Key Laboratory of Space Utilization, Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences, Haidian District, Beijing 100094, China
Auto-generated profile, based on publicly available information
Yang Wang
Key Laboratory of Space Utilization, Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences, Haidian District, Beijing 100094, China
In the real scenario of engineering, the failure time of rotating machinery is generally much less than when it is in a healthy condition. Considering the cost, it is unrealistic to conduct the large-sample and long-time failure tests. This results in the problem of data imbalance in fault diagnosis, i.e. the number of normal samples far exceeds that of the fault ones, which seriously affects the accuracy and stability of fault diagnosis. For the settlement of the above problem, an auxiliary classier Wasserstein generative adversarial network with gradient penalty (ACWGAN-GP) is proposed in this paper, which is capable of generating high-quality samples for the minority classes stably utilizing an imbalanced training set. In the experiment of fault diagnosis, the generated samples firstly go through the availability verification, and then are employed to augment the imbalanced dataset gradually. The final results show that the proposed method is competent for the generation of data, which is highly similar to real samples, and the accuracy of fault diagnosis has effective improvement as the imbalanced dataset is gradually expanded to equilibrium. Additionally, the ACWGAN-GP model presents better performance in sample generation than other widely-used methods.
SLS (Selective Laser Sintering) has been developing rapidly since its initial invention for non-metal materials by Texas University. Nowadays, Direct Metal Laser Fabrication (DMLF), as a variant of SLS technique, has been investigated intensively which is aimed at rapid manufacturing of end-use metal products with full functions. For describing the stability and properties of DMLF process, Top Surface Quality (TSQ) was put forward in this paper, which could be a unique and crucially important feature compared with traditional manufacturing methods. Through the systematic and detailed analysis of DMLF process using related theories of additive manufacturing technologies, it was revealed that TSQ was the key factor for controlling the stability of DMLF process and thus tailoring final properties of metallic parts. TSQ was defined as the surface morphology in macro and micro scopes in laser scanning area of unit layers during DMLF, and could be characterized by three key elements: flatness, compactness and cleanliness. Only good TSQ could ensure the stability of DMLF process and excellent performance of metal parts in theory. The flatness was the significant factor to assure the shaping during DMLF while the compactness and cleanliness are the decisive factors to assure the final properties of metal part for DMLF. As an example, the typical top surface defects and their contributing factors in DMLF for Cu-based metal powder mixtures were investigated thoroughly according to the proposed definition and requirements. Moreover, the specific controlling methods of TSQ were provided and discussed. Eventually, DMLF of three-dimensional Cu-based metal sample with complicate structure was successfully performed by taking some effective measures for adjusting TSQ parameters.