Study On Statistical Characteristic Of Ground Of Squareroot Of Mean Squares

As the intensity index of ground motion, PGA (peak ground acceleration) is often employed and analyzed. If ground motion is treated as a random process, PGA is not a determinist value and involves randomness, and then its appearance is associated a certainty probability. So PGA does not reflect the destructive effect of the whole earthquake ground motion time history. From the view points stochastic process, PGA only describe the local characteristic of ground motion, and it is mainly dependent of high frequency vibration component, so it can not represent the destructive effect of earthquake ground motion with broad band frequency contents. Beside PGA, being as statistical characteristics of the random earthquake ground motion, the instant mean square root of sum of squares (SRSTMS) in certain time window, correlation function and power spectral are more reasonable parameters. Based on the data banks of strong motion acceleration in USA and in Europe,the attenuation pattern of peak horizontal and vertical acceleration and square root of mean square in short time interval is studied by regressive analysis for the practical seismic records. It is observed that the variance of the square root of mean squares in short time interval is less than that of PGA. Hence we propose using the maximum value of square root of mean squares in short time to describe the intensity of the ground motion acceleration and to study the attenuation characteristics which can be used in seismic risk analysis to replace that of PGA.Envelope function not only reflects the non-stationary characteristics of earthquake ground motion but also includes the influence of duration, which plays important role for the structures' inelastic response to earthquake excitation. The artificial seismic waves are usually expressed as the product of the strength non-stationary function or envelope function f(t) and a stationary random process. So studying of the envelope function of the practical ground motion acceleration records is very important.In the thesis, the time history of square root of mean squares of acceleration in short time period is fitted by Iyengara's envelope which is a certain exponential type function and used by many researches. It is observed that using the time variable curves of square root of sum of squares in short time interval enables to manifest the intensity characteristics of the ground motion's nonstationarity, and it may displace the normally used peak envelope function. Study on the mean square root of sum of squares in short time interval of acceleration by statistics method, the thesis offers an evaluative value parameter to describe the intensity attenuation characteristics to fit the use of seismic hazard analysis.