Support vector machine approach for the switching transient peak overvoltages during line energization

At the time of restoration transmission line switching is one of the major causes, which creates transient overvoltages. Though detailed Electro Magnetic Transient studies are carried out extensively for the planning and design of transmission systems, such studies are not common in a day-today operation of power systems. However it is important for the operator to ensure during restoration of supply that peak overvoltages resulting from the switching operations are well within safe limits. This paper presents a support vector machine approach to classify the various cases of line energization in the category of safe or unsafe based upon the peak value of overvoltage at the receiving end of line. Operator can define the threshold value of voltage to assign the data pattern in either of the class. For illustration of proposed approach the power system used for switching transient peak overvoltages tests is a 400 kV equivalent system of an Indian southern gri

Machine recognition of online handwritten Devanagari characters

In this paper, we describe a system for the automatic recognition of isolated handwritten Devanagari characters obtained by linearizing consonant conjuncts. Owing to the large number of characters and resulting demands on data acquisition, we use structural recognition techniques to reduce some characters to others. The residual characters are then classified using the subspace method. Finally the results of structural recognition and feature-based matching are mapped to give final output. The proposed system Ifs evaluated for the writer dependent scenario.

Site characterization model using least-square support vector machine and relevance vector machine based on corrected SPT data (N-c)

Statistical learning algorithms provide a viable framework for geotechnical engineering modeling. This paper describes two statistical learning algorithms applied for site characterization modeling based on standard penetration test (SPT) data. More than 2700 field SPT values (N) have been collected from 766 boreholes spread over an area of 220 sqkm area in Bangalore. To get N corrected value (N,), N values have been corrected (Ne) for different parameters such as overburden stress, size of borehole, type of sampler, length of connecting rod, etc. In three-dimensional site characterization model, the function N-c=N-c (X, Y, Z), where X, Y and Z are the coordinates of a point corresponding to N, value, is to be approximated in which N, value at any half-space point in Bangalore can be determined. The first algorithm uses least-square support vector machine (LSSVM), which is related to aridge regression type of support vector machine. The second algorithm uses relevance vector machine (RVM), which combines the strengths of kernel-based methods and Bayesian theory to establish the relationships between a set of input vectors and a desired output. The paper also presents the comparative study between the developed LSSVM and RVM model for site characterization. Copyright (C) 2009 John Wiley & Sons,Ltd.

Minimizing total weighted tardiness on a batch-processing machine with non-agreeable release times and due dates

This study considers the scheduling problem observed in the burn-in operation of semiconductor final testing, where jobs are associated with release times, due dates, processing times, sizes, and non-agreeable release times and due dates. The burn-in oven is modeled as a batch-processing machine which can process a batch of several jobs as long as the total sizes of the jobs do not exceed the machine capacity and the processing time of a batch is equal to the longest time among all the jobs in the batch. Due to the importance of on-time delivery in semiconductor manufacturing, the objective measure of this problem is to minimize total weighted tardiness. We have formulated the scheduling problem into an integer linear programming model and empirically show its computational intractability. Due to the computational intractability, we propose a few simple greedy heuristic algorithms and meta-heuristic algorithm, simulated annealing (SA). A series of computational experiments are conducted to evaluate the performance of the proposed heuristic algorithms in comparison with exact solution on various small-size problem instances and in comparison with estimated optimal solution on various real-life large size problem instances. The computational results show that the SA algorithm, with initial solution obtained using our own proposed greedy heuristic algorithm, consistently finds a robust solution in a reasonable amount of computation time.

Adhesion,Presence and Antifouling of Deinococcus geothermalis in Paper Machine Environment

This thesis has two items: biofouling and antifouling in paper industry. Biofouling means unwanted microbial accumulation on surfaces causing e.g. disturbances in industrial processes, contamination of medical devices or of water distribution networks. Antifouling focuses on preventing accumulation of the biofilms in undesired places. Deinococcus geothermalis is a pink-pigmented, thermophilic bacterium, and extremely resistant towards radiation, UV-light and desiccation and known as a biofouler of paper machines forming firm and biocide resistant biofilms on the stainless steel surfaces. The compact structure of biofilm microcolonies of D. geothermalis E50051 and the adhesion into abiotic surfaces were investigated by confocal laser scanning microscope combined with carbohydrate specific fluorescently labelled lectins. The extracellular polymeric substance in D. geothermalis microcolonies was found to be a composite of at least five different glycoconjugates contributing to adhesion, functioning as structural elements, putative storages for water, gliding motility and likely also to protection. The adhesion threads that D. geothermalis seems to use to adhere on an abiotic surface and to anchor itself to the neighbouring cells were shown to be protein. Four protein components of type IV pilin were identified. In addition, the lectin staining showed that the adhesion threads were covered with galactose containing glycoconjugates. The threads were not exposed on planktic cells indicating their primary role in adhesion and in biofilm formation. I investigated by quantitative real-time PCR the presence of D. geothermalis in biofilms, deposits, process waters and paper end products from 24 paper and board mills. The primers designed for doing this were targeted to the 16S rRNA gene of D. geothermalis. We found D. geothermalis DNA from 9 machines, in total 16 samples of the 120 mill samples searched for. The total bacterial content varied in those samples between 107 to 3 ×1010 16S rRNA gene copies g-1. The proportion of D. geothermalis in those same samples was minor, 0.03 1.3 % of the total bacterial content. Nevertheless D. geothermalis may endanger paper quality as its DNA was shown in an end product. As an antifouling method towards biofilms we studied the electrochemical polarization. Two novel instruments were designed for this work. The double biofilm analyzer was designed for search for a polarization program that would eradicate D. geothermalis biofilm or from stainless steel under conditions simulating paper mill environment. The Radbox instrument was designed to study the generation of reactive oxygen species during the polarization that was effective in antifouling of D. geothermalis. We found that cathodic character and a pulsed mode of polarization were required to achieve detaching D. geothermalis biofilm from stainless steel. We also found that the efficiency of polarization was good on submerged, and poor on splash area biofilms. By adding oxidative biocides, bromochloro-5,5-dimethylhydantoin, 2,2-dibromo-2-cyanodiacetamide or peracetic acid gave additive value with polarization, being active on splash area biofilms. We showed that the cathodically weighted pulsed polarization that was active in removing D. geothermalis was also effective in generation of reactive oxygen species. It is possible that the antifouling effect relied on the generation of ROS on the polarized steel surfaces. Antifouling method successful towards D. geothermalis that is a tenacious biofouler and possesses a high tolerance to oxidative stressors could be functional also towards other biofoulers and applicable in wet industrial processes elsewhere.

Cluster based core vector machine

Core Vector Machine(CVM) is suitable for efficient large-scale pattern classification. In this paper, a method for improving the performance of CVM with Gaussian kernel function irrespective of the orderings of patterns belonging to different classes within the data set is proposed. This method employs a selective sampling based training of CVM using a novel kernel based scalable hierarchical clustering algorithm. Empirical studies made on synthetic and real world data sets show that the proposed strategy performs well on large data sets.

Classification of magnetic resonance brain images using wavelets as input to support vector machine and neural network

In this paper. we propose a novel method using wavelets as input to neural network self-organizing maps and support vector machine for classification of magnetic resonance (MR) images of the human brain. The proposed method classifies MR brain images as either normal or abnormal. We have tested the proposed approach using a dataset of 52 MR brain images. Good classification percentage of more than 94% was achieved using the neural network self-organizing maps (SOM) and 98% front support vector machine. We observed that the classification rate is high for a Support vector machine classifier compared to self-organizing map-based approach.

Scalable non-linear Support Vector Machine using hierarchical clustering

This paper discusses a method for scaling SVM with Gaussian kernel function to handle large data sets by using a selective sampling strategy for the training set. It employs a scalable hierarchical clustering algorithm to construct cluster indexing structures of the training data in the kernel induced feature space. These are then used for selective sampling of the training data for SVM to impart scalability to the training process. Empirical studies made on real world data sets show that the proposed strategy performs well on large data sets.

An SIMD machine for low-level vision

This paper presents an SIMD machine which has been tuned to execute low-level vision algorithms employing the relaxation labeling paradigm. Novel features of the design include: 1. (1) a communication scheme capable of window accessing under a single instruction. 2. (2) flexible I/O instructions to load overlapped data segments; and 3. (3) data-conditional instructions which can be nested to an arbitrary degree. A time analysis of the stereo correspondence problem, as implemented on a simulated version of the machine using the probabilistic relaxation technique, shows a speed up of almost N2 for an N × N array of PEs.

DTC Technique for Induction Machine based Integrated Starter-Generator of an Automobile

This paper mainly concentrates on the application of the direct torque control (DTC) technique for the induction machine based integrated startergenerator (ISG) for automobile applications. It also discusses in brief about the higher DC bus voltage requirements in the automobiles i.e. present 14V system vs. 42V system to meet the power requirements, modes of operation of ISG, electric machine and the drive selection for the ISG,description of DTC technique, simulation and experimental results, and implementation.

Experimental estimation of friction force in lubricated cutting of steel

Friction force generated in lubricated cutting of steel is experimentally estimated by recording the tangential force experienced by the spherical face of a pin rubbing against a freshly cut surface. The pin and the cutting tool are both submerged in the lubricant and the pin is situated on the cut-track to record the force. The recording shows an instantaneous achievement of a peak in the force curve followed by a decline in time to a steady state value. The peak and not the steady state friction was found to be sensitive to the structure of the hydrocarbon and addition of additive to the oil. The configuration was designed and tested to demonstrate the influence of a reaction film which develops during cutting, on cutting tool friction. Given the strong correlation between the peak friction and the existence of a tribofilm in the cutting zone, the configuration is used to determine the lower limit of a cutting speed regime, which marks the initiation of lubricant starvation, in cutting of steel using an emulsion as a cutting fluid. (C) 2010 Elsevier B.V. All rights reserved.

Hila- ja kuviotason laskennan vaikutus kuvion kokonaistilavuuteen, tallennustilaan ja laskentaaikaan teoreettista läpimittajakaumaa ja runkolukusarjaa käyttämällä

Metsäsuunnittelussa tarvittavan metsävaratiedon keräämisessä ollaan Suomessa siirtymässä kuvioittaisesta arvioinnista laserkeilaus- ja ilmakuvapohjaiseen kaukokartoitukseen. Tämän tutkimuksen tarkoitus oli selvittää kuvion kokonaistilavuuden ja läpimittajakauman ennustamisen tarkkuus koealan metsikkö- ja puustotunnuksista MSN-, PRM-, ML- ja FMM-menetelmiä sekä Weibull-jakaumaa hyödyntäen seuraavilla tavoilla: 1. PRM-menetelmällä hilatasolla, 2. PRMmenetelmällä kuviotasolla, 3. ML-menetelmällä hilatasolla ja 4. ML-menetelmällä kuviotasolla. Lisäksi kuvion kokonaistilavuuden ennustamisen tarkkuus selvitettiin hyödyntäen kuviolle tuotettua runkolukusarjaa. Tulokset laskettiin puulajikohtaisesti männylle, kuuselle, koivulle ja muille puulajeille. Puulajien tulokset laskettiin kuviotasolla yhteen. Lisäksi selvitettiin menetelmien laskenta-ajan ja tallennustilan tarve. Tutkimuksen aineistona käytettiin Hämeen ammattikorkeakoulun Evon toimipisteen metsistä mitattuja kiinteäsäteisiä ympyräkoealoja, joita oli 249 kappaletta. Hakkuukoneella mitattiin 12kuvion, joiden pinta-alat vaihtelivat välillä 0,2 – 1,94 hehtaaria, puustotiedot. Aluepohjaisen laserkeilausaineiston pulssin tiheys oli 1,8/m2 ja ilmakuvien pikselikoko 0,5 metriä. Kuvion kokonaistilavuus ennustettiin tai estimoitiin laserkeilaus- ja ilmakuva-aineiston piirteiden avulla koealojen puustotunnuksista. Tulokset laskettiin erikseen kaikille kuvioille ja kuvioille, joiden pinta-ala oli yli 0,5 hehtaaria. Yli 0,5 hehtaarin kuvioita oli 8 kappaletta. Kuvion hilojen naapureina käytettiin 1 - 10 koealaa. Menetelmästä ja naapurien määrästä riippuen kokonaistilavuuden suhteellinen RMSE ja harha vaihtelivat välillä 20,76 – 52,86 prosenttia ja -12,04 – 46,54 prosenttia. Vastaavat luvut yli 0,5 hehtaarin kuvioilla olivat 6,74 – 59,41 prosenttia ja -8,04 – 49,59 prosenttia. Laskenta-aika vaihteli menetelmien ja käytettyjen naapurien määrän mukaan voimakkaasti. Kehittyneemmällä ohjelmoinnilla ja ohjelmistolla laskenta-ajat voivat laskea merkittävästi. Tallennustila ei testatuilla menetelmillä ole rajoittava tekijä laajassakaan mittakaavassa. Läpimittajakauman perusteella PRM-menetelmä ennustaa puulajille erittäin kapean läpimittajakauman, jos koeala koostuu vain muutamasta lähes samankokoisesta puusta. Tämä vaikutti tuloksiin erityisesti menetelmällä PRM2.

Estimation of cutting volume with three inventory methods for harvest planning in Canadian boreal forests.

Two methods of pre-harvest inventory were designed and tested on three cutting sites containing a total of 197 500 m3 of wood. These sites were located on flat-ground boreal forests located in northwestern Quebec. Both methods studied involved scaling of trees harvested to clear the road path one year (or more) prior to harvest of adjacent cut-blocks. The first method (ROAD) considers the total road right-of-way volume divided by the total road area cleared. The resulting volume per hectare is then multiplied by the total cut-block area scheduled for harvest during the following year to obtain the total estimated cutting volume. The second method (STRATIFIED) also involves scaling of trees cleared from the road. However, in STRATIFIED, log scaling data are stratified by forest stand location. A volume per hectare is calculated for each stretch of road that crosses a single forest stand. This volume per hectare is then multiplied by the remaining area of the same forest stand scheduled for harvest one year later. The sum of all resulting estimated volumes per stand gives the total estimated cutting-volume for all cut-blocks adjacent to the studied road. A third method (MNR) was also used to estimate cut-volumes of the sites studied. This method represents the actual existing technique for estimating cutting volume in the province of Quebec. It involves summing the cut volume for all forest stands. The cut volume is estimated by multiplying the area of each stand by its estimated volume per hectare obtained from standard stock tables provided by the governement. The resulting total estimated volume per cut-block for all three methods was then compared with the actual measured cut-block volume (MEASURED). This analysis revealed a significant difference between MEASURED and MNR methods with the MNR volume estimate being 30 % higher than MEASURED. However, no significant difference from MEASURED was observed for volume estimates for the ROAD and STRATIFIED methods which respectively had estimated cutting volumes 19 % and 5 % lower than MEASURED. Thus the ROAD and STRATIFIED methods are good ways to estimate cut-block volumes after road right-of-way harvest for conditions similar to those examined in this study.