PARAMETER ESTIMATION FOR BIOLOGICAL MODEL WITH SPIRAL DYNAMICS OPTIMIZATION

• Main Author: MICHELLA (NIM: 10113049), LEVINA
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/22890
• Institution: Institut Teknologi Bandung

Modeling and simulation of a biological phenomenon may involve a system <br /> <br /> <br /> of differential equations which describes the interactions between its biological <br /> <br /> <br /> components. Biological models usually depend on several parameters and initial <br /> <br /> <br /> conditions. Improper parameter values will cause misleading result in simulation. <br /> <br /> <br /> This can be prevented by estimating the unknown parameter values by fitting the <br /> <br /> <br /> model to experimental data. Parameter estimation relates to minimizing an objective <br /> <br /> <br /> function which evaluates the difference between model prediction and experimental <br /> <br /> <br /> data. Generally, the biological model involves the system of nonlinear differential <br /> <br /> <br /> equation. Parameter estimation problem will be brought to a nonlinear optimization <br /> <br /> <br /> problem. Therefore, the problem of parameter estimation will be solved numerically. <br /> <br /> <br /> In this thesis, parameter estimation for the biological model will be solved <br /> <br /> <br /> numerically with Spiral Dynamics optimization. In addition to Spiral Dynamics, <br /> <br /> <br /> Nonlinear Least Square and Genetic Algorithm method will be used to estimate the <br /> <br /> <br /> parameters of the biological model as a comparison. In terms of computing, Spiral <br /> <br /> <br /> Dynamics produces pretty good parameters with the fastest computation time.