Real time video segmentation for recognising paint marks on bad wooden railway sleepers

Wooden railway sleeper inspections in Sweden are currently performed manually by a human operator; such inspections are based on visual analysis. Machine vision based approach has been done to emulate the visual abilities of human operator to enable automation of the process. Through this process bad sleepers are identified, and a spot is marked on it with specific color (blue in the current case) on the rail so that the maintenance operators are able to identify the spot and replace the sleeper. The motive of this thesis is to help the operators to identify those sleepers which are marked by color (spots), using an “Intelligent Vehicle” which is capable of running on the track. Capturing video while running on the track and segmenting the object of interest (spot) through this vehicle; we can automate this work and minimize the human intuitions. The video acquisition process depends on camera position and source light to obtain fine brightness in acquisition, we have tested 4 different types of combinations (camera position and source light) here to record the video and test the validity of proposed method. A sequence of real time rail frames are extracted from these videos and further processing (depending upon the data acquisition process) is done to identify the spots. After identification of spot each frame is divided in to 9 regions to know the particular region where the spot lies to avoid overlapping with noise, and so on. The proposed method will generate the information regarding in which region the spot lies, based on nine regions in each frame. From the generated results we have made some classification regarding data collection techniques, efficiency, time and speed. In this report, extensive experiments using image sequences from particular camera are reported and the experiments were done using intelligent vehicle as well as test vehicle and the results shows that we have achieved 95% success in identifying the spots when we use video as it is, in other method were we can skip some frames in pre-processing to increase the speed of video but the segmentation results we reduced to 85% and the time was very less compared to previous one. This shows the validity of proposed method in identification of spots lying on wooden railway sleepers where we can compromise between time and efficiency to get the desired result.

Machine vision for the automatic classification of images acquired from Non-destructive tests

This project is based on Artificial Intelligence (A.I) and Digital Image processing (I.P) for automatic condition monitoring of sleepers in the railway track. Rail inspection is a very important task in railway maintenance for traffic safety issues and in preventing dangerous situations. Monitoring railway track infrastructure is an important aspect in which the periodical inspection of rail rolling plane is required.Up to the present days the inspection of the railroad is operated manually by trained personnel. A human operator walks along the railway track searching for sleeper anomalies. This monitoring way is not more acceptable for its slowness and subjectivity. Hence, it is desired to automate such intuitive human skills for the development of more robust and reliable testing methods. Images of wooden sleepers have been used as data for my project. The aim of this project is to present a vision based technique for inspecting railway sleepers (wooden planks under the railway track) by automatic interpretation of Non Destructive Test (NDT) data using A.I. techniques in determining the results of inspection.

Gruvrätten vid Stora Kopparberget 1641-1682 : en undersökning över rannsakade brott och utdömda straff

The purpose of this essay is to examine and explain how the Swedish mining court of Stora Kopparberget (the Great Copper Mountain) implemented its judicial legislation between 1641-1682. Questions are asked about which counts of indictments the court tried, which sentences they handed out, in what quantities and how these results looks in comparison with other contemporary courts. The index cards of the court judicial protocols are the primary source of information. The methods are those of quantity- and comparative analysis.The results show that theft of copper ore was the most common crime ransacked by the court. Other common crimes were (in order): sin of omission, transgression of work directions, fights, slander and disdain, trade of stolen ore, failing appearance in court etc.Fines were by far the most common sentence followed by shorter imprisonments, gauntlets, loss of right to mine possession, twig beating, loss of work, penal servitude, banishment, “wooden horse riding” and finally military transcription. Even though previous re-search, in the field of Swedish specialized courts, is almost non existent evidence confirms great similarities between the Stora Kopparberget mining court and Sala mining court. This essay will, hopefully, enrich our knowledge of specialized courts, of 17th century mining industry and society and let us reach a broader understanding of the working conditions of the mountain.

Nonstationary feature extraction techniques for automatic classification of impact acoustic signals

Condition monitoring of wooden railway sleepers applications are generallycarried out by visual inspection and if necessary some impact acoustic examination iscarried out intuitively by skilled personnel. In this work, a pattern recognition solutionhas been proposed to automate the process for the achievement of robust results. Thestudy presents a comparison of several pattern recognition techniques together withvarious nonstationary feature extraction techniques for classification of impactacoustic emissions. Pattern classifiers such as multilayer perceptron, learning cectorquantization and gaussian mixture models, are combined with nonstationary featureextraction techniques such as Short Time Fourier Transform, Continuous WaveletTransform, Discrete Wavelet Transform and Wigner-Ville Distribution. Due to thepresence of several different feature extraction and classification technqies, datafusion has been investigated. Data fusion in the current case has mainly beeninvestigated on two levels, feature level and classifier level respectively. Fusion at thefeature level demonstrated best results with an overall accuracy of 82% whencompared to the human operator.

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.

Identifiering av parametrar till ackumulatortanken Solus1050 från consolar - en metodbeskrivning

This report describes the work done creating a computer model of a kombi tank from Consolar. The model was created with Presim/Trnsys and Fittrn and DF were used to identify the parameters. Measurements were carried out and were used to identify the values of the parameters in the model. The identifications were first done for every circuit separately. After that, all parameters are normally identified together using all the measurements. Finally the model should be compared with other measurements, preferable realistic ones. The two last steps have not yet been carried out, because of problems finding a good model for the domestic hot water circuit.The model of the domestic hot water circuit give relatively good results for low flows at 5 l/min, but is not good for higher flows. In the report suggestions for improving the model are given. However, there was not enough time to test this within the project as much time was spent trying to solve problems with the model crashing. Suggestions for improving the model for the domestic circuit are given in chapter 4.4. The improved equations that are to be used in the improved model are given by equation 4.18, 4.19 and 4.22.Also for the boiler circuit and the solar circuit there are improvements that can be done. The model presented here has a few shortcomings, but with some extra work, an improved model can be created. In the attachment (Bilaga 1) is a description of the used model and all the identified parameters.A qualitative assessment of the store was also performed based on the measurements and the modelling carried out. The following summary of this can be given: Hot Water PreparationThe principle for controlling the flow on the primary side seems to work well in order to achieve good stratification. Temperatures in the bottom of the store after a short use of hot water, at a coldwater temperature of 12°C, was around 28-30°C. This was almost independent of the temperature in the store and the DHW-flow.The measured UA-values of the heat exchangers are not very reliable, but indicates that the heat transfer rates are much better than for the Conus 500, and in the same range as for other stores tested at SERC.The function of the mixing valve is not perfect (see diagram 4.3, where Tout1 is the outlet hot water temperature, and Tdhwo and Tdhw1 is the inlet temperature to the hot and cold side of the valve respectively). The outlet temperature varies a lot with different temperatures in the storage and is going down from 61°C to 47°C before the cold port is fully closed. This gives a problem to find a suitable temperature setting and gives also a risk that the auxiliary heating is increased instead of the set temperature of the valve, when the hot water temperature is to low.Collector circuitThe UA-value of the collector heat exchanger is much higher than the value for Conus 500, and in the same range as the heat exchangers in other stores tested at SERC.Boiler circuitThe valve in the boiler circuit is used to supply water from the boiler at two different heights, depending on the temperature of the water. At temperatures from the boiler above 58.2°C, all the water is injected to the upper inlet. At temperatures below 53.9°C all the water is injected to the lower inlet. At 56°C the water flow is equally divided between the two inlets. Detailed studies of the behaviour at the upper inlet shows that better accuracy of the model would have been achieved using three double ports in the model instead of two. The shape of the upper inlet makes turbulence, that could be modelled using two different inlets. Heat lossesThe heat losses per m3 are much smaller for the Solus 1050, than for the Conus 500 Storage. However, they are higher than those for some good stores tested at SERC. The pipes that are penetrating the insulation give air leakage and cold bridges, which could be a major part of the losses from the storage. The identified losses from the bottom of the storage are exceptionally high, but have less importance for the heat losses, due to the lower temperatures in the bottom. High losses from the bottom can be caused by air leakage through the insulation at the pipe connections of the storage.

Grundläggande förutsättningar för solfångare med interna reflektorer : en kunskapsöversikt

Syftet med föreliggande rapport har varit att visa på de grundläggande egenskaperna för solfångare med interna reflektorer. Vidare tjänar rapporten syftet att ge en bild av dagsläget inom detta område och därigenom fungera som en utgångspunkt för forskning och utveckling kring solfångare med interna reflektorer för svenskt bruk. Arbetet har därför till stor del gått ut på att leta referenser genom tidskrifter och databaser.Den stora fördelen med CPC-solfångare, som är den klart dominerande typen av solfångare med interna reflektorer, består i dess låga värmeförluster, vilket gör dem attraktiva speciellt vid högre driftstemperaturer. De optiska egenskaperna hos olika typer av CPC-solfångare har grundligt studerats sedan mitten av 70-talet, medan studier av värmeförluster varit mer begränsad. Idag har forskningen och intresset för CPC-solfångare mattats av något, men fortsatt forskning pågår t ex i ett flertal länder, t ex USA, Israel, NordIrland och Japan. Endast en kommersiell tillverkare av CPC-solfångare har hittats (Portugal), vilken dock upphört p g a yttre ekonomiska faktorer. Japans CPC-teknologi anses stå närmast kommersiellt genombrott.Idag har utvecklingen av CPC-solfångare inriktats mot i huvudsak två koncept:1. Stationära lågkoncentrerande CPC-solfångare för produktion av värme i området 60-100 grader C. Dessa konstruktioner är ofta enkla och man försöker minimera kostnaderna för dessa konstruktioner genom att minimera behovet av t ex reflektorer och isolering. Syftet med dessa är att konkurera med plana solfångare vilka producerar värme i samma temperaturintervall. Typiska solfångarparametrar 5 som rapporteras för denna typ är UL < 3.0 W/m2 ,°C och n0 = 0.65-0.75.2. Stationära lågkoncentrerande CPC-solfångare för produktion av värme i området 100-300 grader C. Ofta bygger dessa på olika teknik för evakuering av absorbatorn, antingen genom att omsluta cirkulära absorbatorer med glasrör eller genom att införa CPC-reflektorer i heatpipes. Syftet med dessa är ofta att på ett billigt sätt producera högtemperaturvärme för olika industriprocesser, och därmed konkurera med koncentrerande solfångare typ paraboliska tråg eller traditionella heat-pipes.Beräkningar av vad en Svensk CPC-solfångare, baserad på de plana absorbatorer som finns på marknaden idag, kan prestera visar att årsutbytet av energi är jämförbart med de bästa svenska plana solfångaren som finns på marknaden idag då driftstemperaturen är ca 60-65 °C. För högre driftstemperaturer ökar skillnaden till CPC-solfångarens fördel. Vidare visas att årsutbytet har en jämnare fördelning över året jämfört, med vad en plan solfångare med samma prestanda har. Den högre prestandan vid höga driftstemperaturer och den jämnare fördelningen av energiproduktion över året gör solfångare med CPC-reflektorer intressanta för större solfångarfält, kopplade till nät med höga returtemperaturer och/eller solvärmesystem med säsongslager.Det är dock brist på undersökningar av värmeförluster i CPC-solfångare med låga koncentrationer och med plana absorbatorer, vilket är av intresse ifall de svenska erfarenheterna av plana solfångare skall tas tillvara. Potentialen med ytterligare reduktion av värmeförlusterna genom att införa extra konvektionshinder i kombination med plan absorbator har inte heller undersökts tillräckligt.

ClimateWell TDC with District Heat

The PolySMART demonstration system SP1b has been modeled in TRNSYS and calibrated against monitored data. The system is an example of distributed cooling with centralized CHP, where the driving heat is delivered via the district heating network. The system pre-cools the cooling water for the head office of Borlänge municipality, for which the main cooling is supplied by a 200 kW compression chiller. The SP1b system thus provides pre-cooling. It consists of ClimateWell TDC with nominal capacity of 10 kW together with a dry cooler for recooling and heat exchangers in the cooling and driving circuits. The cooling system is only operated from 06:00 to 17:00 during working days, and the cooling season is generally from mid May to mid September. The nominal operating conditions of the main chiller are 12/15°C. The main aims of this simulation study were to: reduce the electricity consumption, and if possible to improve the thermal COP and capacity at the same time; and to study how the system would perform with different boundary conditions such as climate and load. The calibration of the system model was made in three stages: estimation of parameters based on manufacturer data and dimensions of the system; calibration of each circuit (pipes and heat exchangers) separately using steady state point; and finally calibration of the complete model in terms of thermal and electrical energy as well as running times, for a five day time series of data with one minute average data values. All the performance figures were with 3% of the measured values apart from the running time for the driving circuit that was 4% different. However, the performance figures for this base case system for the complete cooling season of mid-May to midSeptember were significantly better than those for the monitoring data. This was attributed to long periods when the monitored system was not in operation and due to a control parameter that hindered cold delivery at certain times.