Uploaded by PSUwind on 23.07.2012

Transcript:

In this video I'm going to show you how to develop a histogram

using a set of wind data. So you can see

over here, the sample number which we won't necessarily be using

the hour, so this is a time record it's

a continuous numer. And the wind speed in meters per

second as well as miles per hour. We're not going to look at the

miles per hour. First I'm going to point out, if you go to the data

tab in excel and on the right hand side

if you don't see data analysis and solver

you may need to go and add these in. You go to file,

options, go to add-ins, and

manage excel add-ins, go, and you can select

these here. So the analysis tool pack as well as the solver add-in

are things that you might find useful for this class.

So next I'm going to define the range of

bins that I want to explore for my histogram.

So I've just developed a column which is just adding

1 to each progressive row down to 25

so these are meters per second. So I'm going to go to data

analysis now and select histogram

and the imput range is going

to be my meters per second column

the bin range is this column

that I just developed

and...

it's going to pop up in a new worksheet.

So it's created sheet 4 here. So now I've got

the frequency

that the wind blows within each bin. Now, what I really want is

to turn this into what will be similar to the Weibull distribution

is to normalize it by the number of observations. So I'm just

going to sum how many observations there are

in this column and there's 4, 464.

This was a month's worth of data, I should

mention. And then I'm just going to just take each of these and

divide them by the number of observations

to get a percentage of the time rather than a number of hours.

And actually this was not hourly data so it wasn't even

a number of hours. Okay, so now if I plot,

these 2 columns

should look something like a Weibull curve.

Something like a Weibull curve...

and similarly we can plot a

Weibull distribution on top of that. And actually I'm going to

change this from a

line chart to a bar chart.

There, that's a little bit better.

using a set of wind data. So you can see

over here, the sample number which we won't necessarily be using

the hour, so this is a time record it's

a continuous numer. And the wind speed in meters per

second as well as miles per hour. We're not going to look at the

miles per hour. First I'm going to point out, if you go to the data

tab in excel and on the right hand side

if you don't see data analysis and solver

you may need to go and add these in. You go to file,

options, go to add-ins, and

manage excel add-ins, go, and you can select

these here. So the analysis tool pack as well as the solver add-in

are things that you might find useful for this class.

So next I'm going to define the range of

bins that I want to explore for my histogram.

So I've just developed a column which is just adding

1 to each progressive row down to 25

so these are meters per second. So I'm going to go to data

analysis now and select histogram

and the imput range is going

to be my meters per second column

the bin range is this column

that I just developed

and...

it's going to pop up in a new worksheet.

So it's created sheet 4 here. So now I've got

the frequency

that the wind blows within each bin. Now, what I really want is

to turn this into what will be similar to the Weibull distribution

is to normalize it by the number of observations. So I'm just

going to sum how many observations there are

in this column and there's 4, 464.

This was a month's worth of data, I should

mention. And then I'm just going to just take each of these and

divide them by the number of observations

to get a percentage of the time rather than a number of hours.

And actually this was not hourly data so it wasn't even

a number of hours. Okay, so now if I plot,

these 2 columns

should look something like a Weibull curve.

Something like a Weibull curve...

and similarly we can plot a

Weibull distribution on top of that. And actually I'm going to

change this from a

line chart to a bar chart.

There, that's a little bit better.