Application Research Of Ceramic Materials Based On Data Mining

With the rapid development of China’s ceramics industry,the technological requirements for analyzing ceramic materials have become more and more sophisticated.Therefore,it is imperative to improve technologies of analyzing ceramic materials.So far,making chemical experiment is a main way for analyzing ceramic materials.However,it costs a lot and its technology is relatively backward.In view of the current situation,this paper introduces a cutting-edge data mining technology to analyze ceramic materials,mainly has the following three aspectsThe first is the application of GA-BP neural network in classifying ceramic materials.The BP neural network model is optimized through Genetic Algorithm so as to improve its prediction accuracy.The optimized model is applied to classify ceramic materials.The results showed that the predicated effect are excellent and to some extent,it is a fast and accurate way to classify a large amount of ceramic materials into different grades which has some practical value.The second is the application of association rule mining in experiments of selecting the ceramic glaze recipe.Ceramic glaze recipes are modeled through Apriori Algorithm.The ceramic glaze recipes are predicated and optimized by the model so as to lessen the glaze recipe experiments,select an excellent recipe quickly and reduce experimental cost.The third is the application in repairing ancient ceramic crack based on cluster analysis which improves Edge Detection Algorithm.The sobel operator of edge detection is optimized through cluster analysis.The gay value in the edge detection diagram of ancient ceramics is clustered so as to change the distribution of gray value.The results show that the outline,crack mark,edge trend of the improved edge detection diagram is enhanced obviously and the image of edge detection diagram is clear.This method is helpful in improving the accuracy of repairing the ancient ceramic crack.