Acier inoxydable

L'acier inoxydable, couramment appelé acier inox ou inox, est un acier (alliage à base de fer et de carbone), comportant moins de 1,2 % de carbone et plus de 10,5 % de chrome, dont la propriété remarquable est d'être peu sensible à la corrosion et de ne pas se dégrader en rouille. La présence de chrome en solution au-delà de 10,5 % dans la matrice d'un acier provoque la formation d'une couche protectrice d'oxyde de chrome qui lui confère son inoxydabilité. D'autres éléments peuvent être ajoutés, notamment le nickel qui améliore les propriétés mécaniques en général et la ductilité en particulier, ou encore le molybdène ou le titane qui améliorent la stabilité de l'alliage pour des températures autres qu'ambiante, ou encore des éléments à haut point de fusion comme le vanadium et le tungstène accompagné en général d'une augmentation de la teneur en chrome, pour obtenir la résistance aux hautes températures au contact d'une flamme (aciers réfractaires). L'acier inoxydable est devenu in

Automation of the diagnosis process in the railway system : Detection of defects in concrete sleepers using vision-based machine learning models

Linah Charif

inspectors that walk over the track and check the defects on the rail surface, fasteners and sleepers. In the case of concrete sleepers, rail inspectors classify defects according to their size and occurrence over 20 sleepers. The manual inspection is error prone, it is time consuming and it can be sometimes risky. In an eort to overcome these issues, SBB has developed a project called AISI (Articial Intelligence Streckeninspection) in order to automatize the rail inspection by collecting and analysing images with diagnosis trains equipped with cameras. The objective of this thesis is integrated within the AISI Project, since its goal is the conception of a deep learning model to detect and classify concrete sleepers defects using computer vision and more specically, object detection. The defects studied in this thesis are cracks and akings (or spallings). In the literature, many researchers have tackled the subject of defect detection in concrete, especially in civil infrastructures. The most common methods a combination of image-processing and Machine Learning (image pre-processing and a classier) whereas the latest methods use deep learning (CNN, FCN and RPN, YOLO). The proposed methodology uses a Fast-RCNN model developed with Tensor ow Object Detection API. The evaluation of the model is given on a total of 996 defects and its precision and recall are given. The model shows average results as the number of false positives is high. It has been noted that the labels provided needed to be improved. The proposed approach has a great potential of development since a large amount of training data can be added thanks to the produced relabelling process. Also, the designed pipeline establishes a base that can be adjustable for various algorithms and congurations.

2021

Modélisation de la tribocorrosion d'aciers inoxydables dans l'eau à haute pression et haute température

Jonathan Perret

Several metallic components of pressurized water reactors (PWR) used in nuclear power plants are exposed to the combined degradation by friction and corrosion (tribocorrosion). In PWR, corrosion arises from the water pressurized at 150 bars and maintained at 300°C and results in the build up of a micrometer thick surface oxide film. Tribocorrosion is of complex nature as it involves multi-scale interactions between mechanical, chemical and material factors. Nevertheless, PWR component manufacturers are interested in understanding the deterioration mechanisms in order to select appropriate materials as well as predicting the component lifetime. This thesis is a contribution towards the elucidation of the tribocorrosion mechanisms of a specific PWR component (austenitic stainless steel control rods). The research is focused on two inter-related aspects related to the subsurface deformation induced by friction and to the high temperature and high-pressure oxidation kinetics of austenitic stainless steel under tribocorrosion conditions. The subsurface deformation is investigated by using a model tribocorrosion system consisting in an alumina ball sliding against a 304L austenitic stainless steel in sulphuric acid at ambient temperature under electrochemical control. This model system allows studying subsurface deformation in absence of a thick high temperature oxide films that can complicate the interpretation of the results. Subsurface deformation was characterized using SEM, EBDS and FIB cross-sections. Good correlation was found with the deformation found in pressurized water. Subsurface deformation was found to result in the formation of a recrystallised layer of nano-grains close to the surface and an underlying layer exhibiting a deformation gradient. The thickness of the recrystallized zone is highly affected by the presence or the absence of a few nm thick oxide in the contact. However, no obvious correlation between the thickness of the recrystallized zone, or any other measured deformation appearance, and wear could be found. The effect of oxidation was studied using a simulator (Aurore), reproducing the chemical and mechanical PWR conditions of control rods. It includes an electrochemical set-up allowing measuring in-situ and in real time the oxidation kinetic. The oxidation kinetic was found to follow a parabolic law (diffusion controlled oxide growth). Rubbing resulted in repeated scratching of the surface leading to a thinning of the oxide film and thus to an enhanced anodic oxidation. A quantitative model of the electrochemical response of stainless steel in PWR conditions was developed. Based on these results and on the third body theory of wear, a mechanistic model was proposed for describing the material behaviour in contacts operating in pressurized water. This model explains a certain number of practical observations obtained in nuclear plants or in simulators concerning chemical and materials effects on component degradation.

EPFL2010

Modélisation de la tribocorrosion d'aciers inoxydables dans l'eau à haute pression et haute température

Jonathan Perret

EPFL2010

Automation of the diagnosis process in the railway system : Detection of defects in concrete sleepers using vision-based machine learning models

Modélisation de la tribocorrosion d'aciers inoxydables dans l'eau à haute pression et haute température

The effect of light water reactor (LWR) irradiation conditions on the microstructure and tensile properties of stainless steel

The effect of light water reactor (LWR) irradiation conditions on the microstructure and tensile properties of stainless steel

Modélisation de la tribocorrosion d'aciers inoxydables dans l'eau à haute pression et haute température

Automation of the diagnosis process in the railway system : Detection of defects in concrete sleepers using vision-based machine learning models