Rectangular steel silos: finite element predictions of filling wall pressures

R.J. Goodey, C. J. Brown, J. M. Rotter

    Research output: Contribution to journalArticlepeer-review

    Abstract

    The pressures exerted on the walls of rectangular planform steel flexible-walled silos by several different stored granular bulk solids are investigated using a validated finite element model that has been used in several previous studies. These pressures and the state of stress in the bulk solid are explored for a range of silo geometries and stored bulk solids. The results show that the horizontal pressure distribution across a silo wall is generally not uniform. This demonstrates that widely used theories may be adequate for stiff concrete silos, are far from suited to flexible-walled steel silos, and the differences can be used to produce much lighter structures. These findings match previously published experimental and analytical results for square planform silos where much larger pressures develop in the corners. The present analyses show that rectangular silos differ from those of square section, in that the mean pressure and degree of pressure variation is different on the two walls. The mechanisms causing these changes are investigated. The results further demonstrate that relatively small changes in the properties of a stored solid can produce significant changes in the pressure magnitudes and patterns, and hence greatly influence the silo structural design. The paper concludes that existing design guidance is seriously deficient and leads to metal silos that are considerably more expensive than is necessary.
    Original languageEnglish
    Pages (from-to)61-69
    JournalEngineering Structures
    Volume132
    Early online date22 Nov 2016
    DOIs
    Publication statusPublished - 1 Feb 2017

    Keywords

    • rectangular steel silos
    • wall pressures
    • finite elements
    • silo design
    • Civil engineering

    Fingerprint

    Dive into the research topics of 'Rectangular steel silos: finite element predictions of filling wall pressures'. Together they form a unique fingerprint.

    Cite this