Glaze Recipe Based On Mathematical Programming And Artificial Neural Network Optimization

Glaze in ceramic glaze production plays a very important significance,because the cost of raw materials in the market will continue to change,so glaze producers often need to follow the market fluctuations to regulate glaze formula,used to determine the glaze composition,finally reached the market minimum,to save costs.Due to the glaze rich in domestic production,even cost formula drop very low,also can bring remarkable economic benefits for the glaze manufacturers,glaze fast calculation of minimum cost formula,in the formulation of the glaze industry formation process computer in the glaze formulation design has an irreplaceable position and role.By all ways and means to achieve component of the minimum requirements,making glaze the price down to the lowest,meet the minimum content requirements,to get the optimal formula,which is the ultimate goal of our research.Four mathematical programming methods are used in this paper,namely,linear programming,objective programming,stochastic programming,fuzzy programming.At the same time,it uses two kinds of mathematical optimization algorithm,namely:simulated annealing algorithm and genetic algorithm,optimization design of the study and glaze formulation,but above all methods have their own advantages and disadvantages,jointly decided by the specific market conditions,glaze and glaze ingredients to select what kind of method,and many other factors.Standard for the cost of the ratio of in this paper,the author chooses glaze composition,glaze,glaze,glaze,from the above four aspects considered glaze formula relations.finally,by use the artificial neural networks(self organizing feature map(SOFM),study of various glaze formula optimization decision,so as to determine the best glaze formula,due to the self organizing feature map neural network(SOFM)on the formula optimization of has a strong applicability,so according to the results of clustering can choose the best formula.