An intelligent manufacturing system is a composite intelligent system comprising humans, cyber systems, and physical systems with the aim of achieving specific manufacturing goals at an optimized level. This kind of intelligent system is called a human–cyber–physical system (HCPS). In terms of technology, HCPSs can both reveal technological principles and form the technological architecture for intelligent manufacturing. It can be concluded that the essence of intelligent manufacturing is to design, construct, and apply HCPSs in various cases and at different levels. With advances in information technology, intelligent manufacturing has passed through the stages of digital manufacturing and digital-networked manufacturing, and is evolving toward new-generation intelligent manufacturing (NGIM). NGIM is characterized by the in-depth integration of new-generation artificial intelligence (AI) technology (i.e., enabling technology) with advanced manufacturing technology (i.e., root technology); it is the core driving force of the new industrial revolution. In this study, the evolutionary footprint of intelligent manufacturing is reviewed from the perspective of HCPSs, and the implications, characteristics, technical frame, and key technologies of HCPSs for NGIM are then discussed in depth. Finally, an outlook of the major challenges of HCPSs for NGIM is proposed.
With the development of modern information technology—and particularly of the new generation of artificial intelligence (AI) technology—new opportunities are available for the development of the intelligent machine tool (IMT). Based on the three classical paradigms of intelligent manufacturing as defined by the Chinese Academy of Engineering, the concept, characteristics, and systemic structure of the IMT are presented in this paper. Three stages of machine tool evolution—from the manually operated machine tool (MOMT) to the IMT—are discussed, including the numerical control machine tool (NCMT), the smart machine tool (SMT), and the IMT. Furthermore, the four intelligent control principles of the IMT—namely, autonomous sensing and connection, autonomous learning and modeling, autonomous optimization and decision-making, and autonomous control and execution—are presented in detail. This paper then points out that the essential characteristic of the IMT is to acquire and accumulate knowledge through learning, and presents original key enabling technologies, including the instruction-domain-based analytical approach, theoretical and big-data-based hybrid modeling technology, and the double-code control method. Based on this research, an intelligent numerical control (INC) system and industrial prototypes of IMTs are developed. Three intelligent practices are conducted, demonstrating that the integration of the new generation of AI technology with advanced manufacturing technology is a feasible and convenient way to advance machine tools toward the IMT.
Intelligent technologies are leading to the next wave of industrial revolution in manufacturing. In developed economies, firms are embracing these advanced technologies following a sequential upgrading strategy—from digital manufacturing to smart manufacturing (digital-networked), and then to newgeneration intelligent manufacturing paradigms. However, Chinese firms face a different scenario. On the one hand, they have diverse technological bases that vary from low-end electrified machinery to leading-edge digital-network technologies; thus, they may not follow an identical upgrading pathway. On the other hand, Chinese firms aim to rapidly catch up and transition from technology followers to probable frontrunners; thus, the turbulences in the transitioning phase may trigger a precious opportunity for leapfrogging, if Chinese manufacturers can swiftly acquire domain expertise through the adoption of intelligent manufacturing technologies. This study addresses the following question by conducting multiple case studies: Can Chinese firms upgrade intelligent manufacturing through different pathways than the sequential one followed in developed economies? The data sources include semistructured interviews and archival data. This study finds that Chinese manufacturing firms have a variety of pathways to transition across the three technological paradigms of intelligent manufacturing in nonconsecutive ways. This finding implies that Chinese firms may strategize their own upgrading pathways toward intelligent manufacturing according to their capabilities and industrial specifics; furthermore, this finding can be extended to other catching-up economies. This paper provides a strategic roadmap as an explanatory guide to manufacturing firms, policymakers, and investors.
In this research, an auxiliary illumination visual sensor system, an ultraviolet/visible (UVV) band visual sensor system (with a wavelength less than 780 nm), a spectrometer, and a photodiode are employed to capture insights into the high-power disc laser welding process. The features of the visible optical light signal and the reflected laser light signal are extracted by decomposing the original signal captured by the photodiode via the wavelet packet decomposition (WPD) method. The captured signals of the spectrometer mainly have a wavelength of 400–900 nm, and are divided into 25 sub-bands to extract the spectrum features by statistical methods. The features of the plume and spatters are acquired by images captured by the UVV visual sensor system, and the features of the keyhole are extracted from images captured by the auxiliary illumination visual sensor system. Based on these real-time quantized features of the welding process, a deep belief network (DBN) is established to monitor the welding status. A genetic algorithm is applied to optimize the parameters of the proposed DBN model. The established DBN model shows higher accuracy and robustness in monitoring welding status in comparison with a traditional back-propagation neural network (BPNN) model. The effectiveness and generalization ability of the proposed DBN are validated by three additional experiments with different welding parameters.