Tutorials

Welcome to the brightwind python library tutorial page. The brightwind library makes analysing wind data easy, fast and repeatable. Here you will learn how to install the Brightwind library, get started with using the Jupyter Notebooks and carry out a range of useful functions.

Prerequisites

Before starting these tutorials, ideally the user should have a background in wind data analysis or wind resource assessment. No prior knowledge of the python programming language is necessary, but is preferable. EdX have a nice introductory course to python and pandas dataframes which will give you a good grounding of some basic python concepts for data science aswell as pandas dataframes which are used extensively by the library.

Data

The data used in these tutorials can be downloaded from Dropbox here along with the actual tutorials themselves.

Tutorial 1: Getting started on windows 10

This tutorial shows you how to download and install the Anaconda python distribution, install the brightwind library and how to get started with the brightwind library in Jupyter notebooks.

Tutorial 2: Introduction to the brightwind library

This tutorial shows you how get started with the brightwind library, load some csv data using the load_csv() from the brightwind demo datasets and use the monthly_means() function to plot the monthly means.

Tutorial 3: How to load data into brightwind

This tutorial shows you how to use the library to load in files in different formats. The user will get familiar with loading csv files using the load_csv() function, Campbell Scientific data using the load_campbell_scientific()function, data in excel using the load_excel() function and data exported from windographer using the load_windographer_txt().

Tutorial 4: How to get some useful stats

This tutorial shows you how to get some useful statistics from your wind data. Here we use the time_continuity_gaps() function to find any time gaps in the data, the basic_stats() function to get some useful stats, the coverage() function to get the monthly or annual coverage of specific instruments and finally the momm() function to get the mean of monthly means.

Tutorial 5: How to apply a cleaning file

In this tutorial, you will learn how to apply cleaning to a raw dataset by reading in a structured cleaning file and using the apply_cleaning() function. You will also learn how to apply cleaning from an windographer flag file exported from windographer using the apply_cleaning_windographer() function.

Tutorial 6: How to get some useful plots

This tutorial shows you how to get useful plots for your data. The plot_timeseries() allows you to plot a time series of your data for any specific period. The plot_scatter() function allows you to plot a scatter plot of any two variables. The plot_scatter_wspd() and plot_scatter_wdir() functions plot a scatter plot of two anemometers or two wind vanes respectively and includes a reference line. The sector_ratio() plot shows the ratio of two anemometers versus wind direction in a polar plot.

Tutorial 7: How to transform your data

In this tutorial you will learn how to transform your data in some useful ways. You will learn: - how the scale_wind_speed() function is used to scale a wind speed timeseries dataset - how the average_data_by_period() function can be used to average the data for a specific period - how to adjust the slope and offset of a timeseries using the adjust_slope_offset() - how to apply a direction offset to a wind vane using the offset_wind_direction() - how the offset_timestamps() function can be used to correct a timestamp offset - how to selectively average two anemometers to remove mast shadow using the selective_avg() function.

Tutorial 8: How to plot a wind speed frequency distribution and a distribution of anything!

This tutorial will introduce you to plotting the wind speed frequency distribution and how to plot the distribution of any two variables using the dist(), dist_of_wind_speed() and freq_distribution() functions.

Tutorial 9: How to calculate air density

This tutorial shows you how to use the calc_air_density() function to calculate the air density from your site data, and how to scale it to a specific height based on a defined lapse rate.

Tutorial 10: Plotting and understanding turbulence

This tutorial shows you how to get insights into your site data by plotting a variety of turbulence plots. The TI.by_sector() function shows the turbulence intensity by sector on a polar plot to examine both mast and local affects on instrumentation. The TI.by_speed() function plots the turbulence intensity by windspeed and shows the data versus the IEC class standards. The TI.by_12x24() shows a 12 month by 24 hour matrix of turbulence to see how turbulence varies diurnally throughout the year.

Tutorial 11: How to calculate shear

In this tutorial, you will learn how to calculate the shear in a variety of different ways and then scale it to hub height. The Shear.Average() function is used to calculate the average shear using either the power law or log law methods. The Shear.BySector() calculates the shear by direction sector, Shear.TimeSeries() calculates the shear alpha or roughness for each timestamp in the data to be used in other functions and Shear.TimeOfDay() function creates line or heat map plots showing the 12 month by 24 hour matrix of shear, or just the 24 hour, to see how shear varies diurnally throughout the year. All these functions have a method to .apply the shear calculated to a new timeseries. There is also a Shear.scale() function which can apply a user-defined shear value (alpha or roughness) to a wind speed timeseries.

Tutorial 12: Becoming a correlation wizard!

This tutorial shows you the true power of the brightwind library by allowing you to run multiple correlations for different time periods and reference datasets. We will use the OrdinaryLeastSquares() function to carry out a ordinary least squares linear regression correlation. We will then show how this can be run in a loop for various reference datasets and time periods. We will also touch on other correlation methods in the library such as OrthogonalLeastSquares(), MultipleLinearRegression(), Speedsort(), SVR() and the SimpleSpeedRatio().

Tutorial 13: Exporting a tab file and saving your data

This tutorial shows you how to generate a frequency table and export a tab file using the export_tab_file() function for use in other packages such as WASP or Openwind. Additionally you will learn how to use the export_csv() function to export your time series data.

Tutorial 14: Tips on working with DataFrames

The brightwind library uses DataFrames from the Pandas package to work with wind data. There are many ways to work with DataFrames and so it can get a little confusing. This is a simple tutorial to show a few examples of how to select rows or columns of a DataFrame.