论文部分内容阅读
BP神经网络虽然是目前应用最广泛的神经网络模型,但其自身有易陷入局部极小值和收敛速度慢的缺点。本文提出了一种利用混沌粒子群算法来改进BP神经网络。该算法的基本思想是用混沌粒子群算法对BP神经网络的初始权值和初始阈值进行优化。对粒子群算法进行混沌优化,提高粒子群算法的全局搜索能力;用混沌粒子群算法优化后得到的最优解作为BP神经网络的初始权值和阈值。通过对非线性函数的拟合实验,发现改进后的结果与普通的BP神经网络的结果相比,具有更高的准确性,提高了拟合的精度。
Although BP neural network is the most widely used neural network model at present, it is easy to fall into local minimum and the convergence speed is slow. This paper presents a chaotic particle swarm optimization algorithm to improve BP neural network. The basic idea of this algorithm is to use chaos particle swarm optimization algorithm to optimize the initial weights and initial thresholds of BP neural network. The particle swarm optimization algorithm is chaos optimized to improve the global search ability of particle swarm optimization algorithm. The optimal solution obtained by the chaos particle swarm optimization is used as the initial weight and threshold of BP neural network. Through the fitting experiments on nonlinear functions, it is found that the improved results are more accurate than ordinary BP neural networks and the accuracy of the fitting is improved.