GENeration

By David St. John. 2015

Three models with different submarine structure were established firstly, and then, flow fields of the three different structures were computed and compared through FLUENT software. As a result, the virtual reality of the flow characteristics of submarines was also researched and presented through some contours in this paper. As known from the result, the flow field speed around the submarine was significantly reduced but the pressure on the submarine was enlarged if the submarine had a podium, mainly due to a certain vortex existing around the podium. Besides, after the adoption of a cruciform tail, a significant improvement was shown in the flow field around the submarine and flow field on the surface, thus indicating the severe impact generated by the submarine tail on the flow field characteristics of the submarine. Therefore, the tail type of the submarine should be designed strictly during the design process of submarine structure. Finally, BP neural network was applied to compute the total resistance coefficient of the submarine and compare it with the calculation result of CFD to obtain a small relative error. It indicated that it was reliable to use BP neural network to predict the total resistance coefficient of the submarine.

Citations in the corpus

• REVERIE: Remote Embodied Visual Referring Expression in Real Indoor Environments (2019, unknown): “Moreover, compared to the task of Referring Expression (RefExp) [14, 22] that selects the desired object from a single image, REVERIE is far more challenging in the sense that the target object is not visible in the initial view and needs to be discovered by actively navigating in the environment.”