2.State Grid Anhui Electric Power Co., Ltd., Hefei,Anhui 230061, China;
3.Electric Power Research Institute of State Grid Anhui Electric Power Co., Ltd., Hefei,Anhui 230088, China)
Abstract:Through the acquisition, processing and characteristic analysis of voiceprint signal of large power transformer (reactor), this paper proposes a detection and recognition algorithm model and system of transformer working condition based on voiceprint recognition technology, which can judge and detect the working state of transformer. Firstly, 73 groups of transformer audio including 1800 minutes are collected by voice acquisition sensor. And then the noise of the collected transformer voiceprint is processed by segmenting, framing and windowing pretreatment methods. Secondly, the features of transformer voiceprint are extracted and effectively fused by using energy feature, frequency feature, Mel frequency cepstrum coefficient and frequency compression method. Finally, in order to solve the problem of additive superposition of sound from stable working condition and unstable instantaneous noise of transformer, this paper proposes a separation method of voice print superposition based on cosine similarity algorithm, and establishes transformer working condition detection and verification analysis system.
Key words:voiceprint recognition; transformer; working condition detection; verification system
[5] DAVIS K H, BIDDULPH R, BALASHEK S. Automatic recognition of spoken digits[J]. Journal of the Acoustical Society of America, 1952, 24(6): 637.
[6] RABINER L R. A tutorial on hidden Markov models and selected applications in speech recognition[J]. Readings in Speech Recognition, 1990,77(2):267-296.
[7] LEE K F, HON H W. An overview of the SPHINX speech recognition system[J]. IEEE Transactions on Acoustics, Speech,and Signal Processing Speech, 1990, 38(1):35-45.
[8] HINTON G E, OSINDERO S, TEH Y W. A fastlearning algorithm for deep belief nets[J]. Neural Computation, 2006,18(7):1527-1554.
[9] MOHAMED A, DAHL G, HINTON G E. Deep belief networks for phone recognition[C]//Nips Workshop on Deep Learning for Speech Recognition and Related Applications. 2009,1(9): 39.
[10]YU D, DENG L. Deep learning and its applications to signal and information processing[J]. IEEE Signal Processing Magazine, 2011,28(1):145-154.
[11]DAHL G E, YU D, DENG L, et al. Contextdependent pretrained deep neural networks for large vocabulary speech recognition[J]. IEEE Transactions on Audio, Speech and Language Processing, 2012,20(1): 30-42.
[12]CHINTON G, DENG L, YU D, et al. Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups[J]. IEEE Signal Processing Magazine, 2012,29(6): 82-97.
[13]ABDELHAMID O, MOHAMED A R, JIANG H, et al. Convolutional neural networks for speech recognition[J]. IEEE/ACM Transactions on Audio,Speech,and Language Processing, 2014,22(10): 1533-1545.