Response Prediction and Detection in Non-linear Clamped Panels

Aircraft panels may often be subjected to high levels ofacoustic pressure loading in flight that may lead to fatigue damage. In some cases, the in-plane restraint provided by the panel boundaries will introduce a geometric stiffening non-linearity when the panel deforms. This Thesis is split into tw...

Full description

Bibliographic Details
Main Author: Acton, Matthew Nicholas Frederick
Published: University of Manchester 2008
Subjects:
Online Access:http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.504772
id ndltd-bl.uk-oai-ethos.bl.uk-504772
record_format oai_dc
spelling ndltd-bl.uk-oai-ethos.bl.uk-5047722015-03-20T03:48:59ZResponse Prediction and Detection in Non-linear Clamped PanelsActon, Matthew Nicholas Frederick2008Aircraft panels may often be subjected to high levels ofacoustic pressure loading in flight that may lead to fatigue damage. In some cases, the in-plane restraint provided by the panel boundaries will introduce a geometric stiffening non-linearity when the panel deforms. This Thesis is split into two main areas in the field of non-linear dynamics. Firstly the prediction ofthe non-linear response ofa structure to acoustic excitation for use in sonic fatigue life calculations is examined. The second area of consideration is the detection ofthe presence of non-linearities in a structure when there may more than one present. The Non-Linear Modal method (NLMOD) previously developed at the University of Manchester, allows the prediction ofthe kinematic and stress response of such panels via a reduced order approach that converts Finite Element results into a model based in linear modal space, with additional terms included to represent the non-linear behavior. In this Thesis, the development ofan experimental clamped panel structure is presented, together with several Finite Element models ofthe same structure. The panel is tested for its modal and static characteristics and exposed to fairly high-level acoustic excitation in a Progressive Wave Tube (PWT). The resulting non-linear strain responses are compared to those predicted using the identified non-linear modal model. Discrepancies are initially found between test and prediction. Reasons for the discrepancies are discussed and the model modified accordingly. Good final agreement with test results is found and a number of areas for further investigation are highlighted. The second part of the Thesis develops a novel, simple method for detecting the presence of non-linear stiffness and non-linear damping, when both may be present, using higher order statistics ofthe time domain response ofa structure. The method is developed using simulation of a single degree offreedom system and is then validated using a clamped panel. In order to ensure the presence of non-linear damping, a novel discrete, non-linear damper is developed making use of a passive shaker with the terminals shorted.620.110287University of Manchesterhttp://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.504772Electronic Thesis or Dissertation
collection NDLTD
sources NDLTD
topic 620.110287
spellingShingle 620.110287
Acton, Matthew Nicholas Frederick
Response Prediction and Detection in Non-linear Clamped Panels
description Aircraft panels may often be subjected to high levels ofacoustic pressure loading in flight that may lead to fatigue damage. In some cases, the in-plane restraint provided by the panel boundaries will introduce a geometric stiffening non-linearity when the panel deforms. This Thesis is split into two main areas in the field of non-linear dynamics. Firstly the prediction ofthe non-linear response ofa structure to acoustic excitation for use in sonic fatigue life calculations is examined. The second area of consideration is the detection ofthe presence of non-linearities in a structure when there may more than one present. The Non-Linear Modal method (NLMOD) previously developed at the University of Manchester, allows the prediction ofthe kinematic and stress response of such panels via a reduced order approach that converts Finite Element results into a model based in linear modal space, with additional terms included to represent the non-linear behavior. In this Thesis, the development ofan experimental clamped panel structure is presented, together with several Finite Element models ofthe same structure. The panel is tested for its modal and static characteristics and exposed to fairly high-level acoustic excitation in a Progressive Wave Tube (PWT). The resulting non-linear strain responses are compared to those predicted using the identified non-linear modal model. Discrepancies are initially found between test and prediction. Reasons for the discrepancies are discussed and the model modified accordingly. Good final agreement with test results is found and a number of areas for further investigation are highlighted. The second part of the Thesis develops a novel, simple method for detecting the presence of non-linear stiffness and non-linear damping, when both may be present, using higher order statistics ofthe time domain response ofa structure. The method is developed using simulation of a single degree offreedom system and is then validated using a clamped panel. In order to ensure the presence of non-linear damping, a novel discrete, non-linear damper is developed making use of a passive shaker with the terminals shorted.
author Acton, Matthew Nicholas Frederick
author_facet Acton, Matthew Nicholas Frederick
author_sort Acton, Matthew Nicholas Frederick
title Response Prediction and Detection in Non-linear Clamped Panels
title_short Response Prediction and Detection in Non-linear Clamped Panels
title_full Response Prediction and Detection in Non-linear Clamped Panels
title_fullStr Response Prediction and Detection in Non-linear Clamped Panels
title_full_unstemmed Response Prediction and Detection in Non-linear Clamped Panels
title_sort response prediction and detection in non-linear clamped panels
publisher University of Manchester
publishDate 2008
url http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.504772
work_keys_str_mv AT actonmatthewnicholasfrederick responsepredictionanddetectioninnonlinearclampedpanels
_version_ 1716782372326735872