Source-Sink Matching Optimization Based on the Proximal Policy Optimization (PPO) Algorithm

Effective matching of CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emission sources and storage sites is a critical prerequisite for the successful implementation of Carbon Capture, Utilization, and Storage (CCUS) technologies. In this study, we employ the Proximal Policy Optimization (PPO) algorithm to solve the source-sink matching optimization problem, and enhance the traditional PPO algorithm by incorporating techniques such as orthogonal initialization, state normalization, gradient clipping, and learning rate decay, realizing the solution of the source-sink matching optimization problem by unconventional mathematical modeling methods. The results demonstrate that the proposed algorithm not only ensures efficient iteration and convergence speed but also consistently delivers high-quality solutions. This advancement provides new technical support for optimizing the matching of carbon sources and sinks.