The invention discloses a ship propeller rotating speed forecasting method based on STFT periodic energy peaks, and relates to the technical field of ship navigational speed identification. The method comprises the following steps of: 1, carrying out STFT (Short Time Fourier Transform) on a propeller noise signal, and determining a window function and a window length by comparing and analyzing characteristic quantities in a time-frequency spectrogram; 2, performing STFT on the propeller noise signal by using the window function and the length parameter of the window function in the step 1 to obtain time-frequency spectrogram information; 3, extracting time-frequency characteristics of the propeller noise signals, and counting time nodes of periodic energy peaks appearing in a time-frequency graph; and 4, according to the obtained energy peak time node, a difference average value method is adopted to solve cycle time, and correlation analysis is carried out on the cycle time and the rotating speed of the propeller. The method utilizes the energy peak characteristics of the STFT spectrogram to forecast the rotating speed of the propeller, is a new thought for researching the noise characteristics of the propeller, and has important directivity value for the ship navigational speed identification technology.
本发明公开了一种基于STFT周期性能量峰的船舶螺旋桨转速预报方法,涉及船舶航速识别技术领域。本发明的步骤为:步骤一:将螺旋桨噪声信号进行STFT变换,通过比较分析时频谱图中特征量来确定窗函数和窗口长度;步骤二:利用步骤一的窗函数及其长度参数,对螺旋桨噪声信号进行STFT变换,获得时频谱图信息;步骤三:提取螺旋桨噪声信号的时频特征,并且统计时频图出现的周期性能量峰的时间节点;步骤四:根据所获得的能量峰时间节点,采用差值平均值法求得周期时间,并与螺旋桨转速进行相关性分析。本发明利用STFT谱图的能量峰特征来预报螺旋桨的转速,是研究螺旋桨噪声特征的一种新思路,对船舶航速识别技术有重要的指向性价值。
Ship propeller rotating speed forecasting method based on STFT periodic energy peak
一种基于STFT周期性能量峰的船舶螺旋桨转速预报方法
2022-06-03
Patent
Electronic Resource
Chinese
SHIP PROPELLER MOUNTING BRACKET, SHIP PROPELLER AND SHIP
European Patent Office | 2023
|