Many devices in electronics are in the form of multilayered structures. These structures can fail catastrophically if they contain defects or cracks. Enhancing their fracture properties is therefore critical to improve the reliability of the systems. The interface-dominated fracture mechanics of multilayered structure was studied using experiments and finite element (FE) modeling by considering two examples: thin films on polymer substrates in flexible electronics and Cu leadframe/epoxy molding compound (EMC) in micro-electronics packaging. In the first example, aluminum-manganese (Al-Mn) thin films with Mn concentration up to 20.5 at.% were deposited on polyimide (PI) substrates. A variety of phases, including supersaturated fcc (5.2 at.% Mn), duplex fcc and amorphous (11.5 at.% Mn), and completely amorphous phase (20.5 at.% Mn) were obtained by adjusting alloying concentration in the film. In comparison with crystalline and dual phase counterparts, the amorphous thin film exhibits the highest fracture stress and fracture toughness, but limited elongation. Based on a fracture mechanism model, a multilayer scheme was adopted to optimize the ductility and the fracture properties of the amorphous film/PI system. Tensile deformation and subsequent fracture of strained Al-Mn films on PI were investigated experimentally and by FE simulations. It was found that by sandwiching the amorphous film (20.5 at.% Mn) between two ductile copper (Cu) layers, the elongation can be improved by more than ten times, and the interfacial fracture toughness by twenty four times with a limited sacrifice of the film's fracture toughness (less than 18%). This design provides important guidelines to obtain optimized mechanical properties of future flexible electronics devices. The reliability of amorphous brittle Al-Mn (20.5 at.% Mn) thin films deposited on PI substrates is strongly influenced by the film/substrate interface adhesion. Some strategies to improve the adhesion of the interface were conducted, including roughening the surface of the PI substrate, adding a buffer layer and then tuning its thickness. Tensile testing and FE analysis of amorphous Al-Mn thin films with and without buffer layers coated on intact and plasma etched rough PI were investigated. It was found that by adding a chromium buffer layer of 75 nm on a rough PI substrate, the interface adhesion of the film/substrate can increase by almost twenty times. The obtained results would thus shed light on the interfacial engineering strategies for improving interface adhesion for flexible electronics. In the second example, a systematic investigation and characterization of the interfacial fracture toughness of the bimaterial Cu leadframe/EMC was carried out. Experiments and FE simulations were used to investigate delamination and interfacial fracture toughness of the biomaterial system. Two dimensional simulations using computational fracture mechanics tools, such as virtual crack closure technique, virtual crack extension and J-integral proved to be computationally cheap and accurate to find the interfacial fracture toughness of the bimaterial structures. The effects of temperature, moisture diffusion and mode-mixity on the interfacial fracture toughness were investigated. Testing temperature and moisture exposure significantly reduce the interfacial fracture toughness, and its relationship with the mode-mixity was achieved by fitting the results with an analytic formula.

Le sujet de ce travail concerne une partie précise des moteurs à explosion : le système soupape /siège de soupape. Les conditions de contact sont sévères : température élevée, choc, glissement, atmosphère agressive, … Le but a été d’évaluer la résistance à l’usure sous différents environnements des aciers obtenus par la métallurgie des poudres pour les sièges de soupape. Une expérimentation sur un dispositif d'essai spécifique d’impact-glissement a permis d’exploiter la dynamique instantanée du contact et la perte de matière en fonction de l’angle de contact (les angles testés sont 30°, 45° et 60°). L’étude s’est appuyée sur : - Une modélisation numérique par éléments finis d’un contact de configuration bille/plan. La reprise du modèle mécanique du simulateur expérimental a mis en avant une évolution de la dynamique du contact d’un glissement alternatif à 30° à une succession de multi-impact à 45° et 60°. Ce résultat a été validé à l’aide des observations par caméra rapide et des mesures de la résistance électrique du contact. Les résultats numériques montrent que les contraintes de cisaillement diminuent pour les grands angles. En revanche, une déformation plastique importante a été induite par les multi-impacts. Ces paramètres sont sensibles à l’augmentation du frottement aux faibles angles. La modélisation numérique a amené des réponses complémentaires aux résultats expérimentaux. - Une analyse tribologique du couple acier fritté/ acier de roulement, modélisé par une configuration bille/plan en mouvement alternatif et sous impact-glissement. Chaque chargement entraine des processus d’endommagement spécifiques. Dans le cas d’impact-glissement à sec, la perte de matière des aciers frittés augmente avec l’angle de contact. L’examen des traces d’usure indique l’importance de l’adhésion, de l’abrasion et de la déformation plastique. L’introduction d’un lubrifiant à l’interface entraine des modifications sur la dynamique du contact et les mécanismes d’usure. La viscosité et la composition chimique du lubrifiant influent différemment sur la détérioration de la surface. Dans ces conditions, cette dernière est associée à la croissance des pores à la surface, la propagation des fissures à la surface et en sous-couche et l’abrasion. Le suivi du volume d’usure en température indique une usure importante à 180°C. Ceci est dû à la cinétique d’oxydation de l’acier fritté. A plus haute température, la surface est protégée contre l’abrasion et l’adhésion grâce à la présence d’une couche de tribo-oxydation dite ‘phase glacée’. Ce travail montre l’importance de la compréhension de la relation entre la microstructure des aciers frittés destinés au siège de soupape et leur comportement. Ces matériaux ont montré une adaptabilité parfaite entre la perte de matière et les conditions de sollicitation. Les mécanismes d’usure montrent une totale dépendance à la fois à l’angle de contact et à l’environnement. / Sintered steel is used as a material for valve seat insert in automotive engines. During operation, a dynamic contact occurs between the valve and its seat. To investigate the wear behavior of sintered steel for this application, we have developed an impact-sliding tester using a ball on flat configuration. Impact-sliding experiments have been conducted at different impact angles (30°, 45°, 60°) with and without lubrication to investigate the surface damage of the sintered steel under this contact loading and to understand the effect of lubrication. As a first step, we investigated numerically the evolution of the contact pressure, stress and strain as a function of time. In fact, owning the experimental bench test, a finite element model was developed. Numerical results show an evolution from of dynamic behavior from permanent reciprocating sliding at low angles to an intermittent motion called multi-impacts at higher angles. Experimental electric resistance measurements seem to confirm these evolutions. As a consequence, shearing stress is reduced when plastic deformation increased with multi-impacts. Wear track observations are in good agreement with these findings. Our results have shown an important variation of the wear rate in relation to impact-sliding angle. In dry condition, a low wear regime is observed for low angles; whereas maximum wear is observed at 60° angle for lubricated contacts. The wear scar in the dry contact is deeper than in the lubricated one. The damaged surface of sintered steel is examined by a Scanning Electron Microscopy (SEM). In dry conditions, the contact area wears out quickly due to an adhesive-abrasive process. Under lubricated conditions, a fatigue crack opening is associated to a lower wear rate. The lubricated impact-sliding condition modifies the main surface damage phenomena. In addition, a comparison of wear volumes produced using pure mineral base oil and the same base oil containing an anti-wear, anti-friction additive (ZDDP), shows that this additive has only a weak effect on wear reduction under squeeze–sliding lubrication. A discussion of basic wear mechanisms is presented to explain the observations. The present research was carried out to study the combined aspects of impact and sliding failure mechanism at different contact temperatures. The tribological behavior was investigated both under reciprocating motion and with a dynamic impact-sliding loading. The measured friction coefficient decreases as the contact temperature increases. The presence of oxides seems to be the key factor of this evolution. When the loading changes to a combined impact with slides, wear rate and mechanisms of the sintered steel vary with temperatures. Scanning electron microscopy observations coupled with EDX analysis were investigated inside and outside of the wear track in order to understand the surface accommodation with temperatures.

Le comportement mécanique du verre silico-sodo-calcique soumis à un contact ponctuel peut être amélioré par trempe. Les effets de trempe se manifestent par des gradients de propriétés sur le comportement du verre rendant la modélisation plus complexe. Ce travail a porté sur la modélisation et la simulation des verres non trempés et trempés soumis aux essais de nano-indentation et de nano-rayage. Les outils de simulation par éléments finis ont été développé et validés sur le verre métallique, choisi en raison de son comportement plastique connu. En comparant avec les données expérimentales, nous avons observé que les paramètres du matériau et les conditions expérimentales pouvaient donner la même réponse sur la courbe force-déplacement et l'empreinte. L’identification des propriétés du matériau doit être basée sur des comparaisons avec des simulations considérant la géométrie réelle de l'indenteur, la souplesse de la machine et le tilt de surface de l'échantillon. Les stratégies développées permettent de faire des identifications sur le comportement du verre silico-sodo-calcique. Dans la famille du verre silicaté, le verre silico-sodo-calcique a un comportement plastique semblable au verre de silice qui est affecté non seulement par le mécanisme de cisaillement mais aussi par la densification. Il est essentiel donc d'étudier le comportement densification/cisaillement du verre de silice car il est largement discuté dans la littérature. La comparaison des résultats de simulation par plusieurs modèles avec les courbes force-déplacement et les images d'empreinte a montré que la modélisation de la densification, en prenant en compte l'écrouissage et la modification des modules élastiques, est suffisante pour décrire le comportement des verres silicatés. Enfin, les connaissances sur les simulations des essais à chargement ponctuel et la modélisation du comportement du verre ont guidé une étude sur les effets du gradient de propriétés dans les verres trempées thermiquement et chimiquement. / The mechanical behavior of soda-lime-silica glass subjected to contact damage can be improved by tempering. The effects of tempering created tailor properties and the glass behavior more complicated. Therefore, this work studied the modeling and simulation of non-tempered and tempered glasses subjected to the nano-indentation and nano-scratch tests. The finite element simulation tools have been primarily validated and the metallic glass was chosen for the studies because of its known plastic behavior. By comparing with the experimental data, we have observed that the parameters of material model and the experimental conditions could give the same response on load displacement curve and imprint. The evaluation of material properties must be based on the comparisons with fully modeled simulation considering the real geometry of the indenter, the compliance of the instrument and the tilting of the sample surface. The developed strategies allow identification of soda-lime-silica glass behavior. Similar to that of silica glass, the plastic behavior of soda-lime-silica glass is not only affected by the shear mechanism but also the densification. Hence, it is essential to study the shear/densification behavior of silica glass as it is widely discussed in the literature. The comparison of simulation results by several models with the load displacement data and the images of imprint showed that the modeling of densification by taking into account the hardening and the change in elastic moduli is sufficient for describing the behavior of silicate glasses. Finally, the knowledge about the simulation of the contact loading tests and the modeling of glass behavior guided a study on the coupling of tailored properties effects in the thermally and chemically tempered glasses.

abstract: There has been a renewed interest to understand the degradation mechanism of concrete under radiation as many nuclear reactors are reaching their expiration date. Much of the information on the degradation mechanism of concrete under radiation comes from the experiments, which are carried out on very small specimens. With the advent of finite element analysis, a numerical predictive tool is desired that can predict the extent of damage in the nuclear concrete structure. A mesoscale micro-structural framework is proposed in Multiphysics Object-Oriented Simulation Environment (MOOSE) finite element framework which represents the first step in this direction. As part of the framework, a coupled creep damage algorithm was developed and implemented in MOOSE. The algorithm considers creep through rheological models, while damage evolves exponentially as a function of elastic strain and creep strain. A characteristic length is introduced in the formulation such that the energy release rate associated with each element remains the same to avoid vanishing energy dissipation with mesh refinement. A creep damage parameter quantifies the effect of creep strain on the damage that was calibrated using three-point bending experiments with varying rates of loading. The creep damage model was also validated with restrained ring shrinkage tests on cementitious materials containing compliant/stiff inclusions subjected to variable drying conditions. The simulation approach explicitly considers: (i) moisture diffusion driven differential shrinkage along the depth of the specimen (ii) viscoelastic response of aging cementitious materials (iii) isotropic damage model with Rankine′s failure initiation criterion, and (iv) random distribution of tensile strengths of individual finite elements. The model was finally validated with experimental results on neutron-irradiated concrete. The simulation approach considers: (i) coupled hygro-thermal model to predict the temperature and humidity profile inside the specimen (ii) radiation-induced volumetric expansion of aggregates (RIVE) (iii) thermal, shrinkage and creep effects based on the temperature and humidity profile and (iv) isotropic damage model with Rankine’s criterion to determine failure initiation. / Dissertation/Thesis / Doctoral Dissertation Civil, Environmental and Sustainable Engineering 2020

Packages are the means of preservation, distribution and convenience of use for food, medicine and other consumer products. Package opening is becoming complicated in many cases because of cutting cost in design and production of opening techniques. Introduction of new package opening technique, material or geometry for better opening experience, forces new design measurements which require a large number of prototype developments and physical testing. In order to achieve more rapid and accurate design, Finite Element Method (FEM) Simulations are widely used in packaging industries to compliment and reduce the number of physical testing. The goal of this work is to develop the building blocks towards complete package opening simulation. The study focuses on testing and simulation of shear fracture and shear delamination of packaging materials. A modified shear test specimen was developed and optimized by finite element simulation. Test method was validated for High-density polyethylene (HDPE) and Polypropylene (PP). The developed method has been accepted by international standards organization ASTM. Based on linear elastic fracture mechanics, a geometry correction factor of shear fracture toughness for the proposed specimen was derived. The study concluded that, for ease of opening, HDPE is a more favorable material for screw caps than PP. When performing the experiment with the shear specimen to find essential work of fracture, the ligament length should be varied between twice of the thickness and half of the width of the specimen ligament. Multi-layered thin laminate of Low-density polyethylene (LDPE) and aluminum (Al), also known as Al/LDPE laminate, is another key object addressed in this study. Continuum and fracture testing of individual layers provided the base information and input for numerical modeling. The propagation of an interfacial pre-crack in lamination in Al-LDPE laminate was simulated using several numerical techniques available in the commercial FEM solver ABAQUS, and it was concluded that using the combination of VCCT technique to model the interfacial delamination and coupled elasto-plastic damage constitutive for Al and LDPE substrates can describe interfacial delamination and failure due to necking. It was also concluded that the delamination mode in a pre-crack tip is influenced by the ratio of fracture energy release rate of mode I and II. To address the challenge in quantifying shear energy release rate of laminate with very thin substrate, a convenient test technique is proposed. Additionally, scanning electron microscopic study provided useful information on fractured and delaminated surfaces and provided evidence that strengthened the conclusions of this work. The proposed test methods in this work will be crucial to measure the shear mechanical properties in bulk material and thin substrates. Laminates of Al and LDPE or similar material can be studied using the developed simulation technique which can be effectively used for decision support in early package development.

Global warming sets a high demand to reduce the CO2 emissions of vehicles. In the European Union heavy-duty road transports account for 6 % of the total greenhouse gases and one of the main factors affecting these emissions is related to the rolling resistance of tyres. The optimal usage of tyres is an important part of solving these challenges, thereby it is important to understand the parameters affecting rolling resistance and the different compromises coupled to them. These compromises could be analysed using computational and experimental methods. To set out the groundwork necessary to minimise the energy consumption of trucks and assess the different parameters affecting tyre behaviour, the following studies have been conducted during this thesis. A framework to model and parametrise truck tyre rubber has been developed for finite element simulations. The presented parallel rheological material model utilises Mooney-Rivlin hyperelasticity, Prony series viscoelasticity, and perfectly plastic networks. A method to reduce tuneable parameters of the model, which significantly simplifies possible parameter studies, is presented. The model has been parametrised using test data from dynamic mechanical analysis of samples from a long haulage heavy truck tyre, and shows a good agreement with the test data. To test the suitability of the modelling technique for tyre simulations, the constitutive model is used in various tyre simulations using the arbitrary Lagrangian-Eulerian method. The material modelling technique is shown to work for static force-deflection as well as dynamic simulations estimating longitudinal force build-up with varying slip levels. Additionally, the modelling technique captures the uneven contact pressure in steady-state rolling, which indicates that the model could also be used in rolling resistance simulations. To study the change of ambient temperature on rolling resistance using experimental methods, a climate wind tunnel is used where the rolling resistance is quantified using a measurement drum. Tests were conducted between -30 °C and +25 °C, and a considerable ambient temperature dependency on rolling resistance was found. Moreover, temperature measurement inside a tyre shoulder is a good indicator for rolling resistance in a broad range of ambient temperatures. Finally, battery-electric long haulage truck driving range calculations are also conducted with varying rolling resistance and air density at different temperatures, showing a significant decrease of driving range with decreasing ambient temperature. / Den globala uppvärmningen ställer höga krav på att minska tunga fordons CO2-utsläpp. Tunga transporter står för 6 % av de totala växthusgaserna i Europeiska unionen och att fokusera på optimal användning av däck är en viktig del för att minska förorenande växthusgaser. Därför är det viktigt att förstå parametrar som påverkar rullmotståndet och olika kompromisser kopplade till dem. Dessa kompromisser skulle kunna analyseras med hjälp av beräkningsmetoder och experimentella metoder. För att lägga grunden för att minimera energiförbrukningen för lastbilar och bedöma olika parametrar som påverkar däckens beteende, har följande studier genomförts i denna avhandling. Ett ramverk för att modellera och parametrisera lastbilsdäcksgummi utvecklades för finita elementmetod-simuleringar. Den presenterade parallella reologiska materialmodellen använder Mooney-Rivlin hyperelasticitet, Prony-series viskoelasticitet och perfekt plastiska nätverk. En metod har utvecklats för att reducera antalet justerbara materialparametrar i modellen, vilket avsevärt förenklar möjliga parameterstudier. Modellen har parametriserats med hjälp av testdata från dynamisk mekanisk analys och visar en god överensstämmelse mellan testdata och simuleringar. Provstavarna skars ut från ett lastbilsdäck för tunga fordon. För att testa modelleringsteknikens lämplighet användes den konstitutiva modellen i olika däcksimuleringar. Materialmodelleringstekniken har visat sig fungera för statisk vertikalstyvhet såväl som dynamiska simuleringar som uppskattar longitudinell kraftgenerering med varierande slipnivåer och olika friktionskoefficienter. Modelleringstekniken fångar ojämnt kontakttryck vid stationär rullning, vilket indikerar att modellen även kan användas i simuleringar av rullmotstånd. För att studera omgivningstemperaturens inverkan på rullmotståndet med experimentella metoder användes en klimatvindtunnel. Tester utfördes mellan -30 °C och +25 °C och rullmotståndet bestämdes med en mättrumma. Ett avsevärt beroende av omgivningstemperaturen på rullmotståndet påvisades. Dessutom indikerade provningen att temperaturmätning inuti däckskuldran är en bra indikator för rullmotstånd i ett brett område av omgivningstemperaturer. Räckviddsberäkningar för en elektrisk fjärrtransportlastbil utfördes med varierande rullmotstånd och luftdensitet vid olika temperaturer, vilket visade en signifikant minskning av körräckvidden med sjunkande omgivningstemperatur. / <p>QC 220525</p>

In this research thermal properties enhancement of phase change material (PCM) using boron nitride nanomaterials such as nanoparticles and nanotubes is studied through experimental measurements, finite element method (FEM) through COMSOL 5.3 package and molecular dynamics simulations via equilibrium molecular dynamics simulation (EMD) with the Materials and Process Simulations (MAPS 4.3). This study includes two sections: thermal properties enhancement of inorganic salt hydrate (CaCl2∙6H2O) as the phase change material by mixing boron nitride nanoparticles (BNNPs), and thermal properties enhancement of organic phase change material (paraffin wax) as the phase change material via encapsulation into boron nitride nanotubes (BNNTs). The results of the proposed research will contribute to enhance the thermal transport properties of inorganic and organic phase change material applying nanotechnology for increasing energy efficiency of systems including electronic devices, vehicles in cold areas to overcome the cold start problem, thermal interface materials for efficient heat conduction and spacecraft in planetary missions for efficient thermal managements.

Manufacturing of aero engine components requires attention to residual stress and final shape of the product in order to meet high quality product standards.This sets very high demands on involved manufacturing steps to meet design requirements. Simulation of manufacturing processes can therefore be animportant tool to contribute to quality assurance.The focus in this work is on simulation of a manufacturing process chain comprising of sheet metal forming, welding and a stress relief heat treatment.Simulation of sheet metal forming can be used to design a forming tool design that accounts for the material behaviour, e.g. spring back, and avoid problems such as wrinkling, thinning and cracking. Moreover, the simulation can also show how the material is stretched and work hardened. The residual stresses after forming may be of local character or global depending on the shape that is formed. However, the heat affected zone due to welding is located near the weld.The weld also causes large residual stresses with the major component along the weld. It is found that the magnitude of the residual stresses after welding is affected by remaining stresses from the previous sheet metal forming. The final stress relieve treatment will relax these residual stresses caused by e.g. forming and welding. However, this causes additional deformations.The main focus of this study is on how a manufacturing process step affects the subsequent step when manufacturing a component of the nickel-based super alloy 718. The chosen route and geometry is a simplified leading edge of an exhaust case guide vane. The simulations were validated versus experiments. The computed deformations were compared with measurements after each manufacturing step. The overall agreement between experiments and measurement was good. However, not sufficiently accurate considering the required tolerance of the component. It was found from simulations that the residual stresses after each process affects the subsequent step. After a complete manufacturing process chain which ends with a stress relief heat treatment the residual stresses were not negligible. VIII Special experiments were performed for studying the stress relief in order to understand how the stresses evolve through the heat treatment cycle during relaxation. It was found that the stresses were reduced already during the beginning of the heating up sequence due to decreasing Young´s modulus and yield stress with increasing temperature. Relaxation due to creep starts when a certain temperature was reached which gave a permanent stress relief.

No description available.

The automotive industry is reliant on accurate finite element simulations for developing new parts or machines and to achieve this, accurate material models are essential. Material cards contain input about the material model, and are significant; however, time-consuming to calibrate with traditional methods. Therefore a newer method involving Machine Learning (ML) and Feed-Forward Neural Networks (FFNN) is studied in the thesis. The direct application of calibration with FFNN has never been applied to calibrate the Swift hardening law and Barlat yield 2000 criteria, which is done in this thesis. All steps for calibration are performed to achieve a high-fidelity database capable of training the FFNN. The outline of the thesis involves four different phases; experiments, simulations, building the high-fidelity database, and building and optimizing the FFNN. The experiment phase involves tensile testing of three different types of specimens in three material directions with Digital Image Correlation (DIC) to capture local strain. The simulation phase is to replicate all the experiments in LS-DYNA and perform finite element simulation. The finite element models are simulated 100 times and, respectively, 1000 times with different material parameters within a specific range. This range has a lower and upper bound that covers the experimental results. The database phase involves extracting the data from a huge amount of simulations and then extracting the key characteristics from the force-displacement curve. The last phase is building the FFNN and optimizing the network to find the best parameters. It’s first optimized based on Root Mean Square Error (RMSE) and then points from the Swift hardening curve and Barlat yield 2000 criteria are compared with experimental points. The result shows that the FFNN with the high-fidelity database can predict material parameters with an accuracy of over 99 % for the hardening law at the points chosen for optimization and the anisotropy parameters are optimized to 97 % accuracy for the yielding points and Lankford coefficients. The thesis concludes that the FFNN can accurately predict the material parameters with real experimental data. The effectiveness of using this method is significantly faster than traditional methods because only one type of test is needed. / Bilindustrin är beroende av trovärdiga och noggranna finita element simuleringar för utveckling av nya komponenter eller maskiner, och för det behövs noggranna materialmodeller. Materialkort innehåller information om materialmodellerna och är av stor betydelse, men är tidskrävande att kalibrera med traditionella metoder. Därför är en ny metod som involverar Maskininlärning (ML) och Feed-Forward Neurala Nätverk (FFNN) undersökt i avhandlingen. Applikationen av att kalibrera med FFNN har aldrig blivit undersökt för ”Swift hardening law” och anisotropi kriteriet ”Barlat yield 2000”. Alla steg för att kalibrera materialet är utförda för att uppnå en högkvalitativ databas som är kapabel att träna ett FFNN. Arbetets översikt involverar fyra faser som är; experiment, simulationer, databasensuppbyggnad och utvecklingen samt optimeringen av FFNN. Experimentfasen involverar dragprov för tre olika geometrier i tre materialriktningar tillsammans med Digital Image Correlation (DIC) för att fånga lokala töjningspunkter. Simulationsfasen går ut på att replikera experimentfasen genom finita element simuleringar i LS-DYNA. Finita element modellerna är simulerade 100 respektive 1000 gånger med olika materialparametrar inom ett specifikt intervall med en övre och undre gräns som ska täcka experimentdatan. Databasfasen handlar om att extrahera data från de massiva antalet simuleringar och extrahera nyckelbeteenden från kraft-förflyttningskurvan. Den sista fasen är att bygga FFNN och optimera för att hitta bästa möjliga parametrar. Det är först optimerat baserat på Root Mean Squared Error (RMSE) och sedan punkter från Swift härdningskurvan och beteenden genererat från Barlat yield 2000 som är jämförda med experimentella värden som Lankfordkoefficienter och sträckgränser för rullningsriktningarna. Resultatet visar att ett FFNN med en högkvalitiativ databas kan estimera materialparametrar med en noggrannhet över 99 % för härdningskurvan för jämförelsepunkterna och med en 97 % noggrannhet för anisotropipunkterna som Lankfordkoefficienter och sträckgränser i rullningsriktningarna. Exjobbet avslutas med att dra slutsatsen att FFNN kan estimera riktiga materialparametrar med en viss noggrannhet. Effektiviteten av att använda metoden är betydligt snabbare än traditionella metoder eftersom det endast tar några sekunder att estimera parametrarna när datan är extraherad och enbart en typ av test behövs.