This paper proposes a speech enhancement method using the multi-scales and multi-thresholds of the auditory perception wavelet transform, which is suitable for a low SNR (signal to noise ratio) environment. This method achieves the goal of noise reduction according to the threshold processing of the human ear's auditory masking effect on the auditory perception wavelet transform parameters of a speech signal. At the same time, in order to prevent high frequency loss during the process of noise suppression, we first make a voicing decision based on the speech signals. Afterwards, we process the unvoiced sound segment and the voiced sound segment according to the different thresholds and different judgments. Lastly, we perform objective and subjective tests on the enhanced speech. The results show that, compared to other spectral subtractions, our method keeps the components of unvoiced sound intact, while it suppresses the residual noise and the background noise. Thus, the enhanced speech has better clarity and intelligibility.

The construction process of the tunnel ground deformation regularity of surrounding rock, stability, deformation control of tunnel surface based on the requirements, combined with the characteristics of shallow tunnel with large-span. Used three-dimensional numerical simulation software, established a dynamic tunnel analysis program to simulate the construction process of center cross-diagram method and double sidewall drift method. Based on the stratum deformation, supporting force and analysis of plastic zone distribution, comparative analysis of engineering adaptability of different construction methods from the construction process and construction mechanics, get optimization tunnel construction scheme.

Due to the importance of uranium and uranium alloys to national defence and nuclear industrial applications, it is necessary to understand dendrite formation in their solidification structures and to control their microstructures. In this study, a modified cellular automaton model was developed to predict 2-D and 3-D equiaxed dendrite growth in U-Nb alloys. The model takes into account solute diffusion, preferential growth orientation, interface curvature, etc., and the solid fraction increment is calculated using the local level rule method. Using this model, 2-D large-scale and 3-D equiaxed dendrite growth with various crystallographic orientations in the U-5.5Nb alloy were simulated, and the Nb micro-segregation behaviour during solidification was analysed. The simulated results showed reasonable agreement with the as-cast microstructure observed experimentally.