HUCP Instances Generator (HIG) is a generator of instances for the Hydro Unit Commitement Problem.
The generator requires some external data about prices from http://www.mercatoelettrico.org/. A bash script is available among the provided files in order to download authomatically and in the right forlder the files with price data (An equivalent batch script for Windows will soon be available). It is sufficient to execute it, however the Users should convert the excel files into semicolon separated CSV files. The necessary spreadsheet is "Prezzi-Prices". However, we remind you to accept the terms of use of http://www.mercatoelettrico.org/. The folder named "Prices" must contain separate CSV files, each one with data separated by semicolons, all in the same folder. There should be one CSV file for each considered year, and each file should be named “Anno YYYY.csv" where YYYY is the year in question, for example, “Anno 2004.csv". Subfolders, zip files, or Excel files are not required.
To facilitate code usage and enhance understanding of the format, this repository contains examples of files in the correct format within the "Prices" folder. However, it's important to note that the provided data is realistic but not real. For actual price data, you must download the files directly from http://www.mercatoelettrico.org/ or use the script mentioned above while following the instructions below.
To install the genearator on any Linux O.S. it is sufficient to run " make ". Then to execute it is sufficient to run " ./HIG " and follow the instructions.
Detailed instructions for Linux O.S.:
1 Decide if realistic data is sufficient or real data are required
1.1 If realistic is sufficient, skip to step 5
1.2 If real data is required, continue with step 2
2 Execute "chmod +x getPrices.sh"
3 Execute "./getPrices.sh" ;
4 Convert the .xls files in semicolon separated .csv files (only the sheet "prezzi-prices" is necessary),
replacing the obtained files with the existing ones, containing only realistic data"
5 Execute "make"
6 Execute "./HIG"
7 Enjoy creating instances
More detailed instructions are available in this pdf file.
The generator requires price data and inflow data. The available data go from 2004 to 2012.
The price data format that the generator needs to generate the instances is a separate csv file for each year that user wants to use called “Anno YYYY.csv" where YYYY is the year in question, for example, “Anno 2004.csv". An example csv (with ";" as separator) is
| Date (YYYYMMDD) | delta_t | PUN |
|---|---|---|
| 20040401 | 1 | 17.45 |
| 20040401 | 2 | 23.55 |
| 20040401 | 3 | 32.65 |
In the table are represented only the first 3 periods that in the case of the Italian electricity market correspond to one hour each. The first column includes the date in year, month, day format, i.e., YYYYMMDD. The second column indicates the reference time (between 1 and 24). Finally, the third column shows the national unit price (PUN) in €/MWh. The already downloadable files contain additional columns containing the market prices of individual zones and neighboring markets. However, these values are not used explicitly by the generator.
The inflow data format that the generator needs to produce the instances is a csv file (with ”;” as separator) composed of three columns, i.e., date, hour, and inflow, respectively. An example is
| Date (YYYYMMDD) | delta_t | Inflow |
|---|---|---|
| 20040401 | 1 | 2.48 |
| 20040401 | 2 | 2.31 |
| 20040401 | 3 | 2.17 |
The date is expressed as year, month, day, i.e., YYYYMMDD. The second column indicates the reference time (between 1 and 24). Finally, the third column shows the inflows expressed in [m^3/s]. Additional rows can be added to the existing file with the condition of entering the data in ascending date order. Data continuity is not required.
The code and data are freely available for non-commercial use. However, if you choose to utilize them, it is necessary to provide proper attribution and reference.
This work was supported by a public grant as part of the Investissement d’avenir project, reference ANR-11-LABX-0056-LMH, LabEx LMH. This research benefited from the support of the FMJH Program PGMO and from the support of EDF.
For any questions, concerns, or inquiries regarding this dataset or repository, please reach out to dimitri[dot]thomopulos[at]unipi[dot]it.