An Electronic Tongue Based on Genetic Algorithm Improving BP Neural Network

Electronic tongue is a device which is used to classify different taste by multi-sensor. In this work, we had measured the production of chemical composition of five different mineral water by four kinds of selected ion array (sensitive to H⁺, Na⁺, Ca²⁺ and K⁺, respectively). Principal component analysis, a kind of multivariate data analysis was used to educing of total number of the sensors in the array. An adaptive genetic BP neural network is used as a classifier. Compared to other recognition methods, an adaptive genetic algorithm is used to optimize the BP network initial weight first, and to carry out the BP network training process. The application results show that the performance of the proposed method has surpasses the traditional BP algorithm, can improve convergence and the learning capability of the network, and make the electronic tongue has a higher aggregate classification rate.