Civil Engineering Reference
In-Depth Information
The physical characteristics of a stochastic process are described by its statistical
properties. If it is the cause of another process, this will also be a stochastic process. I.e.
if a physical event may mathematically be described by certain laws of nature, a
stochastic input will provide a stochastic output. Thus, statistics constitute a
mathematical description that provides the necessary parameters for numerical
predictions of the random variables that are the cause and effects of physical events. The
instantaneous wind velocity pressure (see Eq. 1.1) at a particular time and position in
space is such a stochastic process. This implies that an attempt to predict its value at a
certain position and time can only be performed in a statistical sense. An observed set of
records can not precisely be repeated, but it will follow a certain pattern that may only be
mathematically represented by statistics.
Since wind in our built environment above ground level is omnipresent, it is
necessary to distinguish between short and long term statistics, where the short term
random outcome are time domain representatives for the conditions within a certain
weather situation, e.g. the period of a low pressure passing, while the long term
conditions are ensemble representatives extracted from a large set of individual short
term conditions. For a meaningful use in structural engineering it is a requirement that
the short term wind statistics are stationary and homogeneous. Thus, it represents a
certain time-space-window that is short and small enough to render sufficiently constant
statistical properties. The space window is usually no problem, as the weather conditions
surrounding most civil engineering structures may be considered homogeneous enough,
unless the terrain surrounding the structure has an unusually strong influence on the
immediate wind environment that cannot be ignored in the calculations of wind load
effects. The time window is often set at a period of
T
= 10 minutes.
Fig. 1.2
Short term stationary random process
