Transient thermal modeling of bioprocess equipment

• Published in: International journal of heat and mass transfer Vol. 184; p. 122064
• Main Authors: Cummings, Cody M., Wang, Hailei, Smith, Mark Thomas
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01-03-2022; Elsevier BV
• Subjects: Bioprocessing; Bioreactors; Control equipment; Environmental control; Fluid flow; Heat transfer; Heat transfer coefficients; Mathematical models; Parameters; Temperature control; Temperature dependence; Temperature profiles; Thermal analysis
• ISSN: 0017-9310; 1879-2189
• DOI: 10.1016/j.ijheatmasstransfer.2021.122064

•Created a thermal model of bioreactors, TCU, and PID controllers.•Validated U values against range of Bioreactor and TCU combinations.•Model enables selection and tuning of control parameters in-silico.•Model predicts heat and cool rates of Bioreactor, TCU, and controller combinations.
Bioprocessing is leveraging cells to produce high value, lifesaving products. Precise environmental control is essential to maintain integrity of the bioprocessing production process, which requires both appropriate equipment choice of the temperature control unit (TCU) and proper control parameter selection in order to reach the targeted process temperature in the desirable rate. To optimize the TCU selection and the associated control parameters, a transient thermal model of typical bioprocessing system was developed to help predict the process temperature profiles. The model captures the heat transfer processes and temperature dependent fluid and flow properties. The control systems for both the bioreactors and TCUs were modeled in detail to reflect their system response in silico. Physical experiments were also conducted across a range of bioreactors and other vessels, from 50 L to 2000 L, in order to validate the model. Various TCU size ranged from 1.2 kW to 18 kW were used in the experiment in order to broaden the model application. The measured time associated with each temperature was compared with the model prediction, which shows in good agreement. When compared to the total of 42 experiments, the predicted overall heat transfer coefficients match reasonably well with the experimental data. Multiple jacket-side Nusselt number correlations were also compared against the experimental data to provide additional insight of the heat transfer process.