Methodology for the Analysis of Series Arc Fault Algorithms

Solar energy has been growing in recent years, becoming increasingly common in family homes. However, improper installation, lack of preventive maintenance, and aging can accelerate the degradation of solar system components, increasing the risk of serious failures. Arc faults are among the most significant risks for residential solar systems due to the high risk of fire that can be initiated. The use of AI in fault detection systems is growing to mitigate these effects. Because of this, a typical scenario is needed for all algorithms to test their capabilities. Due to the existence of various arc fault models and the inherent components of the system aiding in the difficulty of detection, standardization in the evaluation of algorithms is necessary. This paper reviews the types of arc fault detection and proposes data collection to assemble a database to evaluate emerging detection algorithms.