scholarly journals Robust optimization of stiff delayed systems: application to a fluid catalytic cracking unit

Author(s):  
Jonas Otten-Weinschenker ◽  
Martin Mönnigmann

AbstractWe apply a robust steady state optimization method for stiff delay differential equations to the economic optimization of a fluidized catalytic cracking unit. Stiff systems of differential equations appear in this case due to the different time scales in the gas and fluid phase. Delays result from the catalyst hold-ups in the standpipes connecting riser and regenerator. We show that the proposed robust optimization method can cope with stiffness and delays. Moreover, the proposed method is capable of simultaneously optimizing the process parameters and tuning controller parameters.

Author(s):  
Tamas Insperger

It is known that stability properties of delay-differential equations are not preserved by Taylor series expansion of the delayed term. Still, this technique is often used to approximate delayed systems by ordinary differential equations in different engineering and biological applications. In this brief, it is demonstrated through some simple second-order scalar systems that low-order Taylor series expansion of the delayed term approximates the asymptotic behavior of the original delayed system only for certain parameter regions, while for high-order expansions, the approximate system is unstable independently of the system parameters.


Author(s):  
Sun Yi ◽  
Patrick W. Nelson ◽  
A. Galip Ulsoy

In this paper, we consider the problem of feedback controller design via eigenvalue assignment for systems of linear delay differential equations (DDEs). Unlike ordinary differential equations (ODEs), DDEs have an infinite eigenspectrum and it is not feasible to assign all closed-loop eigenvalues. However, we can assign a critical subset of them using a solution to linear DDEs in terms of the matrix Lambert W function. The solution has an analytical form expressed in terms of the parameters of the DDE, and is similar to the state transition matrix in linear ODEs. Hence, one can extend controller design methods developed based upon the solution form of ODEs to systems of DDEs, including the design of feedback controllers via eigenvalue assignment. We present such an approach here, illustrate using some examples, and compare with other existing methods.


Author(s):  
Д.А. Желтков ◽  
Р.М. Третьякова ◽  
В.В. Желткова ◽  
Г.А. Бочаров

Системы уравнений с запаздываниями широко применяются в различных областях современного математического моделирования. В ходе разработки структуры математической модели и идентификации ее параметров приходится многократно решать задачу Коши для подобных систем. В случае высокой размерности системы, а также при условии жесткости задачи численное решение уравнений с запаздываниями может требовать значительных вычислительных и временных затрат. Таким образом, разработка и реализация эффективных алгоритмов численного решения различных классов уравнений с запаздывающими аргументами является актуальной задачей. В настоящей статье представлена модифицированная версия программного комплекса DIFSUBDEL, в которой реализованы методы численного решения дифференциальных уравнений с запаздываниями на основе линейных многошаговых методов. Переработанная версия разработана с применением принципов структурного программирования и является значительно более удобной в эксплуатации, чем исходная, а также обладает свойством потокобезопасности, что позволяет использовать комплекс в качестве блока в системах, основанных на технологиях параллельного программирования с общей памятью. Был проведен сравнительный анализ производительности переработанной системы DIFSUBDEL c другими существующими программными реализациями численных методов решения дифференциальных уравнений с запаздыванием и показана ее эффективность. The systems of delay differential equation are widely used in modern mathematical modeling. The process of mathematical model development and identification requires the repeated solution of initial value problems for such systems. The numerical solution of delay differential equations may be computationally expensive, especially when the problem is stiff and high-dimensional. Therefore, it is important to develop and implement the efficient algorithms for the numerical solution of different classes of delay differential equations. In this paper, a new implementation of DIFSUBDEL program package for the numerical solution of delay differential equations based on linear multistep methods is discussed. The modified version is based on structured programming principles to make the program thread safe and user-friendly. The modified program performance is compared with the existing implementations of numerical methods for the solution of delay differential equations.


2010 ◽  
Vol 16 (7-8) ◽  
pp. 943-960 ◽  
Author(s):  
Y.N. Kyrychko ◽  
S.J. Hogan

This paper is a review of applications of delay differential equations to different areas of engineering science. Starting with a general overview of delay models, we present some recent results on the use of retarded, advanced and neutral delay differential equations. An emerging area for modeling with the help of delay equations is real-time dynamic substructuring, or hybrid testing. We introduce the main idea of this technique together with the latest advances. Special emphasis is given to the development of the theory and applications of partial delay differential equations. The review concludes with a summary of some open problems and questions concerning the analysis of spatially extended delayed systems.


Author(s):  
J.K. Lampert ◽  
G.S. Koermer ◽  
J.M. Macaoy ◽  
J.M. Chabala ◽  
R. Levi-Setti

We have used high spatial resolution imaging secondary ion mass spectrometry (SIMS) to differentiate mineralogical phases and to investigate chemical segregations in fluidized catalytic cracking (FCC) catalyst particles. The oil industry relies on heterogeneous catalysis using these catalysts to convert heavy hydrocarbon fractions into high quality gasoline and fuel oil components. Catalyst performance is strongly influenced by catalyst microstructure and composition, with different chemical reactions occurring at specific types of sites within the particle. The zeolitic portions of the particle, where the majority of the oil conversion occurs, can be clearly distinguished from the surrounding silica-alumina matrix in analytical SIMS images.The University of Chicago scanning ion microprobe (SIM) employed in this study has been described previously. For these analyses, the instrument was operated with a 40 keV, 10 pA Ga+ primary ion probe focused to a 30 nm FWHM spot. Elemental SIMS maps were obtained from 10×10 μm2 areas in times not exceeding 524s.


Author(s):  
Clifford S. Rainey

The spatial distribution of V and Ni deposited within fluidized catalytic cracking (FCC) catalyst is studied because these metals contribute to catalyst deactivation. Y zeolite in FCC microspheres are high SiO2 aluminosilicates with molecular-sized channels that contain a mixture of lanthanoids. They must withstand high regeneration temperatures and retain acid sites needed for cracking of hydrocarbons, a process essential for efficient gasoline production. Zeolite in combination with V to form vanadates, or less diffusion in the channels due to coke formation, may deactivate catalyst. Other factors such as metal "skins", microsphere sintering, and attrition may also be involved. SEM of FCC fracture surfaces, AEM of Y zeolite, and electron microscopy of this work are developed to better understand and minimize catalyst deactivation.


Sign in / Sign up

Export Citation Format

Share Document