Optimisation of rotorcraft fuel tank for crashworthiness based on a neural network. Issue 3 (3rd May 2016)

• Record Type: Journal Article
• Authors: Kim, Hyun-Gi; Kim, Sung Chan; Kim, Sung Joon
• Is Part Of: International journal of crashworthiness. Volume 21:Issue 3(2016)
• Page Start: 242
• Page End: 251
• Publication Date: 2016-05-03
• Subjects: Fuel tank -- crashworthiness -- crash impact test -- US military specification -- artificial neural network -- simulated annealing; Automobiles -- Crashworthiness -- Periodicals; Airplanes -- Crashworthiness -- Periodicals; 629.2042
• DOI: 10.1080/13588265.2016.1165447
• Languages: English
• ISSNs: 1358-8265

Abstract

ABSTRACT: Crashworthy fuel tank has been widely implemented among rotorcraft, and they have served a valuable contribution to improving the survivability of crews and passengers. From the early stages of military rotorcraft history, the US Army has developed and implemented a detailed military specification documenting the unique crashworthiness requirements for rotorcraft fuel tank with the aim of reducing the high incidence of fatalities due to post-crash fires. International manufacturers have followed this information to develop their own fuel tank, and have reflected the results of crash impact tests in trial-and-error design and manufacturing processes. Since the crash impact test itself requires lengthy preparation together with costly fuel cell specimens, a series of numerical simulations of the crash impact test with digital mock-ups is necessary, even at the early design stage, in order to minimise trial-and-error testing with full-scale fuel tank. In this study, a number of numerical simulations on fuel cell crash impact tests are performed with the crash simulation software, ANSYS/Autodyn. The resulting equivalent stresses are further analysed to evaluate a number of appropriate design parameters and the artificial neural network and simulated annealing method are simultaneously implemented to optimise the crashworthy performance of fuel tank.