Droop model

Model description: 

The behavior of phytoplankton cells in a continuous reactor is usually described by the Droop model. Cell growth is limited by a nutrient with concentration $S$. The biomass has a concentration $N$ and $Q$ represents the cell quota of assimilated nutrient, expressed as the amount of intracellular nutrient per biomass unit. The dilution rate $D$ corresponds to the flow rate of renewal medium over the volume of the reactor, and $D$ is the input of the system.

We denote $D = D_0 + u$, and the system fits

$$\sum_D \begin{cases} \dot{x}_i = f(x) + ug(x)\\ y=h(x_1) \end{cases}$$

with

$f(x)=\begin{pmatrix} a_2\left(1-\dfrac{1}{x_2}\right)x_1 - D_0x_1\\ a_3\dfrac{x_3}{a_1+x_3} - a_2(x_2 - 1)\\ D_0(1-x_3)-\dfrac{x_1x_3}{a_1+x_3} \end{pmatrix}$

$g(x)=\begin{pmatrix} -x_1\\ 0\\ 1-x_3 \end{pmatrix}$, and $h(x_1)=x_1$, where

$ x_1 = (\rho_m N/S_i);\\ x_2 = (Q/K_Q);\\ x_3 = (S/S_i);\\ a_1 = (K_{\rho}/S_i);\\ a_2 = \mu_m;\\ a_3 = (\rho_m/K_Q). $

Type: 

Form: 

Model order: 

3

Time domain: 

Linearity: 

Publication details: 

TitleNonlinear observers for a class of biological systems: application to validation of a phytoplanktonic growth model
Publication TypeJournal Article
Year of Publication1998
AuthorsBernard, O., Sallet G., and Sciandra A.
JournalIEEE Transactions on Automatic Control
Volume43
Start Page1056
Issue8
Pagination1056-1065
Date Published08/1998
ISSN0018-9286
Accession Number6002262
Keywordsbiocybernetics, living systems, nonlinear systems, observability, observers, physiological models
AbstractThe authors construct nonlinear observers in order to discuss the validity of biological models. They consider a class of systems including many classical models used in biological modeling. They formulate the nonlinear observers corresponding to these systems and prove the conditions necessary for their exponential convergence. They apply these observers on the well-known Droop model which describes the growth of a population of phytoplanktonic cells. The validity of this model is discussed based on the performance of the observers working on experimental data
DOI10.1109/9.704977