Exhaust Heat Recovery

Model-based control of a heat engine to convert heat into usable power for energy efficient vehicles.
[Photo: Daimler AG]


Shrinking energy resources and growing traffic volume increase the pressure to enhance the energy efficiency of vehicles. Especially for heavy duty vehicles used for long-distance transport, which contribute significantly to the total traffic volume and which still depend on internal combustion engines, small energy savings means a large overall benefit.

This brought up the desire to utilize the unused exhaust gas heat, which, in spite of modern diesel technologies, contains about 34% of the primary energy, and convert it into mechanical power. This can be fed back to the crankshaft or stored in a battery, using a generator, which improves the overall efficiency of the vehicle.

The conversion of heat into mechanical power is done economically with a heat engine (Rankine process). It contains of four processes (see fig. 1), where the exhaust gas heat is used to evaporate the working fluid at high pressure. The superheated vapor is then expanded in an expander which finally provides the mechanical power. Organic working fluids are especially appropriate for the temperature ranges in the vehicle, which is the reason for the name Organic Rankine Cycle (ORC).


                                   Fig. 1. Organic Rankine Process - Components.


                                   Fig. 2. T-s-Diagram of the Rankine Process.

Since the efficiency of the conversion of heat into mechanical power is in general quite low, a maximum efficiency of the process is desirable. This requires an accurate control of the vapor quality at position 3 (see Fig. 1 and 2) with help of the feed pump. The great challenge lies in fact that the exhaust gas heat depends directly on the engine load, which is subject to large transients. In addition, the variable to be controlled must be kept within a small range. The transient operation of the heat engine distinguishes it from similar waste heat recovery applications in power engines which usually operates at steady state.

The main objective of the research project is the solution of this control problem and contains the following topics:

  • Mathematical modelling with parametric, physically motivated approaches (for easy interpretation and adaption to similar system configurations); for this the part feed pump - evaporator - expander is of special importance as here the transient heat input takes place
  • Parameter identification with measurement data
  • Validation
  • Simulation studies and model analysis
  • Application of model-based control methods
  • Implementation and test on the prototype system
  • D. Seitz, O. Gehring, C. Bunz, M. Brunschier & O. Sawodny, “Design of a Nonlinear, Dynamic Feedforward Part for the Evaporator Control of an Organic Rankine Cycle in Heavy Duty Vehicles”, IFAC Symposium on Advances in Automotive Control, Norrköping, Sweden, 2016, doi:10.1016/j.ifacol.2016.08.091
  • D. Seitz, O. Gehring, C. Bunz, M. Brunschier & O. Sawodny, “Dynamic Model of a Multi-Evaporator Organic Rankine Cycle for Exhaust Heat Recovery in Automotive Applications”, IFAC Symposium on Mechatronic Systems, Loughborough, UK, 2016, doi:10.1016/j.ifacol.2016.10.508
  • D. Seitz, O. Gehring & O. Sawodny, “Modellbildung und Identifikation für ein Abgaswärmenutzungssystem in schweren Nutzfahrzeugen”, 7. VDI/VDE Fachtagung AUTOREG Auf dem Weg zum automatisierten Fahren, Baden-Baden, Germany, 2015


To the top of the page