Year: 2026
Type: preprint
Abstract: Stochastic reaction networks governed by Chemical Langevin Equations (CLE) exhibit pronounced multiscale dynamics spanning fast molecular reactions, intermediate transport, and slow cellular regulation, posing significant challenges for efficient and accurate simulation. Although operator splitting naturally decouples fast and slow subsystems, a rigorous error characterization for CLE splitting schemes has been lacking. We propose a modular operator-splitting framework with adaptive discretization that enables reliable and efficient simulation across fast-slow dynamics with explicit control of discretization error. Using stochastic logarithmic representations, we present a complete error analysis of the fast-slow Lie-Trotter splitting method, decomposing the one-step error into stochastic flow truncation error, commutator errors due to subsystem noncommutativity, and numerical discretization errors from fast and slow integrations. Guided by this analysis, we develop a proportional-integral (PI) adaptive controller that jointly selects macro time steps and fast microsteps, achieving substantial efficiency gains while maintaining accuracy.
Source: arXiv (Cornell University)
Authors: Zeng, Yuming, Xie, Wei, Wang, Keqi
Institution: Northeastern University
Subfield: Numerical Analysis
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
Awards: G3387330441, PURPOSE: THE PURPOSE OF THIS GRANT IS TO DEVELOP A BIOMANUFACTURING MODELING AND ONTOLOGY PLATFORM TO FACILITATE CONTINUOUS MANUFACTURING THAT ACCELERATES DIGITAL TWIN MODEL DEVELOPMENT AND DEPLOYMENT; HAS MODULAR DESIGN SUPPORTING THE ASSEMBLY OF FLEXIBLE MANUFACTURING PROCESSES AND FACILITATING THE REUSE OF DATA COLLECTED FROM VARIOUS SOURCES; AND SUPPORTS MANUFACTURING SYSTEM INTEGRATION, INTEROPERABILITY, AND REGULATORY COMPLIANCE.ACTIVITIES TO BE PERFORMED: THIS PROJECT WILL CREATE AN INTEGRATED BIOMANUFACTURING PROCESS ONTOLOGY FOR MANUFACTURING OPERATIONS THAT CAN FACILITATE THE DATA INTEGRATION BETWEEN DIFFERENT PRODUCTION SYSTEMS; DEVELOP AN ONTOLOGICALLY ENCODED MULTI-SCALE MODEL AND DIGITAL TWIN WITH MODULE DESIGN CHARACTERIZING CAUSAL DEPENDENCE FROM INPUTS TO OUTPUTS AND FUNDAMENTAL MECHANISMS AT MOLECULAR, CELLULAR, AND MACRO-SCOPE LEVELS; AND CONSTRUCT A MACHINE LEARNING ONTOLOGY THAT MODELS THE DIGITAL THREAD OF MODEL DEVELOPMENT, TRAINING, AND PREDICTION ALONG WITH ASSOCIATED DATA PROCESSING ALGORITHMS.EXPECTED OUTCOME: THIS PROJECT WILL DELIVER A MULTI-SCALE MODEL BASED DIGITAL TWIN AND MACHINE LEARNING ONTOLOGY PLATFORM THAT CAN FACILITATE MANUFACTURING SYSTEMS INTEGRATION, ENHANCE DATA INTEROPERABILITY, AND SUPPORT THE REUSE OF DATA FROM DIFFERENT PHYSICAL AND DIGITAL SYSTEMS.INTENDED BENEFICIARIES: THIS PROJECT WILL BENEFIT BIOMANUFACTURERS AND THEIR SUPPLIERS TO IMPROVE MANUFACTURING SYSTEMS INTEGRATION.SUBRECIPIENT ACTIVITIES: THE RECIPIENT DOES NOT INTEND TO SUBAWARD FUNDS., CAREER: Mechanism-Informed AI for Biological Systems-of-Systems to Accelerate Biomanufacturing Systems Integration and Innovations, THE NATIONAL INSTITUTE FOR INNOVATION IN MANUFACTURING BIOPHARMACEUTICALS (NIIMBL), G6671297155, THE NATIONAL INSTITUTE FOR INNOVATION IN MANUFACTURING BIOPHARMACEUTICALS (NIIMBL)
OpenAlex ID: W7148424905
Cites:
Cited by:
Related to: