Simulation Studies of Chemical Process Applications
Aditya Kumar, Pródromos Daoutidis
Abstract
Aditya Kumar, Pródromos Daoutidis
Abstract
In this chapter, the authors study, through simulations, the control of several chemical process applications that are modeled by high-index DAEs. While these processes can also be modeled by index-one DAEs or ODEs by appropriately modifying the modeling assumptions, they illustrate the limitations of controllers designed on the basis of such models. In the first example, the authors illustrate the problem of controller ill-conditioning, where the control action evaluated by the controller designed on the basis of the rate-based ODE model is highly sensitive to small modeling errors. The second example illustrates the problem of closed-loop instability for slightly nonminimum phase systems, where the zero dynamics of the process is also a two-time-scale system with the fast subsystem being unstable.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this chapter, the authors study, through simulations, the control of several chemical process applications that are modeled by high-index DAEs. While these processes can also be modeled by index-one DAEs or ODEs by appropriately modifying the modeling assumptions, they illustrate the limitations of controllers designed on the basis of such models. In the first example, the authors illustrate the problem of controller ill-conditioning, where the control action evaluated by the controller designed on the basis of the rate-based ODE model is highly sensitive to small modeling errors. The second example illustrates the problem of closed-loop instability for slightly nonminimum phase systems, where the zero dynamics of the process is also a two-time-scale system with the fast subsystem being unstable.
Key concepts: Process (computing), Computer science, Process engineering, Engineering, Programming language