In daily life, cascading failures in complex networks pose a significant threat to the reliability and stability of systems. To address the limitations of traditional static complex networks in accurately depicting node interactions and failure mechanisms, cascading failures based on temporal networks were investigated. Firstly, temporal networks were constructed using activity driven models, memory-based activity driven models, and attraction activity driven models, and their connectivity characteristics were analyzed to reflect the changes of nodes and edges over time. Secondly, the cascading failure process in temporal networks was analyzed, and corresponding robustness index was proposed. Finally, cascading failure analyses were conducted on both randomly constructed temporal networks and real email temporal networks. The results indicate that temporal networks outperform static networks in simulating cascading failure processes, contributing to a better understanding of the operational mechanisms of real networks and their temporal evolution.