论文部分内容阅读
针对BP神经网络在大坝监测数据预测模型中后期预测精度不高的问题,基于小生境蚁群算法的智能搜索能力和强鲁棒性、BP神经网络对大量的输入-输出模式的非线性映射关系的学习存贮能力,将两种方法结合,用小生境蚁群算法优化BP神经网络的建模方法建立了水平位移观测数据的预测模型,并与ACA-BP神经网络和传统BP神经网络进行了对比分析。结果表明,本文方法可加快BP神经网络收敛速度、增强局部搜索能力,具有更高的预测精度。
Aiming at the low accuracy of BP neural network in the prediction model of dam monitoring data, based on the intelligent search ability and robustness of niche ant colony algorithm, the nonlinear mapping of BP neural network to a large number of input-output modes The relationship between the learning and storage capacity, the two methods combined with niche optimization of niche algorithm BP neural network modeling method to establish a horizontal displacement of the observed data prediction model, and with the ACA-BP neural network and the traditional BP neural network Comparative analysis. The results show that the proposed method can speed up the convergence of BP neural network and enhance the local search ability with higher prediction accuracy.