Intelligent signal validation system for cupola furnace, Part II: system integration, testing and analysis

摘要

In Part I of this research (Paper 99-77), the motivation behind the signal validation system for the cupola furnace was presented. A methodology for developing an artificial neural network rule-based filter and trend estimator (ANN-RBFTE) and inferential sensors for the molten-iron temperature was described. In this paper, we present the testing results of the filter and inferential sensors, using cupola experimental data. A methodology for building a signal validation system for the cupola, by fusing the data from the ANN-RBFTE and the inferential sensors, is also described. The signal validation system is tested using experimental data from a cupola furnace. The testing results verify the excellent performance provided by the validation system. This research provides a solid foundation for the implementation of the signal validation system to other measurands in the cupola and to other industrial plants.