A Fuzzy ART versus Hybrid NN-HMM methods for lithology identification in the Triasic province

We combine neural networks (NNs) and hidden Markov models (HMMs) techniques in order to obtain the lithology identification of wells situated in the Triasic province (Sahara). For the same aim, two systems based on adaptive resonance theory (ART), ART1 and fuzzy ART, are also developed. Our objective is to facilitate the work of the geological experts by permitting them to obtain quickly the structure and the nature of lands around the drilling. Lithology identification supplies qualitative information about rocks. Lithofacies represent rock textures and are important components of hydrocarbon reservoir characterisation. In this paper, we show that it is interesting to combine the respective capacities of the HMMs and NNs to produce a new effective hybrid models that draw their source in the two formalisms and can provide us a more reliable reservoir model. Comparisons are established to show that the results obtained by the NN-HMM hybrid system are close to those obtained by the fuzzy ART approach applied to the same borehole with the same well logs

[1]  Paul R. Cohen,et al.  Segmenting time series with a hybrid neural networks - hidden Markov model , 2002, AAAI/IAAI.

[2]  Marcelo M. Wanderley.  Proceedings of the 2003 conference on New interfaces for musical expression , 2003 .

[3]  Michael Cohen,et al.  Context-Dependent Connectionist Probability Estimation in a Hybrid HMM-Neural Net Speech Recognition , 1994 .

[4]  Stephen Grossberg,et al.  Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system , 1991, Neural Networks.

[5]  Salim Chikhi,et al.  Probabilistic neural method combined with radial-bias functions applied to reservoir characterization in the Algerian Triassic province , 2004 .

[6]  Ibrahim Matta,et al.  End-to-End Inference of Loss Nature in a Hybrid Wired/Wireless Environment , 2002 .

[7]  Afzal Ballim,et al.  A hybrid approach robust text analysis , 2007 .

[8]  James L. Flanagan,et al.  Telephone speech recognition using neural networks and hidden Markov models , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[9]  A. Waibel,et al.  Connectionist Viterbi training: a new hybrid method for continuous speech recognition , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[10]  Graham Russell,et al.  LHIP: Extended DCGs for Configurable Robust Parsing , 1994, COLING.

[11]  Yochai Konig,et al.  Combining Neural Networks And Hidden Markov Models For Continuous Speech Recognition , 1992 .

[12]  David Haussler,et al.  A Generalized Hidden Markov Model for the Recognition of Human Genes in DNA , 1996, ISMB.

[13]  Nicola Orio,et al.  Score Following: State of the Art and New Developments , 2003, NIME.