论文部分内容阅读
针对模糊系统中规则结论为数值和线性函数的两种表示方式 ,找到了它们的共同点 ,将它们置于同一网络结构中 ,形成规则结论为数值和线性函数 (T -S模型 )的两种模糊神经网络 (FuzzyNeuralNetworks,简称FNN) ,导出了它们的网络模型及其学习算法。并首次将其应用于高强混凝土强度预测和配合比设计中。文章还介绍了一种简单有效地从样本数据中提取模糊规则及确定FNN参数初值的方法。运算结果表明 ,FNN不仅具有很高的预测精度 ,而且网络的结点和权值均具有明确的物理意义 ,可以借此深入分析高强混凝土综合性能与影响它们的因素之间的非线性关系
Aiming at the two representations of rule conclusion in fuzzy system which are numerical value and linear function, we find their common ground and place them in the same network structure to form two kinds of rule conclusion that are numerical and linear function (T -S model) Fuzzy Neural Networks (FNN), and derive their network models and their learning algorithms. And for the first time it is applied to the strength of high-strength concrete strength prediction and mix design. The article also introduced a simple and effective method of extracting fuzzy rules from sample data and determining initial values of FNN parameters. The calculation results show that FNN not only has high prediction accuracy, but also has clear physical meaning for the nodes and weights of the network, so that the nonlinear relationship between the comprehensive performance of high-strength concrete and the factors influencing them can be deeply analyzed