引入粗糙集理论对钢丝绳断丝损伤信号原始特征值数据进行属性约简,去掉不必要的属性,并将粗糙集与神经网络有机结合,构造了优化的神经网络结构,进而识别钢丝绳的断丝损伤。试验表明,该方法有效提高了神经网络的性能及钢丝绳断丝损伤定量识别的速度和准确性。 Rough set theory is introduced to reduce the attribute of the original eigenvalue data of the broken wire rope damage signal, and the unnecessary attributes are removed. The rough set and neural network are combined organically to construct the optimized neural network structure and identify the broken wires . Experiments show that this method can effectively improve the performance of neural network and wire rope damage quantitative identification of the speed and accuracy.