In order to identify the tilt direction of the self-mixing signals under weak feedback regime interfered by noise,a deep learning method is proposed.The one-dimensional U-Net(1D U-Net)neural network can identify the d...In order to identify the tilt direction of the self-mixing signals under weak feedback regime interfered by noise,a deep learning method is proposed.The one-dimensional U-Net(1D U-Net)neural network can identify the direction of the self-mixing fringes accurately and quickly.In the process of measurement,the measurement signal can be normalized and then the neural network can be used to discriminate the direction.Simulation and experimental results show that the proposed method is suitable for self-mixing interference signals with noise in the whole weak feedback regime,and can maintain a high discrimination accuracy for signals interfered by 5 dB large noise.Combined with fringe counting method,accurate and rapid displacement reconstruction can be realized.展开更多
A large-scale large eddy simulation in high performance personal computer clusters is carried out to present unsteady mixing mechanism of film cooling and the development of films. Simulation cases include a single-ho...A large-scale large eddy simulation in high performance personal computer clusters is carried out to present unsteady mixing mechanism of film cooling and the development of films. Simulation cases include a single-hole plate with the inclined angle of 30° and blowing ratio of 0.5, and a single-row plate with hole-spacing of 1.5D and 2D （diameters of the hole）. According to the massive simulation results, some new unsteady phenomena of gas films are found. The vortex system is changed in different position with the development of film cooling with the time marching the process of a single-row plate film cooling. Due to the mutual interference effects including mutual exclusion, a certain periodic sloshing and mutual fusion, and the structures of a variety of vortices change between parallel gas films. Macroscopic flow structures and heat transfer behaviors are obtained based on 20 million grids and Reynolds number of 28600.展开更多
文摘In order to identify the tilt direction of the self-mixing signals under weak feedback regime interfered by noise,a deep learning method is proposed.The one-dimensional U-Net(1D U-Net)neural network can identify the direction of the self-mixing fringes accurately and quickly.In the process of measurement,the measurement signal can be normalized and then the neural network can be used to discriminate the direction.Simulation and experimental results show that the proposed method is suitable for self-mixing interference signals with noise in the whole weak feedback regime,and can maintain a high discrimination accuracy for signals interfered by 5 dB large noise.Combined with fringe counting method,accurate and rapid displacement reconstruction can be realized.
基金partially supported by the National Science and Technology Major Project(2013CB035700)the National Natural Science Foundation of China(11672225,11511130053)the Funds for the Central Universities(xjj2014135)
文摘A large-scale large eddy simulation in high performance personal computer clusters is carried out to present unsteady mixing mechanism of film cooling and the development of films. Simulation cases include a single-hole plate with the inclined angle of 30° and blowing ratio of 0.5, and a single-row plate with hole-spacing of 1.5D and 2D （diameters of the hole）. According to the massive simulation results, some new unsteady phenomena of gas films are found. The vortex system is changed in different position with the development of film cooling with the time marching the process of a single-row plate film cooling. Due to the mutual interference effects including mutual exclusion, a certain periodic sloshing and mutual fusion, and the structures of a variety of vortices change between parallel gas films. Macroscopic flow structures and heat transfer behaviors are obtained based on 20 million grids and Reynolds number of 28600.
