The evolving tree-analysis and applications

In this paper, we enhance and analyze the Evolving Tree (ETree) data analysis algorithm. The suggested improvements aim to make the system perform better while still maintaining the simple nature of the basic algorithm. We also examine the system's behavior with many different kinds of tests, measurements and visualizations. We compare the ETree's performance against classical data analysis methods and very similar modern systems. We find that the ETree is a suitable method for unsupervised analysis of huge data sets.     

Advanced flaw production method for in-service inspection qualification mock-ups

One of the key issues in in-service inspection qualification is the representativeness of the defects used in qualification specimens. The best representativeness is achieved with realistic defects. However, present specimen production techniques have some significant weaknesses, such as unrealistic defects or additional alterations induced in the surrounding material. Specimens manufactured, for example, by weld implantation or with weld solidification defects always result in one or more extra weld interfaces. These interfaces can be detected by NDT. To overcome problems with the current specimens, a new defect manufacturing technique was developed. The new technique produces natural, representative defects without introducing additional weld metal or other unwanted alterations to the specimen.The new method enables artificial production of single, separate fatigue cracks by thermal loading. The method is based on a natural thermal fatigue damage mechanism and enables production of real cracks directly into the samples. Cracks are produced without welding or machining and without any preliminary surface treatment or artificial initiator such as a notch or a precrack. Single crack or a network of cracks can be induced into the base material, welded areas, HAZ, weld claddings, threaded areas, T-joints, etc. The location, orientation and size of produced cracks can be accurately controlled. Produced cracks can be used to simulate different types of service-induced cracks such as thermal fatigue, mechanical fatigue and stress corrosion cracks. It is shown that artificially produced thermal fatigue cracks correspond well with the real, service-induced cracks and overcome the problems of traditional qualification specimen manufacturing techniques.     

Nonlinear dynamical factor analysis for state change detection

Changes in a dynamical process are often detected by monitoring selected indicators directly obtained from the process observations, such as the mean values or variances. Standard change detection algorithms such as the Shewhart control charts or the cumulative sum (CUSUM) algorithm are often based on such first- and second-order statistics. Much better results can be obtained if the dynamical process is properly modeled, for example by a nonlinear state-space model, and then the accuracy of the model is monitored over time. The success of the latter approach depends largely on the quality of the model. In practical applications like industrial processes, the state variables, dynamics, and observation mapping are rarely known accurately. Learning from data must be used; however, methods for the simultaneous estimation of the state and the unknown nonlinear mappings are very limited. We use a novel method of learning a nonlinear state-space model, the nonlinear dynamical factor analysis (NDFA) algorithm. It takes a set of multivariate observations over time and fits blindly a generative dynamical latent variable model, resembling nonlinear independent component analysis. We compare the performance of the model in process change detection to various traditional methods. It is shown that NDFA outperforms the classical methods by a wide margin in a variety of cases where the underlying process dynamics changes.     

Wood inspection with non-supervised clustering

Abstract. The appearance of sawn timber has huge natural variations that the human inspector easily compensates for mentally when determining the types of defects and the grade of each board. However, for automatic wood inspection systems these variations are a major source for complication. This makes it difficult to use textbook methodologies for visual inspection. These methodologies generally aim at systems that are trained in a supervised manner with samples of defects and good material, but selecting and labeling the samples is an error-prone process that limits the accuracy that can be achieved. We present a non-supervised clustering-based approach for detecting and recognizing defects in lumber boards. A key idea is to employ a self-organizing map (SOM) for discriminating between sound wood and defects. Human involvement needed for training is minimal. The approach has been tested with color images of lumber boards, and the achieved false detection and error escape rates are low. The approach also provides a self-intuitive visual user interface. Received: 16 December 2000 / Accepted: 8 December 2001 Correspondence to: O. Silvén     

Energetics of Plastic Deformation of Metastable Austenitic Stainless Steel

The change in the internal energy during uniaxial tensile deformation of austenitic stainless steels EN 1.4301 (AISI 304) and EN 1.4318 (AISI 301LN) was determined by measuring the extent of γ→α'‐martensite transformation and the temperature increase of the samples. From the results the fraction of the stored energy of cold work and the free energy change related to the strain‐induced γ→α'‐martensite transformation were determined. The fraction of stored energy varied around 0.4. With the metastable steel grades the free energy change related to the γ→α'‐martensite transformation was found to vary between ‐98 MJ/m³ and ‐206 MJ/m³ depending on the austenite stability of the steel. Furthermore, the magnitude of the mechanical driving force was estimated by comparing the results with the free energy change of thermally induced transformation.     

Prevention of musculoskeletal disorders at work: Validation and reliability in a multicenter intervention study

During the years 1986–1989 The Finnish Work Environment Fund initiated a multicenter, intervention study (programme) to prevent musculoskeletal disorders at work. The programme consisted of four different projects comprising methodological, epidemiological, and interventive themes. The main results were as follows: (1) More than half of the occupational physiotherapists and physicians involved in the project felt that they did not have adequate training for testing the performance capacity of the musculoskeletal system. Several of the most commonly used musculoskeletal tests had either poor validity in predicting future musculoskeletal troubles or had poor reliability. (2) A controlled intervention study showed that active training of neck-shoulder muscles can prevent musculoskeletal troubles, even more than ergonomic actions. (3) It is possible to reduce sick-leave due to low-back disorders by intervention measures directed toward both the work (environment, tools) and the workers (working methods, positive attitude to work, preparedness to keep fit). (4) The identification of ergonomic hazards with the help of teamwork proved to be feasible. The practical improvements, however, were difficult to implement in the prevailing conditions. Some of the improvements increased the work load and the stress of the workers. This aspect needs further consideration.     

Application of a multiline material characterization method to inkjet printed electronics

Material characterization has become increasingly important with the adoption of simulation‐based workflow for microwave electronics design. This article focuses on the characterization of printable electronics materials using transmission‐line measurements and demonstrates the capability of the multiline characterization method to separate dielectric and conductor losses when line geometry and material properties vary between the lines. The effects of multiline algorithm, number of line standards, and methods for selecting optimal line lengths are demonstrated. Consistent characterization results are obtained from inkjet‐printed transmission lines on two different substrate materials. In addition, local conductor thickness variations are demonstrated as an effective way to decrease losses. Finally, the simulation‐based procedure for determining the material properties is outlined and applied to printable electronics characterization. © 2013 Wiley Periodicals, Inc. Int J RF and Microwave CAE 24:177–183, 2014.     

Increasing profit by improving gas handling systems

The availability of a smelter’s gas handling systems is crucial to the profitability of the plant. In new projects, the best technical and economic result is achieved if the gas handling systems are integrated into the process. This integration requires close cooperation between the smelter process designer and the gas handling designer. In modernization projects, environmentally, economically, and technically feasible solutions can be found, and smelter productivity can be increased when imagination and new technology are applied.     

Classification Based on Hybridization of Parametric and Nonparametric Classifiers

Parametric methods of classification assume specific parametric models for competing population densities (e.g., Gaussian population densities can lead to linear and quadratic discriminant analysis) and they work well when these model assumptions are valid. Violation in one or more of these parametric model assumptions often leads to a poor classifier. On the other hand, nonparametric classifiers (e.g., nearest-neighbor and kernel-based classifiers) are more flexible and free from parametric model assumptions. But, the statistical instability of these classifiers may lead to poor performance when we have small numbers of training sample observations. Nonparametric methods, however, do not use any parametric structure of population densities. Therefore, even when one has some additional information about population densities, that important information is not used to modify the nonparametric classification rule. This paper makes an attempt to overcome these limitations of parametric and nonparametric approaches and combines their strengths to develop some hybrid classification methods. We use some simulated examples and benchmark data sets to examine the performance of these hybrid discriminant analysis tools. Asymptotic results on their misclassification rates have been derived under appropriate regularity conditions.     

A study of 3D Network-on-Chip design for data parallel H.264 coding

In this paper, we implement, analyze and compare different Network-on-Chip (NoC) architectures aiming at higher efficiencies for MPEG-4/H.264 coding. Two-dimensional (2D) and three-dimensional (3D) NoCs based on Non-Uniform Cache Access (NUCA) are analyzed. We present results using a full system simulator with realistic workloads. Experiments show the average network latencies in two 3D NoCs are reduced by 28% and 34% respectively, comparing with 2D design. It is also shown that heat dissipation is a trade-off in improving performance of 3D chips. Our analysis and experiment results provide a guideline to design efficient 3D NoCs for data parallel H.264 coding applications.