This paper aims to introduce a new method for the analysis of non-linear and non-stationary signals. This method is based on empirical mode decomposition technology by decomposing signals into multiple eigenfunctions. But there is modal aliasing problem in this method. We combine the auxiliary noise analysis method with the frequency analysis method to eliminate the modal aliasing. The modified EMD method is applied to the shock response signal analysis. By studying each eigenfunction, the components and characteristics of shock response acceleration signal are analyzed, which provides a reference for the analysis of shock signal of underwater explosion.
Under the underwater explosion load, the ship structure produces a response signal with a large bandwidth of frequency spectrum. The signal spectrum is generally complex. A better time-frequency analysis method is necessary to analyze the signal characteristics and obtain the real information of the signal. A new method based on the empirical mode decomposition (EMD) technology is proposed in this paper. The nonlinear signal analysis program is compiled based on this method. Vibration signal is decomposed into several eigenfunctions. The Hilbert-Huang transform method based on EMD is an advanced method, but EMD method can cause modal aliasing in time-frequency analysis. In this paper, the ensemble empirical mode decomposition (EEMD) method is used to solve this problem. Combined with auxiliary noise analysis method, white noise with uniform spectrum distribution is added to the analysis signal, so that white noise and signal are uniformly distributed. The signal components include the signal itself and the white noise signal. Because the mean value of the white noise signal is zero, the noise signals will counteract with each other after several iterative operations. Only the original signal itself is stable, and the mode aliasing is eliminated. In this paper, the response of the ship structure subjected to the underwater explosion load is simulated. Signal analysis program based on the EEMD method was compiled. The proposed program was used to analyze the structure response signal, and the time-frequency information of the signal was obtained. By studying each eigenfunction, the components and characteristics of the shock acceleration signal of structure response were analyzed. The proposed program provides a reference for the analysis and study of the shock signal of ships subjected to underwater explosion.