输电杆塔振动识别方法的分析和应用.pdfVIP

A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Master in Engineering The research and application of method to identify vibration of transmission tower Master Canidate ： Zheng Lei Major ： Control theory and control engineering Supervisor ： Prof. Wang Bin Wuhan University of Science and Technology Wuhan, Hubei 430081, P.R.China May 20, 2014 万方数据 万方数据 摘 要 输电杆塔振动一直是影响输电网络正常运行的重要因素。输电杆塔振动的类 型主要有风激励（覆冰和无覆冰情况下）以及人为破坏引起的振动。目前，对输 电杆塔的振动分析主要采用快速傅立叶变换的方法得出其频率特性。本文在分析 杆塔振动频率特性的基础上，利用基于信号质量中心的方法求取振动信号的阻尼 比。利用以振动频率和阻尼比为输入参数的BP 神经网络模型对杆塔振动进行识 别。仿真实验证明：基于信号质量中心的振动信号阻尼比估计方法的精度明显高 于其他求取振动信号阻尼比方法的精度；利用随机顺序的训练样本对 BP 神经网 络进行训练，得到的模型的识别精度更高。物理实验证明：基于 BP 神经网络的 杆塔振动识别系统的实现能够有效地对输电杆塔振动进行综合分析。 关键词：输电杆塔；振动识别；神经网络；特征参数；信号质心 I 万方数据 Abstract Transmission tower vibration has long been an important factor affecting the normal operation of transmission network. The two main types of transmission tower vibration are wind excitation (ice and without ice) vibration caused by man-made sabotage. Currently, FFT is the most commonly used method to vibration analysis of transmission towers for drawing frequency characteristics. This paper aims to calculate the damping ratio of vibration signal using the approach of signal quality centeron based on analyzing the vibration frequency characteristics of the tower. Vibration frequency and damping ratio were further used as input parameters to identify the tower vibration for BP network model.