Simulatoren zur Unterstützung der Prozeß- und Betriebsführung

Simulators for support of process and plant management. Process simulators are increasingly used for plant and process operation purposes. These simulators differ from those used in process design in the user interface, which is identical to the usual operator interface of distributed control systems. The progress of the numerical solution parallels the dynamic behaviour of the real process, and not that of the numerical procedure. The simulation-supported tasks of plant and process operation are: education of operators, training and instruction of operators, support of plant operation by interpretion of plant states and transitions and by helping to formulate operating instructions. These tasks, the available simulation concepts and examples are discussed in detail.     

Expertensysteme zur Unterstützung der analytischen Chemie: Intelligente Anwenderberatung für die ICP-Massenspektrometrie

Expert Systems to Assist Analytical Chemistry: Intelligent User Counseling in ICP-Mass Spectrometry Because ICP-mass spectrometry is a sensitive and rather new technique, some operators may lack the required thorough expertise. An intelligent system is presented which at first helps to identify any problems originating from the components of the actual sample. By demand of the user an experimental strategy to acquire any lacking information is provided. Based on this knowledge possible interferences are prognosticated. An appropriate operating procedure is then proposed, including suitable details of sample preparation. After the analytical measurement a scanned mass spectrum may be automatically checked for plausibility and for evidence of unexpected sources of error. The system is strictly organized in separated modules. It is open for easy introduction of additional knowledge and for the cooperation with the software of the spectrometer. The expert system shell Nexpert has been used as a building tool.     

Expertensysteme zur Unterstützung der Analytischen Chemie: Verwechslungsprüfung und Schrott-Identifizierung in der Qualitätssicherung

Expert Systems to Assist Analytical Chemistry: Confounding Tests and Scrap Identification in Industrial Quality Control Quality control in the steel industry has to ensure the identity of semi-finished products and of scrap. This means to make a decision among about 2000 or more possibilities. The inductive learning algorithm ID3 is used to build the necessary decision tree, the main problem being overlap of the metal concentrations specifying the different alloys. Local attributes are introduced to split up remaining clusters without destroying the other parts of the tree. The system ist designed for use with industrial (mobile) spectrometers.     

Modellgestützte Prozeßleittechnik in der verfahrenstechnischen Produktion

Model-based process control engineering in chemical engineering production . Modern process control engineering nowadays provides almost ideal tools for the implementation of even highly sophisticated control concepts. The very reluctant acceptance of these tools is due to the following bottlenecks:

• – insufficient understanding of chemical engineering control processes;
• – limitations in the acquisition of sensorial primary information;
• – insufficient performance of control engineering concepts.

The present articles considers how these bottlenecks can be overcome.     

Expertensysteme zur Unterstützung der Analytischen Chemie: Störungserkennung und Plausibilitätsprüfung in der Ionenchromatographie

Expert Systems to Assist Analytical Chemistry: Troubleshooting and Plausibility Testing in Ion Chromatography An expert system for automated troubleshooting and plausibility testing in ion chromatography is introduced. The program is able to detect some major problems in the chromatograms. First it evaluates the incoming data to get the retention time, baseline drift and noise, the peakshape and other criteria. Based on this information conclusions are drawn whether an analysis is formally correct or not. Even more problems can be solved depending on the inputs of the users. The expert system is able to deal with many standard problems (523 are implemented). To evaluate the plausibility of the data the system uses known information about the sample (analytes and concentrations) to calculate ranges of credible results. The troubleshooting routines lead to at least 90% correct conclusions. The plausibility check needs many informations for the inference process, it succeeds best with routine problems.     

Expertensysteme zur Unterstützung der Analytischen Chemie: Realisierung von Lernfähigkeit und Plausibilitätsprüfung am Beispiel der ICP-Massenspektrometrie

Expert Systems to Assist Analytical Chemistry: Realization of Learning Ability and Plausibility Checking Demonstrated for ICP Mass Spectrometry Certain, uncertain and lacking knowledge has to be considered for intelligent counseling. In the ICP mass spectrometry the composition of the actual sample and the ionization rates are the most important parameters to be prognosticated. The way of storage and retrieval of data and of decision making should be automatically checked and improved with respect to the success of previous guesses. The Dempster-Shafer theory is used for the combination and propagation of uncertainties.     

Erhaltung der Margarinequalität von der Produktion bis zur Auslieferung

Maintenance of the Quality of Margarine from Production to Delivery Margarine, which is a sensitive food-stuff, requires utmost care in order that it reaches the consumer in uniform quality. Apart from the time factor, there are numerous other external influences, such as heat, light, pressure, impact etc., which impair the quality of margarine and its packing. These influences have to be encountered by preventive measures, such as proper choice of composition, manufacturing process and packaging on one hand, and maintenance of optimum conditions during storage and transport on the other. Selected examples are given to show, as to how impairing influences during storage and transport can be traced back by series of tests and the technical and organisational changes made accordingly.     

Beiträge der Prozeßleittechnik zur Qualitätssicherung in der chemischen Produktion

Contributions of Process Control to Quality Assurance in the Chemical Industry. Through their functions of measurement, i.e. the acquisition of quantities relating to the process and the product, and adjustment, i.e. the modification of control values, process control engineering and process automation represent the basis of all operative control of quality. The functions of control, regulation, and operator support are likewise instruments ensuring the maintenance of quality-relevant process and product specifications. Utilisation of the considerable potential of these tried and tested control methods is logical in the case of chemical processes and should receive more attention in quality management.     

Prozeßführung zur Sicherung reproduzierbarer Qualität in der verfahrenstechnischen Produktion

Process engineering to assure reproducible quality in production. This paper presents an hierarchically structured process control concept in which product quality is regarded as the central orientation factor. The application of control engineering methods is intended to assure maintenance of the product quality specifications. The second part of the article briefly considers the present state of the art of quality assurance. Fundamental considerations concerning a well-defined quality concept as a basis for process control are presented in the third part, prior to explanation of the process control concept and its illustration with the aid of an example.