Show simple item record

dc.contributor.authorMandal, S.
dc.contributor.authorHegde, G.
dc.contributor.authorGupta, K.G.
dc.date.accessioned2008-07-02T04:53:48Z
dc.date.available2008-07-02T04:53:48Z
dc.date.issued2004
dc.identifier.citationProceedings of National Seminar on Construction Management: Latest Trends and Developments, 26-27 March 04, eds. Pal, S.C.; Vinaykumar, C.H.; Nagaraj, K.; Morankar, D.V.; Jhamnani, B. 323-332p.
dc.identifier.urihttp://drs.nio.org/drs/handle/2264/1135
dc.description.abstractThe safety of an offshore platform depends on the predicting environmental phenomena such as wind, current, wave, seismic loadings and accurate calculation of responses of the structure to these loads and determining the strength of the structure. This paper deals with the prediction of stress resultant deflections of fixed offshore platform to the varying environmental loading conditions using neural networks. The manual estimation of stress resultant to the varying loading conditions involves tedious calculation methods and time consuming. The most difficult task is to represent the problem mathematically, involving all possible forcing functions. With the time dependent nature of the environmental loads, estimation of platform response may not be the actual one. Neural networks have the ability to recognize the hidden pattern and accordingly make prediction. Built in dynamism, data error tolerance, lack of any exogenous data and model free solutions makes them even more attractive. Hence an attempt is made to use neural networks in the estimation of deck responses to varying water depth, wave and wind conditions, which will be very useful for the design of offshore structures. With the availability of advanced computer software, the 3D analysis reveals the response behavior of offshore platforms. With the complexity and random nature of environmental loads, the calculated responses may not be accurate to the extent required at all loading state for practical considerations. The back-propagation neural network (BPN) employed here for prediction of response is to refinement of design procedure to make it more economical. This can be a useful tool in periodic inspection of structure to detect structural damages as a requirement for certification. This experimental work depicts prediction of deck displacement resultant of 4 legged offshore jacket platform in 200 m water depth using neural network. With the data of deck displacements at various loading state, the BPN was first trained and later prediction of the deck displacements is obtained for any loading state
dc.language.isoen
dc.publisherCollege of Civil Engineering, Pune; India
dc.rightsCopyright [2004]. It is tried to respect the rights of the copyright holders to the best of the knowledge. If it is brought to our notice that the rights are violated then the item would be withdrawn.
dc.subjectpiled platforms
dc.subjectfixed platforms
dc.subjectloads (forces)
dc.subjectwave forces
dc.subjectwater depth
dc.subjectapproximation
dc.subjectoffshore structures
dc.subjectwave height
dc.subjectwave period
dc.titleEvaluation of stress resultant of offshore jacket platform using neural network
dc.typeConference Article


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record