To address common challenges in the manufacturing of high-end polymer materials—such as significant product quality fluctuations and long RD cycles—we have established multi-scale modeling technologies and developed real-time control methods driven by both data and mechanisms. This has resulted in a closed-loop optimization system connecting Process, Structure, and Performance.These achievements have been implemented by leading enterprises such as SINOPEC Shanghai Petrochemical and Zhenhai Refin

Targeting the demands of high-radiation space environments and complex industrial scenarios, achieving high-precision manipulation and autonomous maintenance driven by embodied intelligence remains a critical challenge. We have innovatively developed fine-grained scene–object modeling methods for complex space and industrial environments, along with cross-scenario, cross-object generalizable embodied collaborative manipulation techniques. Furthermore, we have built a domain-specific embodied fou

The development and large-scale deployment of emerging biotherapyand biomedicine technologies encompass multiscale biological processesfrom molecular interactions to macroscopic bioreactor operations. Within this context, microfluidic technology serves as the cornerstone for achieving precise bioprocess control. To address the common demands of biosensing, bioediting, bioculture, and high-throughput screening, we have conducted innovative research spanning microfluidic actuation, sensing and det

Feature Extraction and Recognition Methods for Brain-Computer Interfaces（Jinjing）Brain-Computer Interfaces (BCIs) have reshaped the paradigm of human-computer interaction and represent a frontier field within national science and technology strategies.By investigating the inherent mechanisms of neural signal distortion, this work innovatively designs optimal encoding combination methods, breaking through the limitations of global information dependency and the lack of neurophysiological constrai