Fault diagnosis for rolling elements in rotating machinery persistently receives high research interest due to the said machinery's prevalence in a broad range of applications. State-of-the-art methods in such setups focus on effective identification of faults that usually involve a single component while rejecting noise from limited sources. This article studies the data-based diagnosis of mixed faults coming from multiple components with an emphasis on model robustness against a wide spectrum of external perturbation. A dataset is collected on a rotor and bearing system by varying the levels and types of faults in both the rotor and bearing, which results in 48 machine health conditions. A duplet classifier is developed by combining two 1-D convolutional neural networks (CNNs) that are responsible for the diagnosis of the rotor and bearing faults, respectively. Experimental results show that the proposed classifier can reliably identify the onset and nature of mixed faults. In addition, one-vs-all classifiers are built using the features generated by the developed 1-D CNNs as predictors to recognize previously unlearned fault types. The effectiveness of such classifiers is demonstrated using data collected from four new fault types. Finally, the robustness and ability to reject external perturbation of the duplet classification model are analyzed using kernel density estimation. The code for the proposed classifiers is available at https://github.com/siyuanc2/machine-fault-diag.
Algorithms for current automatic train operation (ATO) focus mainly on reducing the mechanical energy of motion for a single train within an existing timetable. However, the reuse of regenerative energy is another factor that contributes to energy consumption and conservation in multitrain networks. To improve regenerative energy receptivity and energy savings in a bidirectional metro transit network, this study formulated a coordinated train control algorithm that was based on genetic algorithm techniques. The energy saving potential of different station departure time intervals between two opposing trains (synchronization time) was tested. Simulation on the Visual C++ platform demonstrated that the algorithm could provide an optimal train speed profile with better energy performance while also satisfying operational constraints. Different synchronization times have different optimization ratios. This research was another step to facilitate the development of an ATO control algorithm that considers overall energy consumption. Increased knowledge of the influence of synchronization time at stations on energy consumption in regenerative multitrain networks will also aid in the design of more energy-efficient timetables.
Recent proposals for expanded intercity passenger rail service in the United States have included plans for incremental improvements to existing Amtrak service. Improvements to existing services aim to accommodate faster and more frequent passenger train operation, generally on trackage owned and operated by freight railways. There are various projects and approaches to consider when decreasing the running time of passenger trains on a particular corridor. Raising the maximum operating speed can yield different benefits on different sections of the route and conditions on adjacent sections can interact with each other. For instance, the marginal travel time benefit of improving segments of a line from 79 to 110 mph maximum speed is less than the benefit of other improvements to eliminate segments currently restricted to lower speeds. Therefore, to maximize the potential of limited resources, project investments must be selected carefully to improve performance in a cost-effective manner. This paper presents a methodology for optimally selecting projects or establishing program budgets to reduce running time on a passenger rail corridor with consideration of capital, maintenance and operating cost. The proposed project selection model is formulated with Genetic Algorithms (GAs). In the model, a route is divided into sections that can be independently upgraded and the objective function is formulated as minimization of running time along the route. This model can aid in quickly and efficiently developing a strategic plan for improving running time on passenger rail corridors.
A method of detecting atmospheric ducts using a wind profiler radar (WPR) and a radio acoustic sounding system (RASS) is proposed. The method uses the RASS to measure the virtual temperature profile and calculate the Brunt–Väisälä frequency; it also uses the WPR to measure the spectral width of the atmosphere and the atmospheric refractive index structure constant. Then the profile of the atmospheric refractive index gradient and modified refractivity are calculated using virtual temperature, spectral width, and the atmospheric refractive index structure constant. Finally, the height and intensity of the atmospheric duct are calculated to achieve continuous monitoring of the atmospheric duct. To verify the height and intensity of the atmospheric duct, comparison experiments between WPR-RASS and radiosondes were carried out from June 2014 to June 2015 in Dalian, Liaoning Province, China. The results show that the profile of modified refractivity by WPR-RASS has exactly the same trend as the radiosondes, the two methods have a good consistency, and the atmospheric duct value from WPR-RASS is in good agreement with that from radiosondes.
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.