Using Artificial Intelligence to Classify IEDs’ Control Scope from SCL Files

IEC 61850 is one of the most accepted standards worldwide for the automation of electrical substations. This standard uses Substation Configuration Language (SCL) for describing the data model and services from electrical substation components, and SCL files are used for the integration of these components throughout the substation. In this context, the integration of bay level Intelligent Electronic Devices (IEDs) into the station level demands a detailed analysis of the IED’s control scope in SCL files and advanced know-how in IEC 61850, increasing the complexity in the engineering process. Hence, this work presents a method to automate the analysis of the control scope of IEDs using SCL files, generating their respective control system object. This is achieved via Machine-Learning (ML) concepts, such as supervised learning and classification algorithms. IEDs used for control and protection of feeder and transformer systems were analyzed, and control system objects were generated for them. The results indicate that the developed method makes it possible to classify the control scope of IEDs using the SCL files from the bay level. This method is a unique development for application in the engineering process of digital substations, reducing the complexity of a critical step towards substation automation.