An Optimization Cost Strategy for Storage-enabled Hydrogen Flow Network Using Monte Carlo Simulation

This article presents an innovative approach to address the optimization and planning of hydrogen network transmission, focusing on minimizing computational and operational costs, including capital, operational, and maintenance expenses. The mathematical models developed for gas flow rate, pipelines, junctions, and storage form the basis for the optimization problem, which aims to reduce costs while satisfying equality, inequality, and binary constraints. To achieve this, we implement a dynamic algorithm incorporating 100 scenarios to account for uncertainty. Unlike conventional successive linear programming methods, our approach solves successive piecewise problems and allows comparisons with other techniques, including stochastic and deterministic methods. Our method significantly reduces computational time (56 iterations) compared to deterministic (92 iterations) and stochastic (77 iterations) methods. The non-convex nature of the model necessitates careful selection of starting points to avoid local optimal solutions, which is addressed by transforming the primal problem into a linear program by fixing the integer variable. The LP problem is then efficiently solved using the Complex Linear Programming Expert (CPLEX) solver, enhanced by Monte Carlo simulations for 100 scenarios, achieving a 39.13% reduction in computational time. In addition to computational efficiency, this approach leads to operational cost savings of 25.02% by optimizing the selection of compressors (42.8571% decreased) and storage facilities. The model’s practicality is validated through realworld simulations on the Belgian gas network, demonstrating its potential in solving large-scale hydrogen network transmission planning and optimization challenges.