【作者】

Andreas Hoepfner,Ingo Koops,Achim Przymusinski

【摘要】

论文已在上海2013年CIMAC大会上发表，论文版权归CIMAC所有。 The future emission legislation for marine applications stipulates a considerable reduction of nitrogen oxide pollutants. In 2016, the IMO will enact the emission standard Tier lIll for emission controlled areas (ECA). Whereas engines operated at open sea will keep the less stringent emission level Tier |Ⅱ. Dual fuel engines provide the possibility to run in different propulsion modes which are suitable for the permit-ted emission levels. On open sea, a dual fuel engine can be operated IMO Tier lI compliant in Diesel mode with cost efficient heavy fuel oil. In ECA, dual fuel engines run on natural gas to fulfill the upcoming emission regulation IMO Tier ll. The operation and inter-action of Diesel and gas mode result in complex control structures for dual fuel engines. In each propulsion mode, the engine shall provide a reliable operation with an excellent transient behavior. At the same time, engine durability and safety aspects may not be neglected. Hence, the development of an engine control system (ECS) for dual fuel engines represents a notable challenge. Generally, the development process for engine control systems includes the elaboration of control software which has to be tested and calibrated on engine test bed. This approach contains two major drawbacks. First, the availability of proto-type engines is limited especially at the initial development phase. Second, engine test bed operation is ex-pensive and increases the overall costs for the engine development. Within this paper, AVL shows a possibility to improve the development process of ECS software. A close interaction of the tools Simulink from The MathWorks and AV's BOOST RT is used to set up a simulation environment BOOST RT provides the plant model of the engine. Its thermodynamics and flow characteristics are simulated in the time domain using steady and unsteady OD and quasi-dimensional component models. Responsiveness of engine sensors as well as the behavior of actuators is also included in the BOOST RT model. ECS control concepts are modeled in Simulink. The BOOST RT inter-face in Simulink allows a model-in-the-loop (MiL) simulation of ECS software and engine plant model. Furthermore, BOOST RT is capable to generate a surrogate of the engine plant model. This also can be done for a part of the model and in combination with the remaining crank angle resolved model. Thus, the computing time can be tailored for real-time applications on a hardware-in-the-loop (HiL) test bench considering also slow computers or complex models. On the Simulink side, autocode and flash tools are used to implement the control functions into the target ECS hardware. As a consequence, a HiL simulation of the control functions running on the ECS hardware is set up to test and calibrate the ECS application software. In summary, the coupling of BOOST RT and Simulink allows a validation of ECS control functions in an early engine development phase. Engine test bed time can be reduced through the use of control functions with a relatively high maturity level and calibration efforts are minimized by pre-parameterized application datasets. The advantages of this improved development process will be shown by means of chosen engine control functions used in Diesel and gas mode of a dual fuel engine.

【会议名称】

第27届CIMAC会议

【会议地点】

上海

【下载次数】

1

返回