In modern industry, vibration, as a common physical phenomenon during equipment operation, contains rich information about the internal state of the equipment. By collecting, processing, and analyzing vibration data, abnormal situations of the equipment can be detected promptly to prevent potential failures. However, the vibration signals of industrial equipment often exhibit nonlinear and non-stationary characteristics. Traditional signal processing methods are difficult to accurately capture these complex features, resulting in insufficient accuracy and low computational efficiency. This study uses a complementary set empirical mode decomposition of vibration signals and introduces sample entropy to make an improvement. At the same time, a high-precision vibration signal acquisition and processing technology for industrial equipment monitoring is designed by using discrete wavelet transform to process vibration characteristics. The results showed that compared with other methods, the proposed algorithm has improved accuracy by 8.06% and 10.49%, reduced rejection rate and false acceptance rate by 3.72% and 5.83%, and significantly reduced computation time, proving its high accuracy and computational efficiency. The coefficient of determination of the proposed algorithm was 0.987, which was higher than that of the comparison method, proving its high-fitting performance in practical applications. This algorithm effectively improves the accuracy and efficiency of industrial equipment vibration signal monitoring, providing strong support for early fault diagnosis and stable operation of industrial equipment. This is of great significance for promoting the development of industrial intelligent manufacturing.
Wei et al. (Sun,) studied this question.