布谷鸟算法优化神经网络的站房结构损伤识别

Optimization of Neural Network by Cuckoo Algorithm for Identification of Station Structural Damage

研究目的:在铁路站房结构健康监测中,采用BP神经网络来识别结构损伤常常因为陷入局部极小值导致误报率高。因此,本文使用布谷鸟算法对神经网络进行优化,并将优化后的神经网络模型应用于站房结构损伤识别,以提高损伤识别的准确率。研究结论:(1)利用布谷鸟算法搜索BP神经网络最优的权重和偏置,可以找到模型的全局最优解,防止陷入局部极小值;(2)本文提出的模型适用于铁路站房结构健康监测数据量大、结构复杂、维数高的特点,可通过数据预处理构建神经网络的训练测试数据集;(3)以昆明南站监测数据为例进行试验验证,结果表明,通过布谷鸟算法优化BP神经网络模型准确率达到99.73%,较BP神经网络模型提高1.88个百分点;(4)本文研究可为实时结构性能评估提供技术支撑,可为铁路站房结构健康监测提供新的途径,具有一定的参考价值。

Research purposes:In structural health monitoring of railway station,BP neural network is used to identify structural damage,which often leads to high false positive rate because it is easy to fall into local minima.The cuckoo algorithm is used to optimize the neural network,and the optimized neural network model is applied to improve the accuracy of damage identification for station structural damage.Research conclusions:(1)Cuckoo algorithm is used to search the optimal weight and bias of BP neural network,so as to find the global optimal solution of the model and prevent the neural network from falling into the local minima.(2)The model proposed in this paper is suitable for the characteristics of large amount,complex structure,high dimension of railway station structure health monitoring data,and the training and testing datasets of neural network can be constructed by data preprocessing.(3)Taking the monitoring data of Kunming South Station as an example for experiment,the results show that the accuracy of BP neural network model optimized by cuckoo algorithm reaches 99.73%,which is 1.88 percentage points higher than that of BP neural network model.(4)The research can provide technical support for real-time structural state assessment,provide a new way for structural health monitoring of railway stations,and have a certain reference value.