INVESTIGADORES
DURAND Guillermo Andres
artículos
Título:
A Stiff Non-Linear ODE Simulation of a Batch/Continuous Plant Using Matlab/Simulink
Autor/es:
GUILLERMO A. DURAND; J. ALBERTO BANDONI
Revista:
MECANICA COMPUTACIONAL
Editorial:
Asociación Argentina de Mecánica Computacional
Referencias:
Lugar: Santa Fé; Año: 2003 vol. XXII p. 1373 - 1388
Resumen:
Abstract. A sugarhouse is one stage in the production of beet sugar. The purpose of the sugarhouse is to crystallize and separate sugar from the thick juice obtained in previous stages. These crystallization and separation processes are carried out in batteries of batch units, while the upstream and downstream stages are continuous. The dynamics of such plant generate a highly non-linear model, and the sequencing of the batch and intra-batch operations makes the simulation a problem with heavy stiff characteristics. The simulation is an ODE problem and is carried out in a MATLAB/Simulink© environment. But the choice of the correct ODE solver depends on their performance to handle the model’s stiff and non-linear characteristics. This project studies each ODE solver present in the Simulink suite. The study will consider also how the solver parameters affect its capacity to solve this simulation problem. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers. A sugarhouse is one stage in the production of beet sugar. The purpose of the sugarhouse is to crystallize and separate sugar from the thick juice obtained in previous stages. These crystallization and separation processes are carried out in batteries of batch units, while the upstream and downstream stages are continuous. The dynamics of such plant generate a highly non-linear model, and the sequencing of the batch and intra-batch operations makes the simulation a problem with heavy stiff characteristics. The simulation is an ODE problem and is carried out in a MATLAB/Simulink© environment. But the choice of the correct ODE solver depends on their performance to handle the model’s stiff and non-linear characteristics. This project studies each ODE solver present in the Simulink suite. The study will consider also how the solver parameters affect its capacity to solve this simulation problem. The final purposes of the sugarhouse simulation are: a) provide a test-bed where to apply proposed schedules for the batch units in order to obtain the optimal operation, and b) generate a training tool for operators and plant managers.