Emission-Constrained Optimization of Gas Networks: Input-Convex Neural Network Approach

# Emission-Aware Optimization of Gas Networks   This repository collects the Belgium gas network dataset, details on the training procedure, and codes to replicate the results reported in the following paper:   *Emission-Constrained Optimization of Gas Networks: Input-Convex Neural Network Approach *   accepted for presentation at the 62nd IEEE Conference on Decision and Control, Dec. 13-15, 2023, Singapore.   Materials are released with the Attribution 4.0 International (CC BY 4.0) license.   The repository contains two folders: * ```operation_planning``` folder containing data and codes for neural network training and operation planning optimization  * ```long_term_planning``` folder containing data and codes for neural network training and long-term planning optimization    The models are implemented in ```Julia-1.6``` Language, using ```JuMP.jl``` using ```Flux.jl``` library for machine learning and JuMP.jl library for mathematical programming. Before running the code, make sure to activate the virtual environment from ```Project.toml``` files stored in each folder, e.g., by running  ``` julia> ] (@v1.6) pkg> activate . (operation_planning) pkg> instantiate ``` For experiment settings, refer to ```exp_settings``` dictionary in file ```main.jl```. For network data, refer to ```.../data/case_BE```.