Feasibility of Predicting Airblast-Induced Ground Motions by the Statistical Analysis of Fourier Transforms Calculated from Measured Nuclear and Hest Time-Histories. Volume I.
Abstract
An investigation into the possibility of applying statistical analysis of the Fourier transforms of airblast-induced ground motion time-histories in order to predict the most probable spectral characteristics of the motions in terms of known independent variables is presented. The various analytical and empirical procedures used in the development of the prediction technique are discussed. A model Fourier transform equation is postulated to fit actual transforms calculated from measured nuclear and HEST test time-histories. Constants in the model transform equation are found so that the postulated spectra is made to fit the actual spectra for each measured time-history. A set of regression equations using peak overpressure, pressure-impulse, depth and seismic velocity as the independent variables are derived. A linear regression analysis is conducted to calculate new constants for the model spectra which have a mean squared error when compared to the constants fitted to each actual spectra. A statistical analysis of the model Fourier transform constants so calculated is conducted to determine the 95 percent confidence limits on the population mean, variance, and regression and correlation coefficients. Equations are derived for the vertical acceleration, velocity and displacement of the free-field airblast-induced ground motions based on the predicted most probable Fourier transform of the motions. Application of the transform and time-history equations is discussed in terms of an example. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1969
- Accession Number
- AD0849496
Entities
People
- Peter C. Calcaterra
- Richard D. Cavanaugh