Process simulation

The process simulation is a tool for development and optimization of technical processes in process engineering or chemical plant.

Principle

The process simulation is essentially an image of chemical processes and unit operations in computer programs. For the modeling of a set of knowledge is necessary:

• Chemical substance and mixture properties ( by measurements, database searching, correlations or assessments)
• Mass and heat transport,
• Device properties ( reactors, columns, mixer, condensers, evaporators, etc.),
• Reaction mechanisms and kinetics, refer to Chemical Reaction Engineering,
• Suitable mathematical and chemical models
• Efficient solution procedure.

Process simulation ensures that the mass and heat balances are correct and are brought into a stable equilibrium. Most processes can be visualized simultaneously.

History

The first attempts to simulate processes electronically, has already been made in the 1950s to electronic analog computers. These approaches, however, were quickly abandoned in favor of simulations on digital computers.

Initial developments in the digital process simulation of chemical plants were started in the 1970s, since the first time appropriate hardware and software (especially here the FORTRAN programming language) were available. The modeling of chemical properties has already started much earlier, here are, for example, cubic equations of state ( see, eg, Van der Waals equation) and correlations ( see, eg, Antoine equation ) should be mentioned, which were developed in the 19th century and today partly be used. Also, studies on the kinetics of chemical reactions and reaction mechanisms were advanced. Device properties had also been largely modeled so that all tools were available, in silico modeling complete chemical processes (not including computers by ) and calculate.

At the same time the development process simulation has the development of the various models for the estimation of material properties, reaction mechanisms, their kinetics, device properties, etc., but in particular the development of factual databases greatly accelerated. Factual databases are now used to develop estimation and correlations.

Static and dynamic process simulation

Originally, the process simulation has been applied only to stationary systems. This gives a complete mass balance and energy balance of a steady state on the basis of models. This static simulation is now complemented by the dynamic simulation. Dynamically in this context means that time-dependent results are calculated. In principle, the flow diagram is considered infinitesimal and calculated as a system of differential equations numerically. This process requires a much higher computing power, but also allows the transition to the control and management of chemical plants in real time. A simple example is the filling or emptying of a container. In the dynamic simulation, in particular, control processes (PID controller ), hold-ups and chemical reactions can be represented very realistically.

Phase equilibria

The most common phase equilibrium, the vapor -liquid equilibrium ( VLE generally for vapor-liquid equilibrium hereinafter) is that in particular in gas scrubbers and in the rectification of importance. But in the calculation of the boiling and Tautemperaturen it is applied. Under ideal substances such as alkanes, as the model satisfies the Raoult - Dalton Law, which is based on the definition of the Partialbruches. At non-ideal mixtures is calculated in the liquid phase and the activity coefficient in the gas phase of the fugacity coefficient, and thus corrects the Raoult - Dalton Law. During the fugacity coefficient well from the equation of state ( Soave - Redlich - Kwong frequently ) can be calculated for each individual substance in a mixture, the activity coefficient of the binary interactions is dependent. In a mixture with eg 10 ingredients exist 45 binary interactions. Therefore, in this case 45 VLE would be measured. VLE measurements can be found in databases such as the DETHERM or the DDB and in the literature such as DECHEMA Data Collection. In it you will also find the belonging parameters of suitable models such as Non- Random -Two - Liquid model ( NRTL ). For many binary mixtures that were not measured, the model parameters can be estimated using the UNIFAC method. The UNIFAC model is described inter alia in the VDI Heat Atlas.

The greater the activity coefficients, the clearer the xy diagram is different (x refers to the composition of the liquid, y represents the vapor composition ) from that of an ideal VLE until it finally cut line or an S-curve is what the sign of azeotropism and possibly a miscibility gap is. This can be easily demonstrated at the Porter model.

Finally, it is also possible with the non -random two- liquid model (NRTL ) a liquid -liquid equilibrium (LLE ) for calculating, assuming that the parameters are known. Approximation, one can certainly calculate an LLE with NRTL VLE data. The greater the miscibility gap is, for example, benzene, water, the lower the error. The system of n- butanol-water mixture with a lower hole, the approximation is not acceptable. With appropriate data, even complex LLE such as 3- methylpyridine - water can be calculated with elliptic equilibrium lines.

On the basis of appropriate data for the heat of fusion can be calculated with the NRTL model even Feststofflöslichkeiten ( solid-liquid equilibrium, shortly SLE for Solid - Liquid Equilibrium ). In many materials, such as very narrow boiling substances 1- methyl naphthalene and 2-methyl- naphthalene there is immediately a eutectic, whose position is a good approximation of reality.

Database

The substances used in the process simulation are selected from a database. The database contains gases, liquids, solids. Polymers and electrolytes. It can be extended with your own materials and data by regression. The database provides temperature-independent data, such as critical pressure and temperature functions for example, the vapor pressure, specific heat capacity, etc. Known databases are DETHERM and the Dortmund Data Bank, which essentially include the experimental data for pure substances and mixtures, and the DIPPR database which essentially contains the parameters of equations for pure chemicals. With the help of mixing rules, the material data of mixtures of known pure component properties can be calculated approximately.

Rectification

The rectification, often called distillation, is one of the key unit operations in process simulation but also in the chemical process engineering. The older model FUG ( Fenske - Underwood - Gilliland ), which is a quick and good approximation for ideal mixtures hardly plays a role. Rather, the Simultaneous Correction system (see Perry's Chemical Engineering Handbook ) has prevailed, which can convert nearly all types of rectification model well such as azeotrope, extraction, reactive distillation, separation wall column, gas scrubbing, absorption, desorption, electrolytes, side column. For petrochemical distillation is also certainly not the model inside-out in use, because it converges quickly and the mixture consists predominantly of alkanes.

The batch distillation can be simulated. In most cases here, the algorithms of the continuous distillation can be used. With the batch distillation, a multi-fuel mixture can be divided chronologically into individual fractions. The mathematical description of the batch distillation is carried out by means of the Rayleigh distribution.

Reactors

The most common types are the stoichiometric model, the equilibrium and the kinetic reactor. The equilibrium reactor can be modeled on the one hand according to the Gibbs theory and according to van't Hoff. For the kinetic reactor one usually uses the model of Arrhenius. In combination with VBA to kinetic approaches can represent arbitrary, eg for biological reactions.

Interfaces

For optimal operation of the process simulation used interfaces such as Excel for data transmission in a project database or system. Using the MS -COM Technology process simulation can even be controlled from within Excel, ie be started. Thereby even an online simulation is possible to be continuously supplied with data from the current process simulation systems. The results are designed for optimum process control.

Process simulation software

Simulation programs are available in large numbers, the more significant are eg

• Aspen simulator and HySys ( Aspen Technology, Inc.)
• Pro / II and DYNSIM (both from SimSci ) ( Invensys )
• System 7 ( EPCON )
• RSI ( RSI Group)
• CHEMCAD ( Chemstations )
• IPSEpro ( SimTech )

Larger companies often have in-house developments in use, which are used exclusively in-house. Two examples are

• CHEMASIM (BASF )
• VT Plan ( Bayer)

While the above systems in general can be used to simulate pure fluid processes preferably, the simulation program SolidSim can be specially applied to the simulation of solids processes.

• Technical Chemistry
• Process engineering
• Computer simulation