Biologically-motivated system identification: application to microbial growth modeling.

Area of Science:

• * System identification and control engineering.
• * Computational modeling and simulation.
• * Bioprocess engineering and microbial kinetics.

Background:

• * Traditional system identification methods struggle with models that are linear in parameters but nonlinear in signal operations.
• * Existing approaches often lack robust mechanisms for automatic model structure selection.
• * Accurate modeling of complex biological processes like microbial growth is crucial for optimization and control.

Purpose of the Study:

• * To present a novel hybrid method for identifying system models with linear parametric structure and nonlinear signal operations.
• * To integrate traditional system identification with advanced strategies: linear-time-invariant-in-parameters models, set-based parameter identification, and evolutionary model structure selection.
• * To demonstrate the method's efficacy and performance, particularly the evolutionary model determination aspect, using microbial growth kinetics as a case study.

Main Methods:

• * Development of a hybrid system identification framework.
• * Incorporation of linear-time-invariant-in-parameters (LTIP) models.
• * Application of set-based parameter identification techniques.
• * Utilization of evolutionary algorithms for model structure selection and optimization.

Main Results:

• * Advancements in the theoretical foundation for the proposed identification methods.
• * Successful operation and performance evaluation of the hybrid approach.
• * Effective application to the complex problem of modeling microbial growth using Monod Kinetics.

Conclusions:

• * The presented method offers a powerful new tool for identifying complex system models.
• * Evolutionary model determination significantly enhances the selection of appropriate model structures.
• * The approach shows promise for modeling nonlinear biological systems with linear parametric structures.