Although deep learning technologies have been widely exploited in many fields, they are vulnerable to adversarial attacks by adding small perturbations to legitimate inputs to fool targeted models. However, few studies have focused on intelligent networking in such an adversarial environment, which can pose serious security threats. In fact, while challenging intelligent networking, adversarial environments also bring about opportunities. In this paper, we, for the first time, simultaneously analyze the challenges and opportunities that the adversarial environment brings to intelligent networking. Specifically, we focus on challenges that the adversarial environment will pose on the existing intelligent networking. Furthermore, we investigate frameworks and approaches that combine adversarial machine learning with intelligent networking to solve the existing deficiencies of intelligent networking. Finally, we summarize the issues, including opportunities and challenges, which can allow researchers to focus on intelligent networking in adversarial environments.
Thermoacoustic instabilities often present in gas turbine combustors, especially in land-based heavy-duty gas turbines. Normally, thermoacoustic modes can be classified into two sets: acoustic mode and intrinsic mode (ITA mode). Low order network models are usually used to study thermoacoustic
instabilities and passive control methods such as liners and Helmholtz dampers are often used to suppress these instabilities. The addition of Helmholtz dampers could suppress the original thermoacoustic modes. Meanwhile, it could also introduce a new mode, the eigenfrequency of which is related
to the parameters of the Helmholtz damper, including its resonant frequency and neck mean flow Mach number. The damping effects depend on the system parameters and could be different on the two categories of original thermoacoustic modes. Based on a rational design, both original modes can
be controlled by one damper. The new mode introduced by the damper could interplay with original modes to form new exceptional points with certain parameters. In that case, the eigenfrequency and mode shape of original modes could be vastly influenced. The theoretical results will be verified
by numerical simulations.
Bluff body stabilised non-premixed flames are usually used as pilot flames in lean-premixed combustors. Experiments are conducted to investigate the characteristics of the flame. Typical flame modes are investigated in both stable and unstable conditions. The flow structures, the reaction zone, and the dynamics of unstable flames are measured with PIV, ICCD and a high speed camera, respectively, based on which the inherent mechanisms that influence the configuration and stabilisation of the flame are analysed. Stable flames are apparently influenced by the mixing characteristics in the recirculation zone. Flame detachment, a typical phenomenon of stable flames in a turbulent air flow, can be explained by the distribution of fuel concentration in the recirculation zone. The Reynolds number of air has different effects on different parts of the flame, which results in three unstable flame modes at different Reynolds numbers of air. These results could be helpful for the design of stable burners in practice.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.