Search

Continuous advances of sensor technology and real-time computational capability are leading to data-rich environments to improve industrial automation and machine intelligence. When multiple signals are acquired from different sources (i.e. multi-channel signal data), two main issues must be faced: (i) the reduction of data dimensionality to make the overall signal analysis system efficient and actually applicable in industrial environments, and (ii) the fusion of all the sensor outputs to achieve a better comprehension of the process. In this frame, multi-way principal component analysis (PCA) represents a multivariate technique to perform both the tasks. The paper investigates two main multi-way extensions of the traditional PCA to deal with multi-channel signals, one based on unfolding the original data-set, and one based on multi-linear analysis of data in their tensorial form. The approaches proposed for data modelling are combined with appropriate control charting to achieve multi-channel profile data monitoring. The developed methodologies are demonstrated with both simulated and real data. The real data come from an industrial sensor fusion application in waterjet cutting, where different signals are monitored to detect faults affecting the most critical machine components. [ABSTRACT FROM PUBLISHER]

Byline: B.M. Colosimo, M. Monno, Q. Semeraro Recent researches propose the use of Water Jet technology for peening metallic components. Compared with traditional shot peening (SP), water jet peening (WJP) process presents the main advantage of a negligible influence on surface roughness and topography. Nevertheless, the transition from residual stress generation to the beginning of an erosion phenomenon is deeply influenced by an accurate selection of process parameters. Since by now parameter selection is performed by experimental trials, the paper focuses on the analysis of two main parameters the pressure and the traverse rate of the waterjet. The effect of these parameters on residual stress, roughness and erosion is investigated and empirical relationships which can explain these effects are presented.

We consider the problem of adjusting a machine that manufactures parts in batches or lots and experiences random offsets or shifts whenever a set-up operation takes place between lots. The existing procedures for adjusting set-up errors in a production process over a set of lots are based on the assumption of known process parameters. In practice, these parameters are usually unknown, especially in short-run production. Due to this lack of knowledge, adjustment procedures such as Grubbs' (1954, 1983) rules and discrete integral controllers (also called EWMA controllers) aimed at adjusting for the initial offset in each single lot, are typically used This paper presents an approach for adjusting the initial machine offset over a set of lots when the process parameters are unknown and are iteratively estimated using Markov Chain Monte Carlo (MCMC). As each observation becomes available, a Gibbs Sampler is run to estimate the parameters of a hierarchical normal means model given the observations up to that point in time. The current lot mean estimate is then used for adjustment. If used over a series of lots, the proposed method allows one eventually to start adjusting the offset before producing the first part in each lot. The method is illustrated with application to two examples reported in the literature. It is shown how the proposed MCMC adjusting procedure can outperform existing rules based on a quadratic off-target criterion. [ABSTRACT FROM AUTHOR]