Wensong Liu

Standard data-driven fault diagnosis methods typically rely on the assumption of identical label spaces across source and target domains. However, the emergence of novel failure modes in the target domain invalidates this premise, presenting a complex open-set fault diagnosis (OSFD) challenge where traditional closed-set techniques fail to generalize. To bridge this gap, we propose a novel Adaptive Prototype Network. This framework operates through a progressive two-stage paradigm: a warm-up phase that leverages prototype learning and domain adversarial training to establish coarse-grained alignment, followed by a refinement phase that optimizes the model via adaptive clustering and dual-risk minimization. Specifically, we introduce a dynamic sample weighting mechanism guided by a learnable unknown detector to suppress negative transfer. Extensive experiments on two distinct datasets demonstrate that proposed method effectively aligns shared classes while robustly segregating unknown faults, outperforming state-of-the-art baselines.

With the tremendous necessity of multimodal application services, we have devised an integrated four-levels of question and answering (QA) Robot to handle a wide range of automatic tasks ranging from language, words, voice, image, and so on. The intelligent QA Robot, FMIQA, firstly analyze and normalize the corresponding input queries from the users, and then categorize the application areas of the tasks. After the aforementioned two-levels of operations, FMIQA has identified the intention of multiscale tasks. FMIQA then moves forward to the third phase to use the multiple structure of data storage and deals with the information tasks generated from the second-level phase. Through multiple rounds of queries, FMIQA obtains the final answer and analyzes the forms of output information. In the fourth-level phase, FMIQA aligns the form of input and output, and then provides the corresponding form of the answer. The information processing with four-levels allows FMIQA to have less querying and searching time, and it has higher precision and correlation in terms of the answers to the queries.