Bagging, optimized dynamic mode decomposition (BOP-DMD) for robust, stable forecasting with spatial and temporal uncertainty-quantification

Abstract

Dynamic mode decomposition (DMD) provides a regression framework for adaptively learning a best-fit linear dynamics model over snapshots of temporal, or spatio-temporal, data. A diversity of regression techniques have been developed for producing the linear model approximation whose solutions are exponentials in time. For spatio-temporal data, DMD provides low-rank and interpretable models in the form of dominant modal structures along with their exponential/oscillatory behavior in time. The majority of DMD algorithms, however, are prone to bias errors from noisy measurements of the dynamics, leading to poor model fits and unstable forecasting capabilities. The optimized DMD algorithm minimizes the model bias with a variable projection optimization, thus leading to stabilized forecasting capabilities. Here, the optimized DMD algorithm is improved by using statistical bagging methods whereby a single set of snapshots is used to produce an ensemble of optimized DMD models. The outputs of these models are averaged to produce a bagging, optimized dynamic mode decomposition (BOP-DMD). BOP-DMD not only improves performance, it also robustifies the model and provides both spatial and temporal uncertainty quantification (UQ). Thus unlike currently available DMD algorithms, BOP-DMD provides a stable and robust model for probabilistic, or Bayesian forecasting with comprehensive UQ metrics.

arXiv comment12 pages, 8 figures, 2 algorithms

References in corpus (14)

• On Dynamic Mode Decomposition: Theory and Applications
• Spectral proper orthogonal decomposition and its relationship to dynamic mode decomposition and resolvent analysis
• Sparsity-promoting dynamic mode decomposition
• Chaos as an Intermittently Forced Linear System
• Extracting spatial-temporal coherent patterns in large-scale neural recordings using dynamic mode decomposition
• Ergodic theory, Dynamic Mode Decomposition and Computation of Spectral Properties of the Koopman operator
• De-Biasing the Dynamic Mode Decomposition for Applied Koopman Spectral Analysis
• Characterizing and correcting for the effect of sensor noise in the dynamic mode decomposition
• Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
• Randomized Matrix Decompositions using R
• Subspace dynamic mode decomposition for stochastic Koopman analysis
• Characterizing Magnetized Plasmas with Dynamic Mode Decomposition
• Dynamic Mode Decomposition for Plasma Diagnostics and Validation
• Structured Time-Delay Models for Dynamical Systems with Connections to Frenet-Serret Frame

Cited by in corpus (7)

• Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control
• A Dynamic Mode Decomposition Extension for the Forecasting of Parametric Dynamical Systems
• Variational Mode Decomposition-Based Nonstationary Coherent Structure Analysis for Spatiotemporal Data
• A case study of spatiotemporal forecasting techniques for weather forecasting
• Transition to chaos and modal structure of magnetized Taylor-Couette flow
• A Comparison of Parametric Dynamic Mode Decomposition Algorithms for Thermal-Hydraulics Applications
• Fast Subspace Fluid Simulation with a Temporally-Aware Basis