改进粒子群优化神经网络的葡萄酒质量识别

Improved Wine Quality Recognition Based on Particle Swarm Optimization Neural Network

摘要: 随着我国经济的崛起，葡萄酒业也搭上了我国经济崛起的快速列车。葡萄酒产业规模不断壮大，但葡萄酒质量评定却没跟上酒业发展的脚步。现今的葡萄酒质量评定方法远远落后于市场需求。针对这个问题，本文用PSO优化算法代替BP网络自身训练过程，建立PSO优化BP网络模型，进而对葡萄酒质量进行分类评定。经过实证与文献的对比，PSO优化算法的确能够有效的代替BP神经网络自身训练过程。

Abstract: With the rise of China’s economy, the wine industry has also got on the fast train of China’s economic rise. The scale of wine industry has grown, but the quality evaluation of the wine has not kept pace with the development of the wine industry. Today’s wine quality assessment methods lag far behind market demand. Aiming at this problem, particle swarm optimization is used to replace the process of BP network self-training, and to establish the PSO optimization BP network model, and then classify and evaluate the quality of wine. Compared with the literature, particle swarm optimization can effectively replace the process of BP neural network training.