Use this URL to cite or link to this record in EThOS:
Title: Frequency domain fatigue analysis of dynamically sensitive structures
Author: Wang, Ruhuai
ISNI:       0000 0001 3561 1205
Awarding Body: University of London
Current Institution: University College London (University of London)
Date of Award: 1997
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This thesis presents several developments in frequency domain fatigue analysis for dynamically sensitive structures. A sophisticated study is carried out for various factors which affect the prediction of fatigue damage using different spectral fatigue analysis tools. Applications were made to data from wind turbine blade WEG MS1, Howden HWP330 and other data to predict fatigue damage of the structure/component under different loading cases. A new frequency domain based theoretical approach for fatigue prediction has been developed which is applicable for general random, stationary signals. This overcomes the difficulties of assuming Gaussian, narrow band loading or applying correction factors in the existing spectral fatigue analysis tools. Effects of deterministic component (including harmonic and equally spaced spikes) and non-Gaussianality, which are ignored in existing frequency domain fatigue analysis tools, are studied. Extensive computer modeling and Artificial Neural Networks (ANN) have been applied to study the effect of non- Gaussianality. Analysis of WEG, HWP and other data suggested kurtosis as the best description of the degree of the departure from a Gaussian distribution. Several parameters were related to the rainflow cycle range p.d.f. of a non-Gaussian stress signal, these being the root mean square of the signal, the irregularity factor, mean frequency and kurtosis. Among various developed ANN systems, back-propagation, a method of training a neural network to approximate any function, including arbitrarily complex nonlinear functions, was chosen to study the effect of non-Gaussianality. Neural networks were trained to provide a fast, effecient toolbox for practical engineering applications.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Structural engineering