In the context of industrial big data, relying on massive data, the utilization of digital modeling technology and applying data science theoretical methods for the design, optimization, production, and operation of complex engineering systems has gradually become the research paradigm for the next generation of industrial engineering. This project focuses on Prognostics and Health Management (PHM), a signature technology for high value engineering systems. Macro and micro digital PHM models will be constructed employing state-of-the-art deep learning techniques including Transformers, graph neural networks, physics-informed neural networks, and hierarchical reinforcement learning. Additionally, equipment integrated operation & maintenance decision optimization based on PHM digital twins will be explored. By seamlessly integrating PHM with digital twins, complex networks and artificial intelligence, equipment operation processes can be systematically optimized. Relevant outcomes will provide theoretical and technical support for the digitalization and Intelligentization of complex engineering system research, design, manufacturing, testing, and life cycle management.