Energy cogeneration with Modelica Libraries Automotive application in waste heat recovery systems

Last year, Hanon Systems, a leading vehicle thermal management solution provider, was developing a system for recovering the waste heat generated in a vehicle engine. This option allows the use of low temperature heat sources (40-100 oC) to create electricity. To achieve their targets, the team uses Modelica models, which contribute to the understanding of the behavior of the innovative systems that are developed and help master its possibilities, limitations, and control options.

With data for a demo system provided by Hanon Systems, Modelon started an academic collaboration with the Faculty of Engineering (LTH) of Lund University, with the scope to achieve a working model with effective controls for a Rankine Cycle system using Modelica Libraries for Vehicle Thermal Management and the simulation environment - Dymola.

Our role in this project has been focused on overseeing the development of the Rankine Cycle system model. Within this project we had the chance to see three Modelon Libraries used, proved, and further developed. We are here now to report the outcomes.


The building bricks for this project have been: the Vapor Cycle Library, Heat Exchanger Library, and Liquid Cooling Library. These commercial libraries allowed to rapidly compose the whole cycle:

  • Building component and system model, with new or custom component models as needed
  • Parameterization of all components
  • Devising a robust initialization of the Rankine Cycle (RC) system
  • Developing prototype control strategies for the RC system.

Typical engineering tasks the model is devised for include control system design, scenario analysis as well as cycle optimization.

Modeled Rankine Cycle system

We based our project on a standard Rankine Cycle system, as shown in Figure 1: pump, evaporator, turbine, condenser, and tank. The corresponding pressure – enthalpy diagram of the process is also standard, see upper right corner of Figure 1. A working fluid called R134a, or 1,1,1,2-tetrafluoroethane, was used in the cycle.

The pump used was of type diaphragm. The evaporator was a Chevron plate heat exchanger with glycol as secondary fluid. The turbine was turbine scroll type. As with the evaporator model, the condenser was modeled by setting its geometry and boundary conditions in an existing model from Modelon's libraries. The condenser was a tube heat exchanger with air as secondary fluid.

At the outlet of the condenser and before the pump, a tank was situated to make sure that the working fluid was a liquid when it entered the pump, since vapor would damage it. The tank also regulated the amount of working fluid in the cycle.

The whole system model created in Dymola is presented in Figure 1.

Model in Dymola for the whole Rankine Cycle system
Figure 1 - Model in Dymola for the whole Rankine Cycle system

Hanon Systems controlled the test conditions of their experimental test bench like speed of the pump, ambient temperature, air velocity, mass flow, and boiler coolant temperature among others.

In the model we could adjust both boundary conditions and the topology of the system.  Boundary conditions such as the speed of the turbine, speed of the pump, temperature of the heat source for the evaporator, and temperature of the cooling air in the condenser could be easily modified.  In addition, changes in the system topology such as the addition of tanks after the condenser or evaporator.

The amount of working fluid in the cycle could be altered by adjusting the initial level of liquid in the tank.

Alternatively, the amount of working fluid in the cycle could be changed during simulation, by a charge flow source that would either fill or drain the cycle depending on the desired density for the cycle.

Testing scenarios

Different scenarios were tested to understand the system behavior:

  • The different data sets were simulated.
  • Super heating control.
  • The cycle was tested with different torques on the turbine to match the maximum power point tracking diagram the partner company had provided.
  • The cycle was tested with varying amounts of refrigerant.
  • The isentropic efficiency of the turbine was changed in order to see how the evaporation temperature and overall efficiency of the cycle changed.

Control strategy development

A Modelica-based platform provides an excellent test bench for virtual controls prototyping. The following strategies were tested with the physical system model:

  1. Superheating - Dynamic simulation of the model was performed, with a different superheat value set points.  
  2. Torque control – we simulated a line of power point tracking from measurements.
  3. Working fluid drain – this strategy has been simulated for four cases with control over the pump inlet (see Figure 2).

Dymola Simulation results when the Rankine cycle is drained of working fluid
Figure 2 - Simulation results when the cycle is drained of working fluid


The project resulted in a Rankine Cycle system model, applicable to a variety of industrial use-cases:

  • The models accurately capture the maximum power point tracking diagram and thus can be used to identify optimal operating turbine torque.
  • The predicted overall system efficiency is around 2% and agrees well with the measured data and can be changed by altering the speed of the turbine.
  • With the dynamic capabilities of the model, virtual control prototyping can be performed.
  • The simulations with the simple evaporator took between 2 and 10 minutes, certainly fast enough for the work in this project.
  • Modelon’s thermal management libraries have now been proven for Rankine Cycle system simulations. 

So, if you are involved in the development and analysis of novel heat recovery systems for energy production, do not hesitate to contact us.

We would love to hear from you!


Ylva Teleman - Rankine Cycles, Modeling and Control, Master Thesis of Science in Engineering Physics, Lund University, 2016 



Pieter Dermont holds a MSc in Mechanical Engineering from the École Polytechnique Fédérale de Lausanne, focused on the energy industry. Pieter is a modeling and simulation consultant, active in a variety of thermodynamic products and projects, as well as in teaching.

HakJun Kim holds a Ph.D in Mechanical Engineering from Seoul National University at Korea, focused on Computational Fluid Dynamics of automotive engine cooling and thermal cycle analysis.  HakJun is a CAE R&D leader at Hanon systems and doing various thermal cycle analyses like cool-down, heating, powertrain cooling system and multi-stage refrigerant cycle.

by Pieter Dermont & HakJun Kim