Authors:
F. Müller
;
P. O'Leary
;
G. Rath
;
C. Kukla
;
M. Harker
;
T. Lucyshyn
and
C. Holzer
Affiliation:
Montanuniversitaet Leoben, Austria
Keyword(s):
Pattern Recognition, Gibbs Error, Discrete Fourier Spectrum, Orthogonal Complement, Covariance Propagation, Discrete Fourier Transform.
Related
Ontology
Subjects/Areas/Topics:
Applications and Uses
;
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Data Manipulation
;
Health Engineering and Technology Applications
;
Human-Computer Interaction
;
Industrial and Structural Monitoring
;
Methodologies and Methods
;
Multimedia/Audio Signal Processing
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
The production of high quality plastic parts requires in-mould sensors to monitor the injection moulding process.
A novel wireless sensor concept is presented where structure borne sound is used to transmit information
from the inside of an injection mould to the outside surface, eliminating the need for cabling within the mould.
The sound is acquired and analyzed using new algebraic basis function techniques to enable the detection of
temporal occurrence of frequency patterns in the presence of large levels of noise. The temporal occurrence
of the resonators represents the passing melt front. The reduction of spectral leakage is computed by an alternative
method using low degree Gram polynomials. The computation of the pattern matching algorithm
yields both the correlation coefficients and their covariances which are used to determine the certainty of the
measurement. The paper presents the used mathematical background as well as real measurements performed
on an injection moulding
machine. A test mould was equipped with two different resonant structures. Besides
calculating the correlation coefficients the 3s confidence interval of the coefficients is computed. With
the novel algebraic approach a reliable separation of the two temporal points of occurrence of the resonant
structures was computed.
(More)