Search Results

During the development of an Internal Combustion Engine-based powertrain, traditional procedures for control strategies calibration and validation produce huge amount of data, that can be used to develop innovative data-driven applications, such as emission virtual sensing. One of the main criticalities is related to the data quality, that cannot be easily assessed for such a big amount of data. This work focuses on an emission modeling activity, using an enhanced Light Gradient Boosting Regressor and a dedicated data pre-processing pipeline to improve data quality. First thing, a software tool is developed to access a database containing data coming from emissions tests. The tool performs a data cleaning procedure to exclude corrupted data or invalid parts of the test. Moreover, it automatically tunes model hyperparameters, it chooses the best set of features, and it validates the procedure by comparing the estimation and the experimental measurement.

This work focuses on the development and validation of a data-driven model capable of predicting the maximum in-cylinder pressure during the operation of an internal combustion engine, with the least possible computational effort. The model is based on two parameters, one that represents engine load and another one the combustion phase. Experimental data from four different gasoline engines, two turbocharged Gasoline Direct Injection Spark Ignition, a Naturally Aspirated SI and a Gasoline Compression Ignition engine, was used to calibrate and validate the model. Some of these engines were equipped with technologies such as Low-Pressure Exhaust Gas Recirculation and Water Injection or a compression ignition type of combustion in the case of the GCI engine. A vast amount of engine points were explored in order to cover as much as possible of the operating range when considering automotive applications and thus confirming the broad validity of the model.

Due to the recent anti-pollution policies, the performance increase in Spark Ignition (SI) engines is currently under the focus of automotive manufacturers. This trend drives control systems designers to investigate accurate solutions and build more sophisticated algorithms to increase the efficiency of this kind of engines. The development of a control strategy is composed of several phases and steps, and the first part of such process is typically spent in defining and investigating the logic of the strategy. During this phase it is often useful to have a light engine simulator, which allows to have robust synthetic combustion data with a low calibration and computational effort. In the first part of this paper, a description of the control-oriented ANalytical Engine SIMulator (ANESIM) is carried out.

Water Injection (WI) has become a key technology for increasing combustion efficiency in modern GDI turbocharged engines. In fact, the addition of water mitigates significantly the occurrence of knock, reduces exhaust gas temperatures, and opens the possibility to reach optimum heat release phasing even at high load. This work presents the latest development of a model-based WI controller, and its experimental validation on a GDI TC engine. The controller is based on a novel approach that involves an analytic combustion model to define the spark advance (SA) required to reach a combustion phase target, considering injected water mass effects. The calibration and experimental validation of the proposed controller is shown in detail in the paper.

Knock is a non-deterministic phenomenon and its intensity is typically defined by a non-symmetrical distribution, under fixed operating conditions. A statistical approach is therefore the correct way to study knock features. Typically, intrinsically deterministic knock models need to artificially introduce Cycle-to-Cycle Variation (CCV) of relevant combustion parameters, or of cycle initial conditions, to generate different knock intensity values for a given operating condition. Their output is limited to the percentage of knocking cycles, once the user imposes an arbitrary knock intensity threshold to define the correlation between the number of knocking events and the Spark Advance (SA). In the first part of the paper, a statistical analysis of knock intensity is carried out: for different values of SA, the probability distributions of an experimental Knock Index (KI) are self-compared, and the characteristics of some percentiles are highlighted.