论文部分内容阅读
为了提高对空气质量预测的准确性,提出了一种基于混沌遗传算法(CGA)的BP神经网络改进方法。BP神经网络是目前应用最广泛的神经网络,但存在收敛速度慢和易陷入极小值的缺陷。该改进算法的基本思想是用混沌遗传算法优化BP神经网络的初始权值和阈值。混沌遗传算法结合了混沌运动的遍历性和遗传算法的反演性。将混沌变量加入遗传算法中,进一步提高了遗传算法的全局搜索能力和收敛速度;将混沌遗传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。利用改进后的CGA-BP算法进行空气质量预测,结果表明,该方法对空气质量的预测效果明显好于单纯使用BP神经网络的预测效果。
In order to improve the accuracy of air quality prediction, an improved BP neural network based on chaos genetic algorithm (CGA) is proposed. BP neural network is the most widely used neural network at present, but it has some drawbacks such as slow convergence speed and easy to fall into minimum value. The basic idea of the improved algorithm is to use chaos genetic algorithm to optimize the initial weights and thresholds of BP neural network. Chaos genetic algorithm combines the ergodicity of chaos and the inversion of genetic algorithm. The chaos variable is added to the genetic algorithm to further improve the global search ability and the convergence speed of the genetic algorithm. The optimal solution obtained from the chaos genetic algorithm is used as the initial weight and the threshold of the BP neural network. The improved CGA-BP algorithm is used to predict the air quality. The results show that the proposed method can predict the air quality better than the BP neural network.