TY - GEN
T1 - Emission-Constrained Optimization of Gas Networks: Input-Convex Neural Network Approach
AU - Dvorkin, Vladimir
AU - Chevalier, Samuel
AU - Chatzivasileiadis, Spyros
PY - 2024
Y1 - 2024
N2 - Planning optimization of gas networks under emission constraints prioritizes gas supply with the smallest emission footprint. As this problem includes complex gas flow physical laws, standard optimization solvers cannot guarantee convergence to a feasible solution, especially under strict emission constraints. To address this issue, we develop an input-convex neural network (ICNN) aided optimization routine which incorporates a set of trained ICNNs approximating the gas flow equations with high precision. Numerical tests on the Belgium gas network demonstrate that the ICNN-aided optimization dominates non-convex and relaxation-based solvers, with larger optimality gains pertaining to stricter emission targets.
AB - Planning optimization of gas networks under emission constraints prioritizes gas supply with the smallest emission footprint. As this problem includes complex gas flow physical laws, standard optimization solvers cannot guarantee convergence to a feasible solution, especially under strict emission constraints. To address this issue, we develop an input-convex neural network (ICNN) aided optimization routine which incorporates a set of trained ICNNs approximating the gas flow equations with high precision. Numerical tests on the Belgium gas network demonstrate that the ICNN-aided optimization dominates non-convex and relaxation-based solvers, with larger optimality gains pertaining to stricter emission targets.
U2 - 10.1109/CDC49753.2023.10383948
DO - 10.1109/CDC49753.2023.10383948
M3 - Article in proceedings
T3 - Proceedings of the IEEE Conference on Decision and Control
SP - 1575
EP - 1579
BT - Proceedings of 2023 62nd IEEE Conference on Decision and Control (CDC)
PB - IEEE
T2 - 62<sup>nd</sup> IEEE Conference on Decision and Control
Y2 - 13 December 2023 through 15 December 2023
ER -