Nonlinear problems in control of manufacturing systems

A hybrid mechanistic machine learning approach to model industrial network dynamics for sustainable design of emerging carbon capture and utilization technologies

“…One principal limitation of LCA and MFA methods is the lack of accounting for the nonlinear dynamics between various system components. [15][16][17] Capturing such nonlinear relationship is important for accurate assessment of resource consumption and emissions over time as the system response to changes cannot be linearly extrapolated. While the system dynamics method does account for this dynamic behavior, it relies on causal loop diagrams which are generally unknown for emerging industrial networks.…”