Design And Implementation Of A Computer-aided NMR Experiment Design Expert System

Nuclear magnetic resonance (NMR) spectroscopy is used widely in the area of structure elucidation, due to its unique characteristic of high resolution, quantitativeness, and abundance in experimental methods. Decades of the development in NMR theory and instrumental technology has brought forth a large number of different experimental methods targeting different problems and experimentation conditions. The abundance in methods has helped NMR to expand its utilization greatly. However, it also puzzles the students and researchers who want to use NMR to solve their problems.In this thesis, expertise of NMR experiment experts is collected and combined with knowledge comes from text books, spectrometer manuals to build a exert system used for computer-aided NMR experiment design. Given information on problem to solve, sample characteristics, and spectrometer configurations, the expert system can suggest on the experiments to carry out. When the user feeds back the results of the suggested experiments, the system can give further advices accordingly. This interactive procedure can be repeated. For suggested experiments, the system also provides detailed information on related theory, pulse sequence, parameter settings and items to note. The system is not only useful for users to choose appropriate experiments to solve specific problems, but also beneficial to tutoring of experimental NMR.In the implementation of the system, CLIPS is used as the inference core and Visual C++ is used to develop a graphics user interface. The separation of the inference core and the user interface makes the software interface more user-friendly, and makes the maintenance and extensity of the knowledge-base separate from the user interface, thus easier.