Fuzzy clustering analysis of batch production recipe

According to the recipe information model specified in batch standard (ISA S88), via the modified recipe similarity measurement method, an improved fuzzy clustering algorithm (IFCA) is proposed based on fuzzy c-means (FCM) algorithm in the PNN network. IFCA can effectively improve the choice of clustering center and has a good clustering effect for the recipe of batch production. Then use IFCA to cluster for production recipe of an equipment unit wort evaporator in beer production. Finally, the recipe classification result is compared with the classification of FCM algorithm, and the effectiveness of IFCA is proved.