Making fault data and non-SCADA data accessible for predictive analysis in data historians

Abstract

Southern Company consists of four electric utilities; Alabama Power, Georgia Power, Gulf Power and Mississippi Power. In these four states, Southern Company has 120,000 square mile of service territory and approximately 3400 substations. Analyzing power system disturbances is a critically important task for any electrical utility, especially Southern Company. When a power system event occurs, such as an outage, Southern Company needs a clear picture of the root cause of the event in order to take the optimal corrective action. Appropriate responses to system events will allow the utility to improve grid reliability, safety, profitability and to increase customer satisfaction. Smart relays have been widely deployed by electrical utilities to protect the power system. Smart relays usually contain SCADA data, non-SCADA data (useful real-time data not collected by SCADA system) and fault files which capture detailed information about power system disturbance and assist utilities in determining the root cause of the disturbance. But often the valuable non-SCADA data and fault files are not collected, as they require costly time and labor expenses to manually retrieve this additional information from each relay. This collection is time consuming, and delays the utility's ability to respond quickly to system events. Collecting this information is further complicated by the many different relay vendors and product models. These differences include determining when a fault file is available, knowing how to collect this fault file from the device and deciphering the format of the fault file. As a result, Southern Company has sought a solution that not only centrally archives non-SCADA data, but also fault file data in a data historian. The solution provides archiving of non-SCADA data by directly interfacing with the data historian. Archiving of fault file data, however, requires secure automated collection of fault files from numerous smart relays into a centralized corporately-accessible location. Centrally storing these files in a common location provides for system-wide review of disturbance data. The centralized location contains event viewing tools for authenticated users to analyze the fault files. Additionally, the system extrapolates the key time series data from the fault files, performs calculations (for example; breaker analytics) and sends all data (extrapolated and calculated) to the data historian for further historical analysis. Historical analysis on both the fault data and non-SCADA data allows Southern Company to perform predictive maintenance, event analysis, load contingency studies, reliability improvement studies, substation transformer life studies, power quality studies and much more. Overall, Southern Company's expected results from this initiative are to better prioritize money, prioritize labor resources and improve system reliability. The paper outlines the process for automated fault file collection highlighting the importance and benefits that a company can receive by archiving both fault file data and non-SCADA data in a data historian.