论文部分内容阅读
在桥梁工程领域大量使用预应力梁预制技术,其构件中现有预应力大小的识别对预应力结构和构件的性能检测鉴定起到至关重要的作用。初步探讨了基于自振频率测试与神经网络技术识别简支梁的预应力方法。通过简支梁模型试验,测得不同张拉力条件下的自振频率,面向BP神经网络识别技术,对测试数据进行处理。分别探讨了以部分试验数据构造神经网络训练样本,然后应用构造的BP网络来识别试验结果的方法。研究结果表明,就所试验研究的情况而言,80%的识别结果其精度在可接受范围,对于预应力预制构件的检测具有实用意义。
The extensive use of prestressed beam precasting technology in the field of bridge engineering, the identification of the existing prestressing size in its components plays a crucial role in the performance testing of prestressed structures and components. The method of prestressing simply-supported beam based on natural frequency test and neural network technology is preliminarily discussed. Through the simple beam model test, the natural frequencies under different tension conditions are measured, and the BP neural network identification technology is applied to process the test data. The training samples of neural network constructed with some experimental data are discussed separately, and then the BP network constructed is used to identify the experimental results. The results show that the accuracy of 80% of the recognition results is within the acceptable range for the experimental study, which is of practical significance for the detection of prestressed precast members.