Deep Neural Koopman Operator-based Economic Model Predictive Control of Shipboard Carbon Capture System
Minghao Han, Xunyuan Yin
TL;DR
This work addresses the challenge of safely and economically operating shipboard post-combustion carbon capture by marrying data-driven learning with a convex optimization-based control framework. A Deep Neural Koopman Operator (DNKO) model learns a time-varying, latent representation that predicts future economic costs and key outputs from partial-state measurements, while neural networks adapt Koopman matrices to improve accuracy. Building on this model, a convex Economic Model Predictive Control (EMPC) is formulated to minimize cost subject to hard-output constraints and carbon capture targets, enabling efficient real-time control. In high-fidelity simulations across four operating conditions, the DNKO-EMPC consistently reduces economic cost, sustains high carbon capture rates (up to 87.12%), and satisfies safety constraints with a mean control step time of roughly 0.69 seconds, highlighting practical potential for shipboard PCC deployments.
Abstract
Shipboard carbon capture is a promising solution to help reduce carbon emissions in international shipping. In this work, we propose a data-driven dynamic modeling and economic predictive control approach within the Koopman framework. This integrated modeling and control approach is used to achieve safe and energy-efficient process operation of shipboard post-combustion carbon capture plants. Specifically, we propose a deep neural Koopman operator modeling approach, based on which a Koopman model with time-varying model parameters is established. This Koopman model predicts the overall economic operational cost and key system outputs, based on accessible partial state measurements. By leveraging this learned model, a constrained economic predictive control scheme is developed. Despite time-varying parameters involved in the formulated model, the formulated optimization problem associated with the economic predictive control design is convex, and it can be solved efficiently during online control implementations. Extensive tests are conducted on a high-fidelity simulation environment for shipboard post-combustion carbon capture processes. Four ship operational conditions are taken into account. The results show that the proposed method significantly improves the overall economic operational performance and carbon capture rate. Additionally, the proposed method guarantees safe operation by ensuring that hard constraints on the system outputs are satisfied.