Gaussian Filtering Technique Research Articles

The climate of Atndalen

The climate of Atndalen Valley is described by data collected by the standard network of stations currently run by the Norwegian Meteorological institute. It consists of one weather station (Sorneset) with full instrumental equipment, and three precipitation stations with a reduced set of equipment. At the latter stations, precipitation, snow depth, and snow cover are observed. Most of the stations, including the weather station, are situated in the central part of the watershed, near Atnsjoen. The climate of the Atndalen is of a continental type with precipitation minimum in late winter or spring and maximum during summer. The mean annual precipitation is about 500 mm near Atnsjoen (701 m a.s.l.) for the 30 year normal period 1961–1990. At the meteorological station Sorneset the warmest month is July, 11.2 °C, while the coldest month is January, −9.9 °C, i.e. an annual amplitude of 21.1 °C. The mean cloud cover varies from 4.5 oktas in February to 5.4 oktas in July, September and October. The highest ratio of relative sunshine is about 50% in spring. The mean snow depth increases during winter and early spring and reaches its maximum of 68 cm in March. The snow cover disappears on 9 May ±12 days and establishes on 5 November ±18 days. Variations in precipitation (since 1904) and temperature (since 1864) were studied on a decadal time scale by Gaussian filtering technique, and the significance of trends on the 0.05 level were studied by the Mann–Kendall test. For the whole period no significant trend in annual precipitation was detected. The maximum value was located to the 1920s and the minimum value to the 1910s. Annual mean temperature has increased significantly since 1864, and the classical temperature optimum in the 1930s was surpassed in the 1990s. By adopting a sinus model including the first Fourier component, trends and variations in climatological periods as well as heat and frost sums were studied. The frost free period has since 1864 increased by 13 days within 100 years based on a linear trend line. Earlier passing dates in spring largely account for the increase. The length of the growth season also increased up to about 1950. The annual heat sum shows a linear increase of about 103 daydegrees per 100 years while the annual frost sum varies considerably from period to period and fitted badly with a linear model.

Mixture Principal Component Analysis Models for Process Monitoring

A mixture principal component analysis (MixPCA) model detector is proposed. The detector creates each cluster by aggregating input exemplars. The number of clusters in the data set is automatically determined by a Gaussian filter technique called heuristic smoothing clustering. The structure building contains two stages: (1) sizing each cluster on the basis of the clustering analysis and (2) setting up the multivariable statistic confidence levels for each cluster. After the clusters are identified, MixPCA, which is a group of PCA models, will then be created. If any input pattern is under normal operation, it would fall into the accepted region whose criteria are set on the basis of the principal component analysis (PCA), T2 and Q charts. In addition, the techniques of similarity measuring and merging between clusters are also developed for the new correlation structure of the process variables when new data sets are included. The proposed model development procedure is favorable because of its simplici...

High-accuracy wavelength-change measurement system based on a Wollaston interferometer, incorporating a self-referencing scheme

A novel wavelength-difference measurement scheme with a Wollaston prism is presented. By using a suitable reference wavelength, a small variation in the signal wavelength can be converted into a relatively larger change in the modulated wavelength, as a result of the so-called fringe beating effect, resulting in enhanced measurement sensitivity by use of autocorrelation and Gaussian filtering techniques. From the results of a simulation carried out, we observed a wavelength variation of 0.01 nm over 15 nm or 0.1 nm over 60 nm for a typical pair of laser diodes with wavelengths of 785 and 810 nm, and wavelength variations of 0.5 nm over 40 nm or 1 nm over 110 nm for 671-and 785-nm wavelengths. These results were partially verified by the experimental results obtained for which a resolution of 0.01 nm over a range of 2.5 nm for the first pair and 0.5 nm over 4 nm for the second pair of laser diodes was seen. The results have applications to the determination of wavelength variations in a wavelength-division multiplexing system or measurement of the wavelength changes induced in a range of optical sensors.