giaccone / cogen_eval

Uncertainty Quantification in the energy management of a CHP system

cogen_eval

Uncertainty Quantification in the energy management of a CHP system

In this repository you will find all the scripts to reproduce the results of the paper:

Giaccone L., Lazzeroni P., Repetto M., “Uncertainty quantification in energy management procedures”, Vol. 9, N.9, September 2020, Article number 1471, pp. 1-10, DOI: 10.3390/electronics9091471

Requirements

Codes included in this repository are written in Python 3, that is the only real requirement. They have been tested with Python 3.7 and 3.8 but also earlier version of Python 3 should work.

Polynomial Chaos Expansion (PCE) simulations are made using this module https://github.com/giaccone/pce that is developed by the same author of this repository. PCE results are validated by Monte Carlo (MC) technique.

This module is deployed through`` the Python Package Index, therefore, it can be easily obtained bu rinning the following command:

pip install pce

Content of the repository

1. cogen_util.py: this file includes two functions that are used by other files. This file is not directely used by the user.
2. CogenEval_PCE_vs_MC.py: this file performs an uncertaninty analysis with PCE and MC and provides the comparison in the form of a plot. User can set uniform or normal distribution of uncertain variables and can decide if the results are saved in npz (numpy) format
3. CogenEval_Sobol_indices.py: this file computes Sobol indices for the case study under analysis. User can set uniform or normal distribution of uncertain variables and decide if the results are saved in npz (numpy) format.
4. plot_input_profiles.py: this file plots the input profile of the thermal energy demand and the electricity cost
5. plot_pce_vs_mc.py: this file plot the results obtained at point 2 (if the user saved them in npz format)
6. plot_sobol.py: this file plots the results obtained at point 3 (if the user saved them in npz format)