TY - JOUR
T1 - Partial Observations and Conservation Laws
T2 - Gray-Box Modeling in Biotechnology and Optogenetics
AU - Lovelett, Robert J.
AU - Avalos, José L.
AU - Kevrekidis, Ioannis G.
N1 - Publisher Copyright:
© 2019 American Chemical Society.
PY - 2020/2/12
Y1 - 2020/2/12
N2 - Developing accurate dynamical system models from physical insights or data can be impeded when only partial observations of the system state are available. Here, we combine conservation laws used in physics and engineering with artificial neural networks to construct "gray-box" system models that make accurate predictions even with limited information. These models use a time delay embedding (cf., Takens embedding theorem) to reconstruct the effects of the intrinsic states, and they can be used for multiscale systems where macroscopic balance equations depend on unmeasured micro/meso-scale phenomena. By incorporating physics knowledge into the neural network architecture, we regularize variables and can train the model more accurately on smaller data sets than black-box neural network models. We present numerical examples from biotechnology, including a continuous bioreactor actuated using light through optogenetics (an emerging technology in synthetic biology) where the effect of unmeasured intracellular information is recovered from the histories of the measured macroscopic variables.
AB - Developing accurate dynamical system models from physical insights or data can be impeded when only partial observations of the system state are available. Here, we combine conservation laws used in physics and engineering with artificial neural networks to construct "gray-box" system models that make accurate predictions even with limited information. These models use a time delay embedding (cf., Takens embedding theorem) to reconstruct the effects of the intrinsic states, and they can be used for multiscale systems where macroscopic balance equations depend on unmeasured micro/meso-scale phenomena. By incorporating physics knowledge into the neural network architecture, we regularize variables and can train the model more accurately on smaller data sets than black-box neural network models. We present numerical examples from biotechnology, including a continuous bioreactor actuated using light through optogenetics (an emerging technology in synthetic biology) where the effect of unmeasured intracellular information is recovered from the histories of the measured macroscopic variables.
UR - http://www.scopus.com/inward/record.url?scp=85078476726&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85078476726&partnerID=8YFLogxK
U2 - 10.1021/acs.iecr.9b04507
DO - 10.1021/acs.iecr.9b04507
M3 - Article
AN - SCOPUS:85078476726
SN - 0888-5885
VL - 59
SP - 2611
EP - 2620
JO - Industrial and Engineering Chemistry Research
JF - Industrial and Engineering Chemistry Research
IS - 6
ER -