On-line concentration estimation during chemical reactions using adaptive heat/ mass balances


On-line concentration estimation during chemical reactions using adaptive heat/ mass balances

Kryk, H.; Hessel, G.; Schmitt, W.; Seiler, T.; Weiß, F.-P.; Hilpert, R.; Roth, M.

In the fine chemical industry, complex strongly exothermic reactions are usually carried out in semibatch mode in stirred tank reactors. These reactors are characterised by nonsteady-state conditions due to the discontinuous operation. Thus, it is difficult to estimate the actual conversion and to identify undesired process states during the chemical reaction, particularly, if the reactor is not equipped with expensive measuring systems for chemical on-line analysis. Therefore, an on-line monitoring system was developed that is able to estimate the concentration profiles without the need of chemical on-line analysis.
The on-line estimation of the concentration profiles is based on calculations of the thermal and chemical conversion courses during the chemical process using heat and mass balances. In order to consider factors of influence, such as heat losses, heat bridges and systematic measuring errors, adaptive model components are included. The adaptation of the complex model to the target plant is done by software-supported calculations of the adaptive parameters using process data of at least one normal batch course in the chemical plant.
The monitoring system was developed, optimised and tested by means of experiments in a miniplant at laboratory scale. As a test process, the strongly exothermic catalytic hydrogenation of an aromatic nitro compound was chosen. For industrial testing, the monitoring system was integrated into a batch-information-management system which was implemented into the process control system of a multipurpose reactor installation in the fine chemical factory at Radebeul (Degussa, Inc.). The tests showed good agreements of the estimated concentration profiles with analytical measurements in both, the laboratory and the industrial scale. Furthermore, it was found that deviant reaction courses and undesired process states can be recognised very earlier using the on-line monitoring system than with exclusive knowledge of process variables like temperatures, pressures and volume flows.

  • Lecture (Conference)
    4th European Congress of Chemical Engineering, 21-25 September 2003 Granada, Spain, Topic 10 Abstracts, O-10.2-001
  • Contribution to proceedings
    4th European Congress of Chemical Engineering, 21-25 September 2003 Granada, Spain, Topic 10 Abstracts, O-10.2-001

Permalink: https://www.hzdr.de/publications/Publ-5242
Publ.-Id: 5242