![]()
Abstract. Operator training, as well as process automation and optimization, are based on increasingly precise and comprehensive simulations of plant operations. For catalytic reforming, a key process for gasoline production, it is now possible to develop unit simulators based on advanced kinetic models of complex mixtures.
For process development and design, more predictive kinetic models are needed to replace the usual correlative approaches. The model is based on molecular kinetics. The rate expressions are of the Langmuir-Hinselwood type. The reactor effluent and feed compositions are described by a set of 39 pseudo-components.
This model has been validated through pilot and industrial runs.
For operator training and unit operation, we have developed process simulators for the entire unit either in steady state or in dynamic mode. The intrinsic kinetic model for reactors provides high fidelity representation of the unit behaviour. Such a simulator has been running since 1994 for operator training at the Elf-Donges refinery.
The advantage of this approach is to use the same basic modelling throughout the process life-cycle.
INTRODUCTION Modelling and simulation are used all along the process life-cycle, namely: development, design, operator training, unit operation monitoring, advanced process control, optimisation. Previously, the variety of goals of the simulation has led to numerous types of models and simulators for the same process at different steps of its life-cycle. Moreover, these models and simulators were developed, often with no links between them.
Modelling and simulation have been drastically moved forward these later years: - By the ‘technology push’: fast increase in computational power, new software development, - By the ‘end-user pull’: refiners are asking for more automation and optimisation of their plants; process development and design staff within R&D centres need accurate models to improve the quality of the unit's design and to provide refiners with safe and productive plants.
The requirements of the various end-users are now converging towards high-fidelity modelling. As the computer tools are available to satisfy these requirements, it is now possible to use the same basic highfidelity modelling for all types of simulation and control of a given process.
KINETIC MODELLING IFP has developed such high fidelity models for catalytic reforming, a key process for gasoline production. The model has been first studied and developed by the team of Professor G. Froment in the Laboratorium voor Petrochemische Techniek of the Rijksuniversiteit (Gent, Belgium). The model was then adapted to the specific needs of IFP and validated by