The EXLIBRIS project was aimed at developing, applying and testing an expert system methodology for optimized design, operation and control of high temperature industrial equipment such as furnaces, kilns and ovens. The developed expert system, named EXLIBRIS consists of an innovative integration of physical and model based knowledge with advanced optimization algorithms.
The predictive capacity of the EXLIBRIS system gives the possibility of operation optimization, equipment design optimization and advanced training.
The EXLIBRIS system is expected to answer to the following needs:
- Reduction of the energy consumption of the high temperature units.
- Reduction of pollution impact from those high temperature systems.
- Important training possibilities.
Under the scope of the EXLIBRIS project an optimization expert system concept was developed. Previously developed: i) physical-based modelling tools, ii) advanced identification models originated from a self-learning fuzzy expert system shell, iii) dedicated database where a fusion of data from sensors, modelling results and process data accessible to a dedicated optimization algorithm, were integrated in an adequate platform. The optimization allows the search of optimal operating conditions. This system is apt to be linked to an industrial process control system for on-line utilization. This connection was successfully tried during the project period. A distributed computational platform was used for this specific installation.
The application of the system to a glass melting furnace along with the feasibility studies to both continuous and discontinuous processes have shown the possibility of the use in a wide range of industrial processes. In addition to the glass-melting process for a full-scale application, a cement-melting kiln and a ceramic oven were considered as case studies for the feasibility studies.