A similarity-based approach to interpretation of sensor data using adaptive resonance theory

Abstract A machine methodology for generating qualitative interpretations (QIs) of 2-D sensor patterns is described. The approach enables a computer to interpret multisensor patterns under transient conditions at a level comparable to that of an experienced plant operator. Adaptive Resonance Theory introduced by Grossberg (1976a, b) is utilized with modification to provide human-like memory attributes. The methodology offers a more robust and adaptable means to interface symbolic knowledge-based systems with numeric plant operating systems. Exemplar-based, supervised learning is utilized to construct a high dimensional QI-map. During run-time, qualitative interpretations are generated for input patterns based on their location on this QI-map. Demonstration results for a dynamically simulated chemical process are presented.

[1]  J. R. Whiteley,et al.  Knowledge-based interpretation of sensor patterns , 1992 .

[2]  Stephen Grossberg,et al.  ART 3: Hierarchical search using chemical transmitters in self-organizing pattern recognition architectures , 1990, Neural Networks.

[3]  G. Stephanopoulos,et al.  Representation of process trends—Part II. The problem of scale and qualitative scaling , 1990 .

[4]  G. Stephanopoulos,et al.  Representation of process trends—Part I. A formal representation framework , 1990 .

[5]  Mark A. Kramer,et al.  Diagnosis using backpropagation neural networks—analysis and criticism , 1990 .

[6]  Stephen Grossberg,et al.  ARTMAP: supervised real-time learning and classification of nonstationary data by a self-organizing neural network , 1991, [1991 Proceedings] IEEE Conference on Neural Networks for Ocean Engineering.

[7]  Stephen Grossberg,et al.  A massively parallel architecture for a self-organizing neural pattern recognition machine , 1988, Comput. Vis. Graph. Image Process..

[8]  James F. Davis,et al.  Qualitative interpretation of sensor patterns , 1993, IEEE Expert.

[9]  Stephen Grossberg,et al.  Classical and Instrumental Learning by Neural Networks , 1982 .

[10]  Stephen Grossberg,et al.  The ART of adaptive pattern recognition by a self-organizing neural network , 1988, Computer.

[11]  S. Grossberg.  On learning and energy-entropy dependence in recurrent and nonrecurrent signed networks , 1969 .

[12]  Ingrid Daubechies,et al.  Ten Lectures on Wavelets , 1992 .

[13]  F. E. Finch,et al.  Qualitative modeling and fault diagnosis of dynamic processes by MIDAS , 1992 .

[14]  James Robert Whiteley.  Knowledge-based interpretation of process sensor patterns / , 1991 .

[15]  Stephen Grossberg,et al.  Art 2: Self-Organization Of Stable Category Recognition Codes For Analog Input Patterns , 1988, Other Conferences.

[16]  Stephen Grossberg,et al.  ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition , 1991, Neural Networks.

[17]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

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

[19]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[20]  Stanley C. Ahalt,et al.  Competitive learning algorithms for vector quantization , 1990, Neural Networks.

[21]  Venkat Venkatasubramanian,et al.  On the Classification Characteristics of Networks with Ellipsoidal Activation Functions: A Comparative Study , 1992 .