Abstract:A machine-learning-based framework for modeling the error introduced by surrogate models of parameterized dynamical systems is proposed. The framework entails the use of high-dimensional regression techniques (e.g., random forests, LASSO) to map a large set of inexpensively computed `error indicators' (i.e., features) produced by the surrogate model at a given time instance to a prediction of the surrogate-model error in a quantity of interest (QoI). This eliminates the need for the user to hand-select a small number of informative features. The methodology requires a training set of parameter instances at which the time-dependent surrogate-model error is computed by simulating both the high-fidelity and surrogate models. Using these training data, the method first determines regression-model locality (via classification or clustering), and subsequently constructs a `local' regression model to predict the time-instantaneous error within each identified region of feature space. We consider two uses for the resulting error model: (1) as a correction to the surrogate-model QoI prediction at each time instance, and (2) as a way to statistically model arbitrary functions of the time-dependent surrogate-model error (e.g., time-integrated errors). We apply the proposed framework to model errors in reduced-order models of nonlinear oil--water subsurface flow simulations. The reduced-order models used in this work entail application of trajectory piecewise linearization with proper orthogonal decomposition. When the first use of the method is considered, numerical experiments demonstrate consistent improvement in accuracy in the time-instantaneous QoI prediction relative to the original surrogate model, across a large number of test cases. When the second use is considered, results show that the proposed method provides accurate statistical predictions of the time- and well-averaged errors.

Dynamic Error Modeling Based on Time Series and Modern Spectrum Analysis

Dynamic Error Spectrum for Estimation Performance Evaluation: a Case Study on Interacting Multiple Model Algorithm

Partial Source Separation from Unknown Correlation Mixture for Eliminating Unknown Periodic Disturbances from Random Measured Signals

Dynamic Linear Models-Based Time Series Decomposition and Its Application on Bearing Fault Diagnosis

Dynamic State Estimation of Power Systems Considering Maximum Correlation Entropy and Quadratic Function

HiFNet: rethinking time series forecasting models from a perspective of error power spectrum

A Data-Driven Fault Detection Framework Using Mahalanobis Distance Based Dynamic Time Warping

Nonlinear Recursive Minimum Model Error Estimation

Volumetric Error Modeling And Analysis For Cnc Machine Tool Based On Multi-Body System

Dynamic State Estimation Method for Generator Considering Measurement of Bad Data

Fourier Transform Based Dynamic Error Modeling Method for Ultra-Precision Machine Tool

Spatio-Temporal Spectral Analysis for General Array Error Models

Source Separation Based On Sparse Representation And Its Application On Machinery Fault Detection

A New Fault Diagnosis Method Based on Adaptive Spectrum Mode Extraction

Dynamic Transmission Error Analysis for CNC Machine Tools under Variable Speed Condition

Q Estimation Via the Discriminant Method Based on Error Modeling

Error modeling for surrogates of dynamical systems using machine learning

Research on intelligent analysis device for dynamic error of electric energy

A New Metric for Target Tracking Performance Evaluation of Nonlinear Systems

Symplectic Geometry Spectrum Analysis Of Nonlinear Time Series

Fault Tracking Algorithm in Complex Electronic System