Data Sources For Ai Analysis Applications In Iec 61850 Based Digital Substations

IEC 61850 has been used for more than 10 years and we already have thousands of digital substations his service all over the world. It brings significant benefits, such as improvements in the reliability, security and efficiency of the operation of electric power systems under different conditions. One of the characteristics of this digitized substations world is the huge amount of data that is becoming available and impossible to process by human beings. That is why there is a growing interest in artificial intelligence applications in electric power systems. The paper starts with an overview of different artificial intelligence technologies. It describes several commonly applied machine learning techniques from the point of view of their possible application in digital substations: • Supervised learning • Unsupervised learning • Reinforcement learning • Deep learning The second part of the paper describes the components of an IEC 61850 based digital substation from the point of view of the requirements for data feeding into artificial intelligence based applications. It first describes the different types of data available in the digital substation, including the sampling rates and other factors that have an impact on the amount of data available: • Raw data represented in the model by streaming sample values from electrical or non- electrical sensors • Status data from the switchgear and other primary equipment in the substation • Synchrophasor measurements from the P class or M class • Status data from multi-functional protection and control IEDs • Event report from multi-functional protection and control IEDs • Time synchronization data from different clocks • Output data from substation level applications • Data received from remote substation • Data received from system integrity protection schemes • Data received from the control center The use of various combinations of the different types of data to be used as inputs to machine learning algorithms is presented based on specific examples. This is followed by the analysis of the different types of devices in the digital substation and the types of data that they produce: • Standalone merging units • Embedded merging units • Switchgear control units • Process interface units • Non-electrical sensor units • Phasor measurement units • Multifunctional intelligent electronic devices The next part of the paper is concentrated on the communications architecture of the substation protection, automation and control system and how the different types of devices described earlier are connected to the substation network. Different topologies, such as star, ring or hybrid are considered, as well as redundant communication protocols such as PRP and HSR providing higher level of reliability and better performance for protection applications. The allocation of the machine learning applications within the components of the substation protection, automation and control system is then considered. Some of them that require extremely high sampling rates might be implemented at the process level, while others may reside in protection IEDs. However, it is expected that most of the artificial intelligence applications will be running at a central substation server that is also receiving data from outside of the substation. Such data can provide additional information about the state of the electric power system that is helpful in the decision-making process and is not available locally at the substation.