Category
Poster - Basic
Description
Free radical polymerization of Polyethylene is highly exothermic, having the largest adiabatic temperature rise of all commercial monomers. At high temperatures, the monomeric unit Ethylene can decompose into various hydrocarbons, producing even more heat. This decomposition can create a vessel pressure thousands of times larger than earth’s atmosphere, potentially leading to reactor explosions on the scale of earthquakes. Obviously, it is very important to ensure Polyethylene reactors are bounded, so their safe operating limits need to be determined. Since this is dangerous to do in experiment, it is often performed using computer simulations. This work uses Computational Fluid Dynamics (CFD) to develop a model for a Low-Density Polyethylene (LDPE) Continuous Stirred Tank Reactor (CSTR), for the purpose of determining safe reactor operating conditions. Because of the numerical approximations CFD employs, it can be difficult to obtain a stable and accurate model. The sensitive LDPE chemistry further complicates the stability requirements of this model. After a host of tests covering numerical settings, startup processes, kinetic simplifications, and reactor design, it was determined that a passive LDPE CSTR cannot be achieved in CFD. To illustrate this, two reactors that only differed in initial temperature by 0.01 K were modeled and compared. The temperature response of these models diverged from each other without bound. Because of this extreme sensitivity, active PID control was necessary for a stable CFD solution. The addition of Fuzzy Logic to the PID control improved the speed and stability of the model. The now stable LDPE CSTR model can be leveraged to determine safe reactor operating conditions.
Need For Control: Sensitivity of Polymerization Reactors
Poster - Basic
Free radical polymerization of Polyethylene is highly exothermic, having the largest adiabatic temperature rise of all commercial monomers. At high temperatures, the monomeric unit Ethylene can decompose into various hydrocarbons, producing even more heat. This decomposition can create a vessel pressure thousands of times larger than earth’s atmosphere, potentially leading to reactor explosions on the scale of earthquakes. Obviously, it is very important to ensure Polyethylene reactors are bounded, so their safe operating limits need to be determined. Since this is dangerous to do in experiment, it is often performed using computer simulations. This work uses Computational Fluid Dynamics (CFD) to develop a model for a Low-Density Polyethylene (LDPE) Continuous Stirred Tank Reactor (CSTR), for the purpose of determining safe reactor operating conditions. Because of the numerical approximations CFD employs, it can be difficult to obtain a stable and accurate model. The sensitive LDPE chemistry further complicates the stability requirements of this model. After a host of tests covering numerical settings, startup processes, kinetic simplifications, and reactor design, it was determined that a passive LDPE CSTR cannot be achieved in CFD. To illustrate this, two reactors that only differed in initial temperature by 0.01 K were modeled and compared. The temperature response of these models diverged from each other without bound. Because of this extreme sensitivity, active PID control was necessary for a stable CFD solution. The addition of Fuzzy Logic to the PID control improved the speed and stability of the model. The now stable LDPE CSTR model can be leveraged to determine safe reactor operating conditions.
Comments
Doctorate - 2nd Place Award Winner