A Systematic Particle Filter for Estimating Time-Varying Parameters in Advection-Diffusion Equations with Source Terms

TL;DR

This paper introduces a systematic two-phase particle filter method for estimating and quantifying uncertainty in time-varying parameters within advection-diffusion PDEs, using limited noisy observations to improve model accuracy.

Contribution

It presents a novel offline-online numerical procedure for estimating time-varying parameters in PDEs, specifically addressing source terms in advection-diffusion models, with demonstrated effectiveness.

Findings

Successfully tracks unknown time-varying parameters from noisy data.

Estimates variance parameters affecting parameter evolution.

Demonstrates effectiveness on one-dimensional test problems.

Abstract

Many real-world systems modeled using partial differential equations (PDEs) involve unknown parameters that must be estimated from limited, noisy system observations. While typically assumed to be constants, some of these unobserved parameters may vary with time. This work proposes a two-phase, offline-online numerical procedure for systematically estimating and quantifying uncertainty in time-varying parameters (TVPs) in time-dependent PDEs, specifically focusing on advection-diffusion models with TVPs involved in the source terms. Numerical results on a set of one-dimensional test problems demonstrate the effectiveness of the proposed estimation procedure in tracking unknown TVPs of different forms, while simultaneously estimating variance parameters affecting the TVP evolution model, from partial, noisy observations of the solution at discrete spatial locations and times.