Search Results

An electrically heated catalyst (EHC) in the three-way catalyst (TWC) aftertreatment system of a gasoline internal combustion engine (ICE) provides cold engine start exhaust pollutant emission reduction potential. The EHC can be started before switching on the ICE, thereby offering the possibility to pre-heat (PRH) the TWC, in the absence of exhaust flow. The EHC can also provide post engine start heat (PSH) when the heat is accompanied by exhaust mass flow over the TWC. A mixed heating strategy (MXH) comprises both PRH and PSH. All three strategies are evaluated under a range of engine start variations using an ICE-exhaust aftertreatment (EATS) simulation framework. It is driven by an engine speed-torque requested trace, with an engine-out emissions model focused on cold-start, engine heating and catalyst heating engine measures and a physics- based EATS with EHC model.

The three-way-catalyst (TWC) is an essential part of the exhaust aftertreatment system in spark-ignited powertrains, converting nearly all toxic emissions to harmless gasses. The TWC’s conversion efficiency is significantly temperature-dependent, and cold-starts can be the dominating source of emissions for vehicles with frequent start/stops (e.g. hybrid vehicles). In this paper we develop a thermal TWC model and calibrate it with experimental data. Due to the few number of state variables the model is well suited for fast offline simulation as well as subsequent on-line control, for instance using non-linear state-feedback or explicit MPC. Using the model could allow an on-line controller to more optimally adjust the engine ignition timing, the power in an electric catalyst pre-heater, and/or the power split ratio in a hybrid vehicle when the catalyst is not completely hot.

Look ahead information can be used to improve the powertrain’s fuel consumption while efficiently controlling exhaust emissions. A passenger car propelled by a Euro 6d capable diesel engine is studied. In the conventional approach, the diesel powertrain subsystem control is rule based. It uses no information of future load requests but is operated with the objective of low engine out exhaust emission species until the Exhaust After-Treatment System (EATS) light off has occurred, even if fuel economy is compromised greatly. Upon EATS light off, the engine is operated more fuel efficiently since the EATS system is able to treat emissions effectively. This paper presents a supervisory control structure with the intended purpose to operate the complete powertrain using a minimum of fuel while improving the robustness of exhaust emissions.

A two-state forward dynamic programming algorithm is evaluated in a series hybrid drive-train application with the objective to minimize fuel consumption when look-ahead information is available. The states in the new method are battery state-of-charge and engine speed. The new method is compared to one-state dynamic programming optimization methods where the requested generator power is found such that the fuel consumption is minimized and engine speed is given by the optimum power-speed efficiency line. The other method compared is to run the engine at a given operating point where the system efficiency is highest, finding the combination of engine run requests over the drive-cycle that minimizes the fuel consumption. The work has included the engine torque and generator power as control signals and is evaluated in a full vehicle-simulation model based on the Volvo Car Corporation VSIM tool.

Existing battery parameter model structures are evaluated by estimating model parameters on real driving data applying standard system identification methods. Models are then evaluated on the test data in terms of goodness of fit and RMSE in voltage predictions. This is different from previous battery model evaluations where a common approach is to train parameters using standardized tests, e.g. hybrid pulse-power capability (HPPC), with predetermined charge and discharge sequences. Equivalent linear circuit models of different complexity were tested and evaluated in order to identify parameter dependencies at different state of charge levels and temperatures. Models are then used to create voltage output given a current, state of charge and temperature. The average accuracy of modelling the DC bus voltage provides a model goodness of fit average higher than 90% for a single RC circuit model.