Data
Session Category : Grid Operations
Usually when a distribution circuit is restored after an extended outage, the load demand is greater than before the outage as a result of loss of diversity in the system. As all the loads are brought online at the same time and the state of the equipment is “cold”, a very high load demand is experienced by distribution feeders. This phenomenon is referred to as cold load pick up. Attempting to pick up this load can be problematic because the demand can exceed the demand that was experienced before the outage and can trigger an overcurrent breaker trip.
Predicting the cold load pickup behaviour enables optimal sequence of operations to restore power quickly while minimizing the likelihood of operating a protective device or damaging equipment. The prediction of cold load pickup values is a complex function influenced by pre-outage behaviour, duration of the outage, time of day, ambient temperature, and type of customer. Recently, Oncor Electric Delivery embarked on a data science journey into advanced data analytics to predict cold load pickup at the feeder level and gain additional insights into the load behaviour utilizing historical AMI and SCADA data with an ultimate goal of providing their Distribution Operators with tools to predict load behaviour during outage restoration