小波分析与神经网络在钢丝绳断丝处理中的应用

Application of Wavelet Analysis and Neural Network under Handling Wire Rope Broken Condition

钢丝绳断丝检测信号中存在大量的噪声信号。在分析了钢丝绳断丝信号的特征后,利用小波分析算法的高分辨率特点,对钢丝绳断丝检测信号进行分解和重构,提取断丝特征信号;并采用基于BP神经网络算法的断丝识别,解决了断丝识别困难的问题。引入Matlab仿真软件对其进行验证,仿真结果表明,该方法对钢丝绳断丝信号的检测和识别十分有效,减小了钢丝绳断丝的误判率,提高了钢丝绳断丝检测的智能化程度。该方法成本低、效率高,具有一定的应用开发前景。

Normally, large amount of noise signals exist in broken wire detection for wire ropes. On the basis of analysis on the features of broken signals, by adopting high-resolution wavelet analysis algorithm, the broken wire detection signals are decomposed and rebuilt to extract eigen-signal of the broken wire. By adopting the broken wire recognition based on BP neural network, the difficulty in identifying broken wire has been overcome. The Matlab simulation software is introduced for verifying, the simulation indicates that the method is high effective. Thus the misjudging ratio is reduced and the intelligent level in broken wire detection is increased. The method is low-cost, highly efficient and worth to be expanded.