Aiming at the characteristics of the fault spectrum of industrial robots, a new phase difference correction method is proposed on the basis of Fourier transform, which combines autocorrelation technology and windowing technology to convert the original signal into a discrete spectrum with fault characteristics, which effectively improves the accuracy of fault spectrum correction and provides important help for robot fault diagnosis. Simulation analysis and example verification show that the new algorithm is quite effective in the extraction of industrial robot fault features, and the algorithm still has a smaller relative error than the traditional algorithm under noise conditions, with high estimation accuracy and strong compatibility and robustness. The algorithm not only has high theoretical value in pattern recognition, but also has great practical significance in engineering fields such as robot diagnosis.

针对工业机器人故障谱的特点，提出了一种基于傅里叶变换的相位差校正新方法，将自相关技术和加窗技术相结合，将原始信号转换为具有故障特征的离散谱，有效提高故障谱校正的准确性，为机器人故障诊断提供重要帮助。仿真分析和实例验证表明，新算法在工业机器人故障特征提取中效果不错，在噪声条件下，该算法仍比传统算法具有更小的相对误差，估计精度高，兼容性和鲁棒性强。该算法不仅在模式识别方面具有较高的理论价值，