MACHINE VISION FOR AUTOMATICVISUAL INSPECTION OF WOODENRAILWAY SLEEPERS USING UNSUPERVISED NEURAL NETWORKS

The motivation for this thesis work is the need for improving reliability of equipment and quality of service to railway passengers as well as a requirement for cost-effective and efficient condition maintenance management for rail transportation. This thesis work develops a fusion of various machine vision analysis methods to achieve high performance in automation of wooden rail track inspection.The condition monitoring in rail transport is done manually by a human operator where people rely on inference systems and assumptions to develop conclusions. The use of conditional monitoring allows maintenance to be scheduled, or other actions to be taken to avoid the consequences of failure, before the failure occurs. Manual or automated condition monitoring of materials in fields of public transportation like railway, aerial navigation, traffic safety, etc, where safety is of prior importance needs non-destructive testing (NDT).In general, wooden railway sleeper inspection is done manually by a human operator, by moving along the rail sleeper and gathering information by visual and sound analysis for examining the presence of cracks. Human inspectors working on lines visually inspect wooden rails to judge the quality of rail sleeper. In this project work the machine vision system is developed based on the manual visual analysis system, which uses digital cameras and image processing software to perform similar manual inspections. As the manual inspection requires much effort and is expected to be error prone sometimes and also appears difficult to discriminate even for a human operator by the frequent changes in inspected material. The machine vision system developed classifies the condition of material by examining individual pixels of images, processing them and attempting to develop conclusions with the assistance of knowledge bases and features.A pattern recognition approach is developed based on the methodological knowledge from manual procedure. The pattern recognition approach for this thesis work was developed and achieved by a non destructive testing method to identify the flaws in manually done condition monitoring of sleepers.In this method, a test vehicle is designed to capture sleeper images similar to visual inspection by human operator and the raw data for pattern recognition approach is provided from the captured images of the wooden sleepers. The data from the NDT method were further processed and appropriate features were extracted.The collection of data by the NDT method is to achieve high accuracy in reliable classification results. A key idea is to use the non supervised classifier based on the features extracted from the method to discriminate the condition of wooden sleepers in to either good or bad. Self organising map is used as classifier for the wooden sleeper classification.In order to achieve greater integration, the data collected by the machine vision system was made to interface with one another by a strategy called fusion. Data fusion was looked in at two different levels namely sensor-level fusion, feature- level fusion. As the goal was to reduce the accuracy of the human error on the rail sleeper classification as good or bad the results obtained by the feature-level fusion compared to that of the results of actual classification were satisfactory.

Energy Efficiency through Thermal Energy Storage : Possibilities for the Swedish Building Stock

The need for heating and cooling in buildings constitutes a considerable part of the total energy use in a country and reducing this need is of outmost importance in order to reach national and international goals for reducing energy use and emissions. One important way of reaching these goals is to increase the proportion of renewable energy used for heating and cooling of buildings. Perhaps the largest obstacle with this is the often occurring mismatch between the availability of renewable energy and the need for heating or cooling, hindering this energy to be used directly. This is one of the problems that can be solved by using thermal energy storage (TES) in order to save the heat or cold from when it is available to when it is needed. This thesis is focusing on the combination of TES techniques and buildings to achieve increased energy efficiency for heating and cooling. Various techniques used for TES as well as the combination of TES in buildings have been investigated and summarized through an extensive literature review. A survey of the Swedish building stock was also performed in order to define building types common in Sweden. Within the scope of this thesis, the survey resulted in the selection of three building types, two single family houses and one office building, out of which the two residential buildings were used in a simulation case study of passive TES with increased thermal mass (both sensible and latent). The second case study presented in the thesis is an evaluation of an existing seasonal borehole storage of solar heat for a residential community. In this case, real measurement data was used in the evaluation and in comparisons with earlier evaluations. The literature reviews showed that using TES opens up potential for reduced energy demand and reduced peak heating and cooling loads as well as possibilities for an increased share of renewable energy to cover the energy demand. By using passive storage through increased thermal mass of a building it is also possible to reduce variations in the indoor temperature and especially reduce excess temperatures during warm periods, which could result in avoiding active cooling in a building that would otherwise need it. The analysis of the combination of TES and building types confirmed that TES has a significant potential for increased energy efficiency in buildings but also highlighted the fact that there is still much research required before some of the technologies can become commercially available. In the simulation case study it was concluded that only a small reduction in heating demand is possible with increased thermal mass, but that the time with indoor temperatures above 24 °C can be reduced by up to 20%. The case study of the borehole storage system showed that although the storage system worked as planned, heat losses in the rest of the system as well as some problems with the system operation resulted in a lower solar fraction than projected. The work presented within this thesis has shown that TES is already used successfully for many building applications (e.g. domestic hot water stores and water tanks for storing solar heat) but that there still is much potential in further use of TES. There are, however, barriers such as a need for more research for some storage technologies as well as storage materials, especially phase change material storage and thermochemical storage.

Energy Efficiency through Thermal Energy Storage - Evaluation of the Possibilities for the Swedish Building Stock, Phase 1

As a first step in assessing the potential of thermal energy storage in Swedish buildings, the current situation of the Swedish building stock and different storage methods are discussed in this paper. Overall, many buildings are from the 1960’s or earlier having a relatively high energy demand, creating opportunities for large energy savings. The major means of heating are electricity for detached houses and district heating for multi dwelling houses and premises. Cooling needs are relatively low but steadily increasing, emphasizing the need to consider energy storage for both heat and cold. The thermal mass of a building is important for passive storage of thermal energy but this has not been considered much when constructing buildings in Sweden. Instead, common ways of storing thermal energy in Swedish buildings today is in water storage tanks or in the ground using boreholes, while latent thermal energy storage is still very uncommon.

Energy savings potential in existing houses : Energy simulation results for Dalarna's single-family housing stock

Vägar till en halverad energianvändning i Dalarnas byggnadsbestånd

Minimization of stock weight during close-die forging of a spindle

In this paper, Finite Element method and full-scale experiments have been used to study a hot forging method for fabri-cation of a spindle using reduced initial stock size. The forging sequence is carried out in two stages. In the first stage, the hot rolled cylindrical billet is pre-formed and pierced in a closed die using a spherical nosed punch to within 20 mm of its base. This process of piercing or impact extrusion leads to high strains within the work piece but requires high press loads. In the second stage, the resulting cylinder is placed in a die with a flange chamber and upset forged to form a flange. The stock mass is optimized for complete die filling. Process parameters such as effective strain distribution, material flow and forging load in different stages of the process are analyzed. It is concluded from the simulations that minor modifications of piercing punch geometry to reduce contact between the punch and emerging vertical walls of the cylinder appreciably reduces the piercing load. In the flange chamber, a die surfaces angle of 52° instead of 45° is pro-posed to ensure effective material flow and exert sufficient tool pressure to achieve complete cavity filling. In order to achieve better compression, it is also proposed to shorten both the length of the inserted punch and the die “tongues” by a few mm.