基于遗传算法的BP神经网络模型在桩孔质量检测中的应用

Application of BP Neural Network Model Based on Genetic Algorithm in Pile Quality Inspection

目的将改进的神经网络模型应用于钻孔灌注桩桩孔质量的智能化识别,从而减少人为的误判、漏判情况.方法将遗传算法与神经网络模型有机地结合起来,建立桩孔质量检测的智能化模型,先利用遗传算法对神经网络的权值和阈值进行优化,再结合训练完成的神经网络模型对桩孔质量进行预测,同时根据现场数据建立三维分析图,通过预测结果与三维分析图的比对来验证模型的准确性.结果测试样本的仿真误差为0.005 75,训练样本的仿真误差为0.022 4;5、6号桩孔的预测结果为(0.001 2,0.999 9),(0.002 7,0.005 1),即5号桩质量为合格,6号桩质量为良好.结论通过预测结果与三维分析图的比对结果,可以得出基于遗传算法的神经网络模型能够较好地对孔灌注桩进行智能判别.

In this paper,the neural network model based on genetic algorithm is used to realize the intelligent identification of pile hole quality of bored pile,so as to reduce the human misjudgment and omission.This paper adopts the method of combining the genetic algorithm with the neural network model to establish the intelligent model of pile hole quality detection.Firstly,the genetic algorithm is used to optimize the weights and thresholds of the neural network,and then the neural The network model is used to predict the quality of the pile hole,and the three-dimensional analysis chart is established according to the field data.The accuracy of the model is verified by the comparison between the prediction result and the three-dimensional analysis.As a result,the simulation error of the test sample is 0.005 75,and the simulation error of the training sample is 0.022 41.And the prediction results of No.5,6 pile are coded as(0.001 2,0.999 9),(0.002 7,0.005 1).According to the result of coding,it can be concluded that No.5 pile is qualified and No.6 pile is good.Conclusion:Based on the comparison between the predicted results and the three-dimensional analysis,it can be concluded that the neural network model based on genetic algorithm can intelligently discriminate the pile-piles.