| Aimed at the high-efficiency and high-credibility of great precious vacuum resin dosing equipment and with the help of engineering practice of VCD-800, the metric pump ,metric valve and fault-auto-diagnosis system were studied, which were key parts of the equipment.First, the way combined with method and experiment were used to study the metric pump. Some design-principle and consult value were successfully gained and the mechanism of self-symmetry-effect upon precision and grind was analyzed.Second ,based on genetic algorithm the design of the metric valve was optimized. In this way the problem that classic method can't be used because of the lack of analytic expression was solved. The metric valve worked out follow the optimized design is superior to the former in the performances of all aspects.At last, based on analysis and classified studying , a fault-auto-diagnosis system was founded. Specially in which a BP artificial neural-network was used to recognize the sequential pressure signal of the dosing system and to identify its state. Which avoided the disadvantage of dealing with the signal as a static value. |
