A time series comprehensive evaluation method of acoustic emission signal based on information entropy weighting and its application of red bed soft rock compression failure critical characteristics study

To deeply understand the self-organization critical phenomenon of red bed soft rock in the process of compression failure, a uniaxial compression test was carried out on silt mudstone. The acoustic emission signal was collected simultaneously during the whole test process. The information entropy theory was applied to weigh the acoustic emission signal. The translation time domain was introduced to analyze the acoustic emission signal time series variation law. Finally, the time series comprehensive evaluation index value variation law of acoustic emission signal based on information entropy weighting was obtained. (1) During the compression loading process, with the increase of time, all the time series comprehensive evaluation index values showed an increasing trend with the saturated cycle increasing. When the time series comprehensive evaluation index value remains stable, it could be considered that the red bed soft rock enters a critical state. (2) The acoustic emission three-dimensional positioning results showed that when the compression failure reaches the critical point, the time series comprehensive evaluation index value would be the largest, and the coalescence crack also begins to form. When entering the critical state, the time series comprehensive evaluation index value would be in a stable state. (3) When the compression failure reached the critical point, the positioning signal would accumulate towards the rupture area until the sample was destroyed. The acoustic emission three-dimensional positioning results revealed the silt mudstone internal defects evolution law. These internal defects evolve from the slow and random development-disordered state to the instantaneous development-ordered state. These evolution laws corresponded well with the time series comprehensive evaluation index value variation trend. (4) During the compressive loading process, the energy probability density function satisfied the power-function distribution under different saturated periods. The research results are expected to provide a theoretical reference for the study of red bed soft rock compression failure critical damage characteristics.