[work] Insinööritieteiden korkeakoulu / ENG
Permanent URI for this collectionhttps://aaltodoc.aalto.fi/handle/123456789/55
Browse
Recent Submissions
Now showing 1 - 20 of 65
- ECAMARIS Operational Concept Template for Autonomous and Automated Maritime Systems - ECAMARIS Project: Enablers and Concepts for Automated Maritime Solutions
School of Engineering | J Muu elektroninen julkaisu(2023) Owen, Douglas; Laarni, Jari; Liinasuo, Marja; Koskinen, HannaThe purpose of the ECAMARIS Operational Concept Template for Autonomous & Automated Maritime Systems is to provide a basis for the development safe and effective autonomous and automated systems. The template is designed to provide a usable framework to support early and comprehensive identification and collation of key information and assumptions about the system and its operational context. This information provides a basis of initial concept development, design, analysis and assurance activities over the design lifecycle of the system. The template aids the identification and critical review of system features and capabilities to enable the development of systems that deliver their objectives and are effectively integrated into their human, technical, organisational, environmental and regulatory context. The Operational Concept template was developed primarily for use with Maritime Autonomous Ship Systems (MASS). It is based on cross-industry guidance with some specific adaptations for the maritime domain. The template may be useful in the development of autonomous and automated systems for other maritime infrastructure and can also be adapted for use in other industries. MS Word version of the template is available in Aaltodoc. The ECAMARIS Operational Concept Template for Autonomous and Automated Maritime Systems is designed to support early and comprehensive collation of key information about an autonomous /automated system and its operational context. This information provides a basis of initial concept development, design, analysis and assurance activities over the design lifecycle of the system. This work is made available for use under the CC BY 4.0 license. - Desphosphorization in ironmaking and oxygen steelmaking
School of Engineering | J Muu elektroninen julkaisu(2018) Saqlain, Muhammad; Owais, Muhammad; Järvinen, Mika; Visuri, Ville-Valtteri; Fabritius, TimoSteelmaking is an extensive industry based on modifying the physical properties of iron to fit a wide range of applications by either adding alloying elements or removal of impurities. Thus, the effects of various elemental components on final steel properties are at the heart of the steelmaking process whether in small concentrations coming from the raw materials or being in larger concentrations as alloying elements. Owing to increased demand of iron ore raw materials, the access to high-quality ores has been under stress and thus, increased usage of low-quality ores, which contains high concentrations of impurities such as phosphorus and sulphur has become more economical. Such impurities have been known to have adverse impacts on the final steel properties and need special management in conventional steel making operation. This report primarily focuses on the management strategies of phosphorus in steelmaking including the behavior of phosphorus and its removal strategies in several core components of the process: blast furnace (BF), basic oxygen furnace (BOF), electric arc furnace (EAF) and argon oxygen decarburization (AOD). The objective of the report is to review underlying research in the area giving a reference to the available literature and support propagation of future research projects addressing various aspects of dephosphorization in steelmaking. - DES turbulenssimallin lisääminen FINFLO virtausratkaisijaan
Helsinki University of Technology | J Muu elektroninen julkaisu(2009) Kaarlonen, Kyösti; Hellsten, AnttiAbstract: Tämä muistio käsittelee Streletsin k-ω-DES-turbulenssimallin käyttöönottoa FINFLO virtausratkaisijassa. Muistiossa raportoidaan työn tämänhetkinen tilanne. Main result: Streletsin k-ω-DES malli on ohjelmoitu FINFLOon ja validointilaskuja on tehty yksinkertaiselle deltasiipimallille suurella kohtauskulmalla sekä virtaukselle ohi kaksiulotteisten periodisesti toistuvien kukkuloiden. Lisäksi on tutkittu virtausta kaksiulotteisten siipiprofiilien ympärillä. - Description of the numerical methodology for the ercoftac test case F3
Helsinki University of Technology | J Muu elektroninen julkaisu(1997) Rautaheimo, Patrik; Ojala, JuhaAbstract: The purpose of thies work is to compare different low-Reynolds-number turbulence models and their suitability for the highly three dimensional return channel with a deswirl cascade. A comparison is made between Chien's k-ε model, Chien's model with a rotational correction, Gatski et al. ARSM and the full Reynolds stress model by Speziale et al. Experiments are provided by Concepts ETI. - Transportation of FINFLO to a cluster of PCs
Helsinki University of Technology | J Muu elektroninen julkaisu(1999) Rautaheimo, Patrik; Kaurinkoski, PetriAbstract: For a long time, the UNIX workstations have been superior against to the personal computers as far as computational speed is concerned. However, the gap between the speeds has been decreasing. Today’s personal computers are fast and relatively cheap, whereas the UNIX workstations are expensive. A comparison between the available personal computers and traditional engineering tools like UNIX workstations, servers and also the supercomputers is needed. In this report such a study is made for a cluster of PCs. It will be shown that with ideal cases the speed and parallel performance figures of a cluster of personal computers are competitive with those of the workstation. Main result: Benchmark runs in a cluster of Linux personal computers, and a comparison with different UNIX-platforms. - FINFLO- ja Fluent-laskennan vertailu 2D U-kanavassa
Helsinki University of Technology | J Muu elektroninen julkaisu(2007) Miettinen, Ari; Siikonen, TimoAbstract: Tässä raportissa kuvataan TKK:ssa kehitetyn FINFLO-ratkaisijan ja kaupallisen Fluent-ohjelmiston vertailulaskentaa kaksidimensioisessa U-putkessa, jonka sisääntulohalkaisijaan referoitu Reynoldsin luku on 522 000 vastaten turbulenttia virtausta. Molempien ratkaisijoiden k − ω SST turbulenssimalleja vertailtiin keskenään identtisen laskentatapauksen avulla. Main result: Laskentatulokset eivät poikenneet merkittävästi toisistaan, eikä konvergenssinopeuksissa ollut suuria eroja. FINFLO-laskenta vaikutti aluksi keskeneräiseltä seinämän läheisten paineiden perusteella. Kävi kuitenkin ilmi, että Fluentin tulostama paine on efektiivinen sisältäen turbulentin paineen ja FINFLOn paine on staattinen. Kun tämä otetaan huomioon, tulokset ovat pääsuureiden osalta hyvin lähellä toisiaan. Turbulenssisuureet ovat kvalitatiivisesti samoja. Tulosten perusteella voidaan kuitenkin päätellä turbulenssimallien olevan toiminnaltaan samoja. - Simple Polynomial Fittings for Steam
Helsinki University of Technology | J Muu elektroninen julkaisu(2007) Miettinen, AriAbstract: The purpose of this report is to develop fittings for saturation curve, steam density and its derivatives to be used in FINFLO Navier-Stokes solver in two-phase flow computations. Main result: A flexible method to fit data - Study of Boundary Conditions in single-Blade Pump Simulation
Helsinki University of Technology | J Muu elektroninen julkaisu(2007) Ala-Juusela, JuhaveikkoAbstract: Single-blade sewage pump is simulated numerically using FINFLO flowsolver. Loy-Reynolds number approach is utilized with a k-ε turbulence closure. The effect of the boundary conditions on a performance prediction, mass balance fluctuation and velocity distributions is studied. Main result: Pump head, efficiency and torque are obtained from the simulated flowfield and compared to measured data. The velocity distributions are also compared to the measurements available. Reversed in- and outflow boundary conditions do not remove the fluctuation in mass balance, it only changes location. Modified impeller surface boundary condition has a small effect on pump performance but more visible on velocity distributions. - Study of Boundary Conditions in Single-Blade Pump Simulation
Helsinki University of Technology | J Muu elektroninen julkaisu(2007) Ala-Juusela, JuhaveikkoAbstract: Single-blade sewage pump is simulated numerically using FINFLO flowsolver. Low-Reynolds number approach is utilized with a k-ε turbulence closure. The effect of the boundary conditions on a performance prediction, mass balance fluctuation and velocity distributions is studied. Main result: Pump head, efficiency and torque are obtained from the simulated flowfield and compared to measured data. The velocity distributions are also compared to the measurements available. Reserved in- and outflow boundary conditions do not remove the fluctuation in mass balance, it only changes location. modified impeller surface boundary condition has a small effect on pump performance but more visible on velocity distributions. - Numerical Simulation of a Small Cooling Blower
Helsinki University of Technology | J Muu elektroninen julkaisu(2006) Ala-Juusela, JuhaveikkoAbstract: A small cooling blower is simulated numerically using FINFLO flowsolver with quasi-steady assumptions. The blower has 13 blades with an outer radius of 6 mm. The angular speed is 1047.2 Rad/s and the Reynolds number 2450. The size of the blower is 15 x 15 mm³. The Low-Reynolds number approach is utilized with a k-ω turbulence closure. The case is also simulated assuming completely laminar flow. Main result: Pressure rise, efficiency and torque are obtained from the simulated slowfield. Far curve is generated from the simulations of four air flow rates. - A CFD Study of the Effect of a Surface Roughness on Heat Transfer in a Clearance and Cooling Fins of an Electric Motor
Helsinki University of Technology | J Muu elektroninen julkaisu(2004) Ala-Juusela, JuhaveikkoAbstract: The effect of a surface roughness on heat transfer is studied in two electric-motor-related cases. A flow in a clearance and between the cooling fins of the ABB HXR 400L electric motor is simulated numerically. The clearance is simplified to a two dimensional slot and, therefore, the calculation is not valid in the ends of the rotor. Furthermore, the clearance is modelled utilizing symmetry and only a sector of 30◦ is modelled. The air flow around the cooling fins is modelled by calculating one fin space and its surroundings. FINFLO flow solver, developed at Helsinki University of Technology, is used in the simulations. For the turbulence modelling k-ω model is utilized. Main result: A surface sand-roughness of 46 µm is used in a case of the clearance of an electric motor. The improvement of the equivalent conductivity is 0.4 % with open key slots and 4 % with filled key slots. A surface roughness of 1.5 µm has no effect on the heat transfer coefficient in a flow around the cooling fins while a roughness of 260 µm improves heat transfer coefficient by 1.8 %. - Simulation of an Air Flow between Adiabatic and Heated Cooling Fins
Helsinki University of Technology | J Muu elektroninen julkaisu(2004) Ala-Juusela, JuhaveikkoAbstract The flow between cooling fins of the ABB HXR 400L electric motor was simulated numerically by calculating one fin space and its surroundings. FINFLO flow solver, developed at Helsinki University of Technology, was used in the simulations. For the turbulence modelling Chien's low-Reynolds number k-ε models were utilized, since in this case k-ω model did not seem to work properly. Main result A suitable grid with 750 000 cells was introduced. Different variations of a computational domain and boundary conditions were tested and they affected significantly to the computational results. The results were compared to measurements made by 'cold' motor, i.e. only the fan was rotated. The calculated velocity was quite near the measured values at the first half of the fin, but differed significantly towards the end of the fin. Also the simulated flow seems to stay between the fins too well. - Laser-Doppler anemometer measurements of air flow between cooling ribs of an electrical motor
Helsinki University of Technology | J Muu elektroninen julkaisu(2004) Saari, KariAbstract: Two dimensional air flow velocities and turbulence parameters between cooling ribs of an electrical motor were measured using a laser-Doppler anemometer. Both the electrical motor and the air flow between the cooling ribs were approximately at the ambient temperature. Main result: The axial air velocity decreases substantially from 19 m/s to 8...9 m/s while flowing through the cooling rib channel. Also the kinetic energy of turbulence has a rapid decrease along the flow channel. - Test Simulations with the NACA 0012 Airfoil
Helsinki University of Technology | J Muu elektroninen julkaisu(2003) Ala-Juusela, JuhaveikkoAbstract: Air-flow around the NACA 0012 airfoil was simulated with the FINFLO flow solver version 7.2 of Finflo Ltd. and the 2D solver FINF2D of Laboratory of Aerodynamics, HUT. The flow conditions are Ma = 0.3, α = 10°. Main result: Pressure and friction coefficients are in a good agreement between the two solvers, whereas the drag and lift coefficients have some differencies. - Simulation of A Vortex Pump
Helsinki University of Technology | J Muu elektroninen julkaisu(2003) Ala-Juusela, JuhaveikkoAbstract: A wortex-pump is simulated numerically using CFX-TASCFlow flow solver. The grid for the simulations is provided by the manufacturer, Grundfos A/S. The k-ε turbulence model with wall functions were utilized in the simulations. Main result: Pump head, efficiency and shaft power are obtained from the simulated flow-field. Observations are made from the graphical visualizations. The figures are shown in the Appendix of this report and are also available on a CD. - Simulation of Two Sewage Pump Geometries
Helsinki University of Technology | J Muu elektroninen julkaisu(2003) Ala-Juusela, Juhaveikko; Heiska, HarriAbstract: Single-blade sewage pumps with two different geometries are simulated numerically using two flow-solver packages, a University-developed FINFLO solver and a commercial CFX-TASCFlow. Both low and high Reynolds number approaches are utilized with k-ε and k-ω turbulence closures. Results are compared to the experimental data provided by Grundfos A/S. Main result: Pump head, efficiency and torque are obtained from the simulated flowfield. Wall-function approach under-estimates the head and the torque and over-estimates the efficiency. Low-Reynolds number simulation under-estimates all the values. - Simulation of an Air Flow in a Clearance of Electric Motor
Helsinki University of Tehnology | J Muu elektroninen julkaisu(2003) Ala-Juusela, JuhaveikkoAbstract: A flow in a clearance of the ABB HXR 400L electric motor is simulated numerically. The clearance is simplified to a two dimensional slot and, therefore, the calculation is not valid in the ends of the rotor. FINFLO flow solver, developed at Helsinki University of Technology, is used in the simulations. For the turbulence modelling k-ω and k-ε models are utilized. The clearance is modelled utilizing symmetry and only a sector of 30 degrees is modelled. The simulations are made with multiple Reynolds numbers. One case is also computed time-accurately in order to see the effect of a quasi-steady assumption. The effect of the slots is studied by calculating a case without slots on rotor and/or stator. Heat transfer is compared with an empirical correlation from literature. Main result: Calculated mean equivalent conductivity for smooth cylinders is near the prediction of the empirical correction. The effect of the slots in studied cases is at the maximum 20 %. The mean equivalent conductivity calculated with the different turbulence models differs up to 40 %. - Turbulence Characteristics of Flow in a Two Dimensional Contraction
Helsinki University of Technology | J Muu elektroninen julkaisu(2002) Rahman, M. M.; Siikonen, T.Abstract: The relaminarization of a fully turbulent flat plate boundary layer subjected to a favorable pressure gradient inside a two-dimensional contraction has been carried out numerically. Two low-Reynolds number isotropic k-~ models are employed in the simulation. The simulated fluid flow and turbulence characteristics are compared with the experiment, demonstrating that the models have good agreement with the data as the flow approaches the laminar state. - Simulation of an Air Flow between Cooling Fins
Helsinki University of Technology | J Muu elektroninen julkaisu(2001) Ala-Juusela, JuhaveikkoAbstract: The flow between the cooling fins of the ABB HXR 400L electric motor was simulated numerically by calculating one fin space and its surroundings. FINFLO flow solver, developed at Helsinki University of Technology, was used in the simulations. For the turbulence modelling k-ω and Chien's low-Reynolds number k-ε models were utilized. Main result: A suitable grid with 605 000 cells was introduced. Turbulence model affected significantly to the computational results. The k-ω model was used as a basic alternative. Results obtained are qualitative, but comparison with measurements is needed. According to the calculations, turbulent viscosity is also created in low-speed areas where one would expect laminar flow. This viscosity is a driving force for the secondary entrainment flow. The k-ε model created large amounts of turbulent viscosity in the secondary flow area, and the resulting flowfield was clearly unphysical. - Simulation of Jets in a Crossflow
Helsinki University of Technology | J Muu elektroninen julkaisu(2001) Ala-Juusela, JuhaveikkoAbstract: A circular jet in the crossflow was simulated numerically using the k-ε- and k-ω-turbulence models. Two different cases were calculated. In the first case the jet to mean flow ratio is R=0.5 and in the other case R=2.0. Overlapping mesh was used in order to utilize a refined grid without an excessive amount of grid cells. Main result: In both cases the velocity was well predicted in the jet area, in the case of R=0.5 also between the floor and the jet. As compared to the k-ε-model the k-ω-turbulence model predicted the velocity distribution better between the jet and the floor in case of R=2.0, but the result in that area was still poor. Temperature distribution was predicted with a quite good accuracy.