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.

Heat losses and thermal performance of commercial combined solar and pellet heating systems

Various pellet heating systems are marketed in Sweden, some of them in combination with a solar heating system. Several types of pellet heating units are available and can be used for a combined system. This article compares four typical combined solar and pellet heating systems: System 1 and 2 two with a pellet stove, system 3 with a store integrated pellet burner and system 4 with a pellet boiler. The lower efficiency of pellet heaters compared to oil or gas heaters increases the primary energy demand. Consequently heat losses of the various systems have been studied. The systems have been modeled in TRNSYS and simulated with parameters identified from measurements. For almost all systems the flue gas losses are the main heat losses except for system 3 where store heat losses prevail. Relevant are also the heat losses of the burner and the boiler to the ambient. Significant leakage losses are noticed for system 3 and 4. For buildings with an open internal design system 1 is the most efficient solution. Other buildings should preferably apply system 3. The right choice of the system depends also on whether the heater is placed inside or outside of the heated are. A large potential for system optimization exist for all studied systems, which when applied could alter the relative merits of the different system types.

Jämförelsestudie avseende stomsystem : Ramverk eller fackverk/balk/pelar-system

Projektet omfattade en jämförelsestudie avseende stomsystem. Studien undersökte ramverk och fackverk/balk/pelar-system och genomfördes självständigt med stöttning av Ramböll AB´s kontor i Falun. Syftet var främst att undersöka vilka skillnader det finns mellan tvåledsramar och fackverk/balk/pelar-system för lätta hallbyggnader och försöka få klarhet i varför fackverk/balk/pelar-system är det dominerande systemet i Sverige eftersom övriga Europa har tagit en annan utveckling och domineras av tvåledsramar. Studien undersöker skillnaderna mellan systemen i en hallbyggnad med förutbestämda mått i stål.Inledningsvis gjordes en litteraturstudie för att få en bredare bakgrund av systemen och en bättre förståelse om förutsättningarna för varje system. Efter litteraturstudien kunde sedan ett typhus och beräkningsunderlag för jämförelsestudien tas fram. Även en enkätstudie gjordes med syftet att skapa en tydlig bild av vilket stomsystem konstruktörer i Sverige oftast väljer och varför. Resultatet av studien visade att tvåledsramar ger en ökad kostnad jämfört med fackverk/balk/pelar-system i materialåtgång och framställning samt att beräkningarna blir mer komplicerade. Skulle fortsatta studier göras med dessa system i byggnader med andra mått skulle det kanske gå att få fram speciella mått på byggnader där kostnaden för tvåledsramar blir densamma som för fackverk/balk/pelar-system och därför är ett likvärdigt alternativ som stomsystem.En viktig slutsats från projektet är att tvåledsramar används mycket mer sällan än fackverk/balk/pelar-system som stomsystem i lätta hallbyggnader i Sverige på grund av att kostnaderna blir mycket högre med tvåledsramar och att det är ett mer komplicerat system i beräkningsarbetet. De viktigaste slutsatserna från jämförelsestudien går att sammafatta som följande: Tvåledsramar är dyrare att använda. Tvåledsramar är ett mer komplicerat system beräkningsmässigt. Traditionen av att använda tvåledsramar finns inte och därför används inte systemet.

The effectiveness of proactive telephone support provided to breastfeeding mothers of preterm infants : study protocol for a randomized controlled trial

Background: Although breast milk has numerous benefits for infants' development, with greater effects in those born preterm (at < 37 gestational weeks), mothers of preterm infants have shorter breastfeeding duration than mothers of term infants. One of the explanations proposed is the difficulties in the transition from a Neonatal Intensive Care Unit (NICU) to the home environment. A person-centred proactive telephone support intervention after discharge from NICU is expected to promote mothers' sense of trust in their own capacity and thereby facilitate breastfeeding. Methods/design: A multicentre randomized controlled trial has been designed to evaluate the effectiveness and cost-effectiveness of person-centred proactive telephone support on breastfeeding outcomes for mothers of preterm infants. Participating mothers will be randomized to either an intervention group or control group. In the intervention group person-centred proactive telephone support will be provided, in which the support team phones the mother daily for up to 14 days after hospital discharge. In the control group, mothers are offered a person-centred reactive support where mothers can phone the breastfeeding support team up to day 14 after hospital discharge. The intervention group will also be offered the same reactive telephone support as the control group. A stratified block randomization will be used; group allocation will be by high or low socioeconomic status and by NICU. Recruitment will be performed continuously until 1116 mothers (I: 558 C: 558) have been included. Primary outcome: proportion of mothers exclusively breastfeeding at eight weeks after discharge. Secondary outcomes: proportion of breastfeeding (exclusive, partial, none and method of feeding), mothers satisfaction with breastfeeding, attachment, stress and quality of life in mothers/partners at eight weeks after hospital discharge and at six months postnatal age. Data will be collected by researchers blind to group allocation for the primary outcome. A qualitative evaluation of experiences of receiving/providing the intervention will also be undertaken with mothers and staff. Discussion: This paper presents the rationale, study design and protocol for a RCT providing person-centred proactive telephone support to mothers of preterm infants. Furthermore, with a health economic evaluation, the cost-effectiveness of the intervention will be assessed. Trial registration: NCT01806480

Evaluation of a versatile energy auditing tool

Energy auditing can be an important contribution for identification and assessment of energy conservation measures (ECMs) in buildings. Numerous tools and software have been developed, with varying degree of precision and complexity and different areas of use.   This paper evaluates PHPP as a versatile, easy-to-use energy auditing tool and gives examples of how it has been compared to a dynamic simulation tool, within the EU-project iNSPiRe. PHPP is a monthly balance energy calculation tool based on EN13790. It is intended for assisting the design of Passive Houses and energy renovation projects and as guidance in the choice of appropriate ECMs.   PHPP was compared against the transient simulation software TRNSYS for a single family house and a multi-family house. It should be mentioned that dynamic building simulations might strongly depend on the model assumptions and simplifications compared to reality, such as ideal heating or real heat emission system. Setting common boundary conditions for both PHPP and TRNSYS, the ideal heating and cooling loads and demands were compared on monthly and annual basis for seven European locations and buildings with different floor area, S/V ratio, U-values and glazed area of the external walls.   The results show that PHPP can be used to assess the heating demand of single-zone buildings and the reduction of heating demand with ECMs with good precision. The estimation of cooling demand is also acceptable if an appropriate shading factor is applied in PHPP. In general, PHPP intentionally overestimates heating and cooling loads, to be on the safe side for system sizing. Overall, the agreement with TRNSYS is better in cases with higher quality of the envelope as in cold climates and for good energy standards. As an energy auditing tool intended for pre-design it is a good, versatile and easy-to-use alternative to more complex simulation tools.