Distillation models, first-principle

In this work, new developments were achieved through the use of new examples, one of which the optimisation of a real crude distillation unit involving 19 decision variables. The performance of the metamodel-based optimisation is compared with results obtained with the optimisation based on a first-principles model, embedded in a sequential-modular process simulator. It is shown that metamodel-based optimisation with adaptation of the metamodels during the optimisation procedure provides results with good accuracy and significant reduction of computational effort. The performance comparison between neural networks and kriging models for chemical processes is another contribution of this work. 

The process model was built using PETROX, a proprietary sequential-modular process simulator from PETROBRAS. The simulation comprises 53 components and pseudocomponents and 64 unit operation modules, including 7 distillation columns and a recycle stream. All modules are built with rigorous, first-principles models. For optimization applications, PETROX was linked to NPSOL, an SQP optimisation algorithm. 

The methodology was tested with an example involving the optimisation of a crude distillation unit, using the first-principles models of a sequential-modular process simulator. The solution of the corresponding optimisation problem with this rigorous model required considerable computational effort. 

A distillation column model was chosen to test the various solvers. First-principle distillation models yield a tridiagonal block structure, each block... 

Objective entities in the process operation system, such as reactor, distillation column units and processes, are used and conducted by the different operation tasks. Modeling methods for the units and processes objects are first principle rules, statistical regression, artificial neural network, and fuzzy logic relation. 

The book is divided into five parts. Part I provides an introduction to process control and an in-depth discussion of process modeling. Control system design and analysis increasingly rely on the availabihty of a process model. Consequently, the third edition includes additional material on process modeling based on first principles, such as conservation equations and thermodynamics. Exercises have been added to several chapters based on MATLAB simulations of two physical models, a distillation column and a furnace. These simulations are based on the book. Process Control Modules, by Frank Doyle, Ed Gatzke, and Bob Parker. Both the book and the MATLAB simulations are available on the book Web site (www.wiley.com/college/seborg). National Instruments has provided multimedia modules for a number of examples in the book based on their Lab VIEW software. 

The principle of the perfectly-mixed stirred tank has been discussed previously in Sec. 1.2.2, and this provides essential building block for modelling applications. In this section, the concept is applied to tank type reactor systems and stagewise mass transfer applications, such that the resulting model equations often appear in the form of linked sets of first-order difference differential equations. Solution by digital simulation works well for small problems, in which the number of equations are relatively small and where the problem is not compounded by stiffness or by the need for iterative procedures. For these reasons, the dynamic modelling of the continuous distillation columns in this section is intended only as a demonstration of method, rather than as a realistic attempt at solution. For the solution of complex distillation problems, the reader is referred to commercial dynamic simulation packages. 

---