A number of catastrophic failures of steel conical tanks occurred during the past few decades in various locations around the globe. Previous investigations attributed the reason of collapse to inadequate thickness of the steel vessel especially at the bottom part. It was shown in a previous study that welding rectangular shaped longitudinal stiffeners to the bottom part would enhance the buckling capacity of existing tanks and improve the design of new ones. Due to the lack of design guidance of such type of stiffened shells, a numerical tool that uses the state-of-the-art knowledge is required in order to achieve the optimum shell thickness, and optimum configuration and cross section of stiffeners. In the current study an optimum design technique of stiffened liquid-filled steel conical tanks subjected to global and local buckling constraints is developed. The proposed technique is based on a numerical tool that couples a non-linear finite element model developed in-house and an evolutionary optimization based on real coded genetic algorithm. The design variables considered in the current study are the shell thicknesses, the geometry of the steel vessel as well as the dimensions and number of stiffeners. The developed numerical tool is capable of selecting the set of design variables that leads to optimum safe design. Both case of retrofitting an existing tank, and case of designing a new stiffened tank are considered in the current investigation. Finally, comparisons and case studies are conducted to prove the adequacy of the proposed technique.
展开▼
