A chaotic system-based signal identification Technology: Fault-diagnosis of industrial bearing system

•A smart fault-diagnosis method with data‐mapping strategy and KNN algorithm is proposed.•The raw data is converted to multi-dimensional chaotic signals-visualization.•Data around 0.4 s are collected for training unit preventing from any potential deviation.•The key features values of EFV and SEFV are collected to further define the boundary threshold.•Results show that this mothed is effective, accuracy and cost less computational resources.