Solar data analysis

General measurement method of an average sunlight power at a given location. The method is used to improve low-energy houses built by Dworek Polski.

Data source: ftp://ftp-cdc.dwd.de/pub/CDC/observations_germany/climate/hourly/solar

Source code: https://github.com/marycha/solar-data-analysis

Sample input and output data for Potsdam (Berlin) station is kept together with the source code.

Program structure

• run.R - invokes main source code
• solar.R - main source code
• test.R - tests code
• tests/* - sample inputs and outputs used by tests
• database/* - input directory with zip files downloaded from ftp server
• output/* - output directory
• unzip/* - intermediate input directory (once filled with unzipped files, we can reuse it and skip unzip phase)

To run the program:

1. Install R: https://www.r-project.org/

2. Clone or download the code from GitHub repository to your working directory .

3. Save zip files from datasource to ./database (or use sample Potsdam data)

4. Open R console and set working directory (it should contain solar.R, run.R and test.R scripts)

    setwd("path/to/your/working/directory")
   

5. Install required packages (first run only)

    install.packages("reshape")
    install.packages("ggplot2")
    install.packages("lubridate")
   

6. Run tests

    source("test.R")
   

7. Run the code

    source("run.R")
   

8. Your results will be in ./output directory