论文部分内容阅读
本文为Fourier系数的确定提出了两种适用于工程应用的求解法：遗传算法（GA）和人工神经网络（ANN）法。应用GA求解Fourier系数时，将Fourier系数作为解向量进行染色体编码，然后通过进化使得Fourier展开充分逼近原函数从而获得最佳解；Fourier级数通过三角变换后，能够用一个标准的三层前馈网络描述，其网络权与Fourier系数相对应，利用BP算法训练该网络即可确定Fourier系数。这两种算法只要求知道原函数的样本，因而具有工程应用价值。
In this paper, two solution methods for engineering applications are proposed for the determination of Fourier coefficients: Genetic Algorithm (GA) and Artificial Neural Network (ANN). When GA is used to solve Fourier coefficients, the Fourier coefficients are used as chromosome vectors to solve the chromosomes. Then the Fourier transform is used to obtain the optimal solution by fully approximating the original functions. Fourier transform can be used to generate a standard three-layer feedforward Network description, the network right and Fourier coefficient corresponding to the use of BP algorithm to train the network to determine the Fourier coefficient. These two algorithms only require to know the samples of the original function, and thus have engineering application value.