Modelon
News
Dymola Sales
Modelica Sales
Consulting
Training
Products
Downloads
Modelica
Partners
Job Positions
Students
Contact


Masters Thesis Projects

Modelon offers Masters thesis projects on selected subjects in Physical Systems modeling. Students that are interested in in one of the topics listed below are asked to contact us.


Implementation of OpenCRG Road Model for Vehicle Dynamics simulation

Published 2009-11-09

The Vehicle Dynamics Library (VDL) is a Dymola-based Modelon solution for simulation of road vehicle chassis dynamics. In vehicle dynamics simulation it is of vital importance to have a good representation of road characteristics, such as geometry, roughness, friction, and visualiztion. This representation is the combination of a data format and mathematical models. Measured road data is often re-used in different tools and environments, and it is therefore of interest to have a standard representation. German automotive industry with Daimler as initiator have therefore developed the OpenCRG open file formats and tools for the detailed description of road surfaces. Roads are currently described in a proprietary format in VDL, and we now want to implement support for the openCRG standard. This thesis work aims at interfacing Dymola to OpenCRG and implement corresponding VDL road models.

Student profile: One or two highly motivated and skilled students with interest in mathematics, modeling of physical systems, and programming (Modelica and C).

Contact: Magnus Gäfvert, Modelon AB


Phase equilibrium models, in particular for bio-fuel applications

Published 2010-01-29

Modelica is a free object-oriented simulation language. Due to its capability of handling dynamic system simulations it is also gaining significant interest in the area of chemical process control. One application would be the rectification in bio-fuel production.

Currently no Modelica Library contains property models for industrially relevant phase equilibria computations. Base classes for multi-component fluids are only available for mixtures of ideal gases.

The aim of this master-thesis project is to extend the existing Modelica library to cover mixtures of real fluids that are described with cubic equations of state or with Helmholtz functions. The following steps should be performed:

  1. Testing and extensions to the existing Real Gas models.
  2. Implementation of a new medium model that uses the Peng-Robinson (PR) equation of state (EOS) and a polynomial approximation of the specific heat capacity of the ideal gas, with a possible extension to construct a Helmholtz equation from that.The phase equilibrium computation for this new model should be taken care of automatically which includes the computation of nodes for a spline approximation for pure substances that can be used as starting values for an analytical phase boundary computation.
  3. Implementation of a mixture model that allows for mixing of at least two fluids with either PR or Helmholtz EOS, including a flash phase equilibrium computation (i.e. total moles of both fluids, pressure and temperature known, computation shall return phase fractions and mole fractions in each phase). Flashes with other input variables are an optional extension, if time remains.

Important features of the implemented medium model will be user-friendliness and numerical efficiency. The models should be tested with industrially relevant phase separations for bio-fuel production.

Student profile: One or two highly motivated and skilled students with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming, preferably from an engineering physics, engineering mathematics or chemical engineering program.

Contact: Hubertus Tummescheit, Modelon AB


Dynamic Modeling of an Air Separation process

Many modern power generation processes, for example Oxyfuel plants designed for "Carbon Capture and Storage" (CCS), depend on a large capacity air separation plant upstream of the power plant process. The air separation is one of the processes that determine the dynamic response of a full scale Oxyfuel plant, and therefore a dynamic model of an air separation plant is a necessary component in a system level simulation model of an Oxyfuel plant. The task in this Master's thesis project will be to develop a dynamic model of an air separation plant based on a number of existing models in Modelon's power plant modeling library CombiPlant. Starting with a literature review, the student will decide for appropriate models for separation columns and work with the phase separation fundamentals for air separation at an industrial scale. The models are going to be implemented in Modelica/Dymola and possibly compared to process data from a pilot size plant.

Student profile: One or two highly motivated and skilled students with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Hubertus Tummescheit, Modelon AB

Advanced Moving Boundary models for Air Conditioning Heat Exchangers

Moving boundary (MB) models have been used for many years to model control-oriented models for simulation of AC- and refrigeration systems. This Master's thesis project tried to address two of the main shortcomings of that model type for more widespread use, in particular for real-time capable models.
  1. A zone in a moving boundary model may vanish when e.g. the superheat (temperature difference to evaporation temperature at outlet) decreases to zero. A modification of a moving boundary model that can cope with a change in the number of zones that are currently active would enlarge the scope of MB models.
  2. Existing MB models are usually coupled with very simplistic, and not geometry based models for the air-side heat transfer, in spite of the fact that the air side dominates the heat transfer resistance.  Coupling improved air side models with the MB concept will improve the predictive capability of MB models considerably.
Student profile: A highly motivated and skilled student with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Hubertus Tummescheit, Modelon AB

Heat Exchanger Design Models

The Modelica language has been used mostly for dynamic system level models. The equation-based, object oriented approach of Modelica is equally well useable for static, design-oriented models. Modelon has built design-oriented models for a few specific heat exchanger types, and would like to investigate the applicability of this method to other types of heat exchangers, in particular standard tube-and-fin type heat exchangers and tube-and-shell heat exchangers.  A library of fundamental components for physical heat exchangers models is available and needs to be extended to include new geometries and correlations.

Student profile:
A highly motivated and skilled student with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Hubertus Tummescheit, Modelon AB

Engine Cooling Library model development


The goal of the thesis work is to construct models of the overall energy balances of the combustion engine, ram air cooling, gear box and climate box heater of a passenger car. The models should be usable for steady-state design computations and for partly dynamic system level models for engine management systems and model based control design. Integration into overall vehicle energy management schemes for hybrid vehicles should be possible. On the air side, the models have to interface with coarse, integrated data from CFD computations of the engine compartment and existing models for inhomogeneous flow at the engine radiator. Data and existing models of a current Mercedes production car will be used for validation. The scope of the thesis can be adapted to either one or two Master's students working as a team.
The models will be built in the Modelica language and can make use of an extensive set of existing object oriented model libraries.

Student profile: One or two highly motivated and skilled students with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Hubertus Tummescheit, Modelon AB

Fuel consumption from automotive air-conditioning

Simulation is being used more and more in the automotive industry to enable faster design and reduce expensive testing. Coming EU regulations put requirements on car makers to declare the added fuel consumption when driving with the air-conditioning unit switched on under different conditions. The regulations make it possible to use validated simulation models for the calculation of fuel consumption.
The focus of this work is to build simple component and system models for vehicle fuel consumption calculation. The models can be based on and compared with components from the Air-Conditioning library, but should be simple enough that drive-cycle simulations can be run very quickly. Component models should be compared with available data and part of the work is to propose approaches for adapting component parameters to the measured data.

Student profile: One or two highly motivated and skilled students with interest in thermodynamics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Hubertus Tummescheit, Modelon AB

Modeling for Hybrid Vehicle Energy Management

With the introduction of hybrid electric propulsion, additional degrees of freedom to control energy flow within a vehicle has been introduced. To reach good fuel efficiency it is critical to make all powertrain components - from combustion engine to electric machines and energy storages - work at their optimal operating point. This is often referred to as energy management. The focus of this work is to design and implement a model library with suitable powertrain components that are capable of describing efficiency and losses, to be used as a tool for analysis of energy management functions. The model library shall be demonstrated with an application example.

Student profile:
Two highly motivated and skilled students with interest in power electrics, control, dynamical systems, modeling of physical systems, and programming.

Contact: Magnus Gäfvert, Modelon AB

Real-time Capable Mean-Value Engine Simulation Models for Diagnostics Applications This thesis work is offered by Modelon AB, Volvo AB (Concept Engine Functions, Volvo Powertrain) and Division of Combustion Engines, LTH, together. The background is that diagnostics functions are growing increasingly complex in modern combustion engines. Validation and verification of algorithms and ECU:s constitute major efforts that may be greatly relieved by the use of simulation technology. Real-time capable simulation models may be used in place of real engines in so called Hardware-In-the-Loop (HWIL) setups. The task of this thesis work is to develop real-time capable simulation models for validation and verification of engine diagnostics functions using HWIL methods. The models shall comprise a large part of the engine system including gas-exchange, combustion, after-treatment, and heat dynamics, with proper simplifications and approximations. The models will be implemented in the Modelica language using the Dymola modeling and simulation environment.

Student profile: Two highly motivated and skilled students with combined good knowledge on combustion engines and thermodynamics, dynamical systems and control theory, modeling of physical systems, and programming.

Contact: Magnus Gäfvert, Modelon AB, or Per Tunestål, Department of Energy Sciences, Division of Combustion Engines, LTH

Multi-Body Mechanisms for Improved Simulation Speed

Simulation is an important tool for road vehicle manufacturers and their suppliers in the research, development, design, and validation processes. For products such as vehicles, robots, and construction equipment, mechanical mechanisms are used to control motion. Mechanisms are traditionally modelled with multi-body primitives such as joints and bodies. This normally yields systems of non-linear equations which are time consuming to solve. In this project, analytical solutions to some common mechanisms are to be derived to improve simulation speed.

Student profile:
Two highly motivated and skilled students with interest in mechanics, dynamical systems, modeling of physical systems, and programming.

Contact: Johan Andreasson, Modelon AB
 
 
 
 
 


MODELON AB • Ideon Science Park • SE-22370 LUND, Sweden • Phone: +46 46 286 2200 • Fax: +46 46 286 2201