Abstract:The balance between accuracy and computational complexity is currently a focal point of research in dynamical system modeling. From the perspective of model reduction, this paper addresses the mode selection strategy in Dynamic Mode Decomposition (DMD) by integrating an embedded fractal theory based on fractal dimension (FD). The existing model selection methods lack interpretability and exhibit arbitrariness in choosing mode dimension truncation levels. To address these issues, this paper analyzes the geometric features of modes for the dimensional characteristics of dynamical systems. By calculating the box counting dimension (BCD) of modes and the correlation dimension (CD) and embedding dimension (ED) of the original dynamical system, it achieves guidance on the importance ranking of modes and the truncation order of modes in DMD. To validate the practicality of this method, it is applied to the reduction applications on the reconstruction of the velocity field of cylinder wake flow and the force field of compressor blades. Theoretical results demonstrate that the proposed selection technique can effectively characterize the primary dynamic features of the original dynamical systems. By employing a loss function to measure the accuracy of the reconstruction models, the computed results show that the overall errors of the reconstruction models are below 5%. These results indicate that this method, based on fractal theory, ensures the model's accuracy and significantly reduces the complexity of subsequent computations, exhibiting strong interpretability and practicality.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to balance accuracy and computational complexity in dynamic system modeling. Specifically, the paper focuses on how to select modes in Dynamic Mode Decomposition (DMD) to reduce the computational burden of high - dimensional systems while maintaining the accuracy of the model. The existing mode selection methods lack interpretability and arbitrariness, especially when choosing the mode dimension truncation level. To solve these problems, the paper proposes a new mode selection strategy by combining the fractal dimension characteristics in fractal theory (such as Box Counting Dimension (BCD) and Correlation Dimension (CD)) to guide the importance ranking and truncation order of modes. This method not only improves the accuracy and interpretability of the model but also significantly reduces the complexity of subsequent calculations. ### Main Contributions 1. **System Fractal Feature Description**: By using the box - counting dimension and the correlation dimension to describe the fractal features of the dynamic system, a new perspective for nonlinear system analysis is provided. Fractal features can better capture the complex structures of modes, thus enabling more effective selection of key modes. 2. **Mode Selection Strategy**: Fractal features provide information about the dynamic characteristics of the original system and can guide the selection of modes that best represent the dynamic behavior of the system. Combining fractal features with DMD mode selection can more accurately capture the dynamic behavior of the system and improve the credibility and prediction ability of the model. ### Application Examples To verify the effectiveness of this method, the paper applies it to the reconstruction of the wake velocity field of a cylinder and the reconstruction of the compressor blade force field. The experimental results show that the proposed mode - selection technique can effectively characterize the main dynamic features of the original dynamic system, and by evaluating the accuracy of the model through the loss function, the results show that the overall error of the reconstructed model is less than 5%. ### Methodology The paper details the fractal dimension theory and the dynamic modal decomposition theory: - **Fractal Dimension Theory**: It includes the definitions and calculation methods of the box - counting dimension and the correlation dimension. - **Dynamic Modal Decomposition Theory**: It introduces the basic principles and calculation steps of DMD, as well as how to extract modes, amplitudes, and eigenvalues from flow - field data. ### Experimental Verification The paper conducts experimental verification through the vortex - field data of the cylinder wake. The experimental data are from the cylinder wake with a Reynolds number of 100, and a total of 150 time - step flow - field snapshots are collected. By performing DMD analysis on these data, the paper shows the eigenvalue distributions and amplitude relationships of different modes, and further explores the fractal characteristics of different modes through fractal analysis. In conclusion, the paper proposes a DMD mode - selection method based on fractal theory, aiming to solve the mode - selection problem in high - dimensional dynamic system modeling and improve the accuracy and interpretability of the model.

Reduced Order Modeling of System by Dynamic Modal Decom-Position with Fractal Dimension Feature Embedding

Data-driven Dimension Reduction for High-Dimensional Random Vibration Systems with Cubic Nonlinearity

Model Reduction via Dynamic Mode Decomposition

A model reduction method for parametric dynamical systems defined on complex geometries

Parametric Dynamic Mode Decomposition for nonlinear parametric dynamical systems

Brief Communication: Dimensionality Reduction in Total Dynamic Mode Decomposition Using A Simple Geometric Method

A Review on Reduced Order Modeling using DMD-Based Methods

A Splitting Approach to Dynamic Mode Decomposition of Nonlinear Systems

A scaled derivative-based DMDc method for modelling multiple-input multiple-output mechanical systems

PF-DMD: Physics-fusion dynamic mode decomposition for accurate and robust forecasting of dynamical systems with imperfect data and physics

Estimation and Control of Fluid Flows Using Sparsity-Promoting Dynamic Mode Decomposition

Reduced-order variational mode decomposition to reveal transient and non-stationary dynamics in fluid flows

Mori-Zwanzig Modal Decomposition

Challenges in dynamic mode decomposition

A wavelet-based dynamic mode decomposition for modeling mechanical systems from partial observations

Dynamic Mode Decomposition and Its Variants

Dynamic Mode Decomposition in Adaptive Mesh Refinement and Coarsening Simulations

A Parametric and Feasibility Study for Data Sampling of the Dynamic Mode Decomposition: Spectral Insights and Further Explorations

A reduced-order model for compressible flows with buffeting condition using higher order dynamic mode decomposition with a mode selection criterion

Data-driven reduced order modeling for parametrized time-dependent flow problems