Pseudo-Time Domain Gusting Wind Sample Clauses

Pseudo-Time Domain Gusting Wind. As shown in Figure 4.7, the hourly maximum gusts at Ketchikan are, on average, 152% of the one-minute average wind speed. The standard error of the regression fit shown in Figure 4.7 is 4.6. Ketchikan Wind Speed Observations Number of Obersvations: 20,440 Regression Coefficient to Linear Fit: 0.78 Gust Wind Speed = 1.52 x One-Minute Wind Speed 80 70 60 Gust Wind Speed (knots) 50 40 30 20 10 0 25 30 35 40 45 50 One-Minute Average Wind Speed (knots) FIGURE 4.7 CORRELATION BETWEEN PEAK GUSTS AND ONE-MINUTE AVERAGE WIND SPEED AT KETCHIKAN AIRPORT Thus, given a one-minute average wind speed, U (knots), the hourly maximum gust is estimated as: hourly maximum gust amplitude = 0.52 U + N(0,4.6) where: the gust amplitude is in knots U is the one-minute average wind speed (knots) N(0,4.6) is a Monte Carlo random sample from the Normal distribution2 with zero mean and standard deviation of 4.6 2 Ang and Tang, Vol. 2, (Reference 3), pp. 284-285, provide a procedure for generating normally distributed Monte Carlo sample variables. This is the hourly maximum gust. For simulation, we need the sample population of wind gusts given a one- minute average wind speed. This is taken to be: gust amplitude = Uniform(0, 1) × 0.52 U + N(0,4.6) where: Uniform(0, 1) uniform distribution between 0 and 1 The duration (persistence) of each gust is modeled using a uniform distribution between 5 seconds and 60 seconds, rounded to the nearest 5 seconds. The gust direction is also a random variable. The direction of the gust is a random variable selected from a uniform distribution extending between 0 degrees and 360 degrees. The instantaneous wind realization is the vector sum of the one-minute average wind vector and gust vector.