Abstract:This paper introduces the concept of abstracted model reduction: a framework to improve the tractability of structure-preserving methods for the complexity reduction of interconnected system models. To effectively reduce high-order, interconnected models, it is usually not sufficient to consider the subsystems separately. Instead, structure-preserving reduction methods should be employed, which consider the interconnected dynamics to select which subsystem dynamics to retain in reduction. However, structure-preserving methods are often not computationally tractable. To overcome this issue, we propose to connect each subsystem model to a low-order abstraction of its environment to reduce it both effectively and efficiently. By means of a high-fidelity structural-dynamics model from the lithography industry, we show, on the one hand, significantly increased accuracy with respect to standard subsystem reduction and, on the other hand, similar accuracy to direct application of expensive structure-preserving methods, while significantly reducing computational cost. Furthermore, we formulate a systematic approach to automatically determine sufficient abstraction and reduction orders to preserve stability and guarantee a given frequency-dependent error specification. We apply this approach to the lithography equipment use case and show that the environment model can indeed be reduced by over 80\% without significant loss in the accuracy of the reduced interconnected model.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: when dealing with model order reduction of complex interconnected systems, how to significantly improve computational efficiency and ensure the accuracy of the reduced - order model while maintaining the system structure and stability. Specifically, the paper proposes a new framework - abstracted model reduction - to address the computational infeasibility or high cost of existing structure - preserving order - reduction methods. ### Problem Background 1. **Complex Interconnected Systems**: Modern engineering systems are usually composed of multiple interconnected systems, such as plant - controller feedback loops, multi - physical systems, networked cyber - physical systems, etc. The high - order models of these systems make dynamic analysis computationally infeasible. 2. **Model Order - Reduction Requirement**: In order to make the analysis and simulation of high - order models feasible, it is necessary to approximate high - order models with low - order substitute models. However, directly reducing the order of the entire interconnected system often leads to excessive computational costs or inaccurate results. 3. **Limitations of Structure - Preserving Order - Reduction Methods**: Although existing structure - preserving order - reduction methods can preserve the structure and stability of interconnected systems, when the system is very complex, the computational cost of these methods is very high, or even infeasible. ### Solutions Proposed in the Paper 1. **Abstracted Model Reduction Framework**: The paper introduces the concept of "abstracted model reduction", effectively reducing computational costs by using low - order environmental models of each subsystem to replace their complete high - order environmental models. The specific steps are as follows: - **Environmental Abstraction**: Abstract the high - order environmental model \( E(s) \) into a low - order model \( \hat{E}(s) \). - **Enhancement**: Add the input - output information of the subsystem to the enhanced environment to form \( \hat{F}(s) \). - **Order - Reduction**: Use the structure - preserving order - reduction method to reduce the order of the enhanced system \( Fl(\hat{F}, \Sigma) \) to obtain \( Fl(\hat{F}, \hat{\Sigma}) \). - **Replacement**: Replace \( \hat{E}(s) \) with the original environmental model \( E(s) \) to obtain the final reduced - order interconnected system \( Fl(E, \hat{\Sigma}) \). 2. **Error Analysis and Stability Assurance**: The paper also proposes a systematic error analysis method to ensure that the reduced - order interconnected system not only has lower computational costs but also can meet the given frequency - dependent error specifications and ensure stability. ### Experimental Verification The paper verifies the effectiveness of the proposed method through a practical case of the structural dynamics model of lithography equipment in the semiconductor industry. The results show that using the abstracted model reduction method can maintain high precision while significantly reducing computational costs. ### Summary The main contribution of the paper is to provide a new framework that can efficiently perform model order - reduction while maintaining the structure and stability of interconnected systems. This provides a powerful tool for the modeling and analysis of complex engineering systems, especially in the case of limited computational resources.

Abstracted Model Reduction: A General Framework for Efficient Interconnected System Reduction

Structure based model reduction in real-time optimization

Reduced-order Modeling of Modular, Position-dependent Systems with Translating Interfaces

Passivity-Preserving, Balancing-Based Model Reduction for Interconnected Systems

Model Reduction for Large Scale Systems

Power System Dynamic Model Reduction Based on Extended Krylov Subspace Method

Structure-preserving Model Reduction for Nonlinear Port-Hamiltonian Systems

Stability-Preserving Model Reduction of Networked Lur'e Systems

Identification of dominant subspaces for model reduction of structured parametric systems

Nonlinear Model Order Reduction via Lifting Transformations and Proper Orthogonal Decomposition

Structured Model Order Reduction of System-Level Power Delivery Networks

Adjacency-based, Non-intrusive Reduced-order Modeling for Fluid-Structure Interactions

A combined model reduction algorithm for controlled biochemical systems

Structured Projection-Based Model Reduction with Application to Stochastic Biochemical Networks

A Study of Model-Order Reduction Techniques for Verification

Model reduction on manifolds: A differential geometric framework

A nonintrusive distributed reduced‐order modeling framework for nonlinear structural mechanics—Application to elastoviscoplastic computations

A Unified Krylov Projection Framework For Structure-Preserving Model Reduction

Exact hierarchical reductions of dynamical models via linear transformations

Discrete Abstraction Of Multiaffine Systems

Electromagnetic Model Order Reduction for System-Level Modeling