Development and Validation of a Virtual Sensor for Estimating the Maximum in-Cylinder Pressure of SI and GCI Engines

Scocozza, Guido Federico; Silvagni, Giacomo; Brusa, Alessandro; Cavina, Nicolò; Ponti, Fabrizio De; Ravaglioli, Vittorio; Cesare, Matteo De; Panciroli, Marco; Benedetti, Cristian

doi:10.4271/2021-24-0026

Cited by 9 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, pressure sensors on-board installation is still uncommon, mainly due to problems related to their reliability and cost. To overcome these, over the past years, several remote combustion sensing methodologies were developed to extract information about the combustion process, such as SOC, CA50, within the engine cycle through the real-time processing of signals coming from low-cost sensors (such as speed sensors or accelerometers) mounted on the engine [33][34][35][36][37][38][39]. One of the most studied approaches for the combustion indexes estimation is based on a accelerometer sensor [33,[37][38][39][40][41][42] which has shown a reliable correlation between engine block vibrations and the way in which the energy release process takes place in the combustion chamber.…”

Section: Introductionmentioning

confidence: 99%

Accelerometer-based SOC estimation methodology for combustion control applied to Gasoline Compression Ignition

Silvagni

Ravaglioli

Ponti

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The European Community’s recent decision to suspend the marketing of cars with conventional fossil-fueled internal combustion engines from 2035 requires new solutions, based on carbon-neutral technologies, that ensure equivalent performances in terms of reliability, trip autonomy, refueling times and end-of-life disposal of components compared to those of current gasoline or diesel cars. The use of bio-fuels and hydrogen, which can be obtained by renewable energy sources, coupled with high-efficiency combustion methodologies might allow to reach the carbon neutrality of transports (net-zero carbon dioxide emissions) even using the well-known internal combustion engine technology. Bearing this in mind, experiments were carried out on compression ignited engines running on gasoline (GCI) with a high thermal efficiency which, in the future, could be easily adapted to run on a bio-fuel. Despite the well-reported benefits of GCI engines in terms of efficiency and pollutant emissions, combustion instability hinders the diffusion of these engines for industrial applications. A possible solution to stabilize GCI combustion is the use of multiple injections strategies, typically composed by 2 early injected fuel jests followed by the main injection. The heat released by the combustion of the earlier fuel jets allows to reduce the ignition delay of the main injection, directly affecting both delivered torque and center of combustion. As a result, to properly manage GCI engines, a stable and reliable combustion of the pre-injections is mandatory. In this paper, an estimation methodology of the start of combustion (SOC) position, based on the analysis of the signal coming from an accelerometer sensor mounted on the engine block, is presented (the optimal sensor positioning is also discussed). A strong correlation between the SOC calculated from the accelerometer and that obtained from the analysis of the rate of heat release (RoHR) was identified. As a result, the estimated SOC could be used to feedback an adaptive closed-loop combustion control algorithm, suitable to improve the stability of the whole combustion process.

show abstract

Section: Introductionmentioning

confidence: 99%

Accelerometer-based SOC estimation methodology for combustion control applied to Gasoline Compression Ignition

Silvagni

Ravaglioli

Ponti

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…Differently from the physical and semiphysical models [10][11][12][13][14][15], a data-driven approach could be helpful, since the processes at the basis of emission formation, such as combustion and turbulence, are quite difficult to model analytically [16] and require much time to run in virtual environments. Despite 0-D models [17][18][19][20][21] being computationally efficient, the analytical formulation of the physical phenomena can be difficult to determine when many independent variables are affecting the output. Some applications of machine learning aimed at emission modeling are already present in the literature [22][23][24].…”

Section: Introductionmentioning

confidence: 99%

Comparative Evaluation of Data-Driven Approaches to Develop an Engine Surrogate Model for NOx Engine-Out Emissions under Steady-State and Transient Conditions

et al. 2022

Self Cite

View full text Add to dashboard Cite

In this paper, a methodology based on data-driven models is developed to predict the NOx emissions of an internal combustion engine using, as inputs, a set of ECU channels representing the main engine actuations. Several regressors derived from the machine learning and deep learning algorithms are tested and compared in terms of prediction accuracy and computational efficiency to assess the most suitable for the aim of this work. Six Real Driving Emission (RDE) cycles performed at the roll bench were used for the model training, while another two RDE cycles and a steady-state map of NOx emissions were used to test the model under dynamic and stationary conditions, respectively. The models considered include Polynomial Regressor (PR), Support Vector Regressor (SVR), Random Forest Regressor (RF), Light Gradient Boosting Regressor (LightGBR) and Feed-Forward Neural Network (ANN). Ensemble methods such as Random Forest and LightGBR proved to have similar performances in terms of prediction accuracy, with LightGBR requiring a much lower training time. Afterwards, LightGBR predictions are compared with experimental NOx measurements in steady-state conditions and during two RDE cycles. Coefficient of determination (R2), normalized root mean squared error (nRMSE) and mean average percentage error (MAPE) are the main metrics used. The NOx emissions predicted by the LightGBR show good coherence with the experimental test set, both with the steady-state NOx map (R2 = 0.91 and MAPE = 6.42%) and with the RDE cycles (R2 = 0.95 and nRMSE = 0.04).

show abstract

“…Several examples of condition monitoring methods applied to diesel engines can be found in the literature. For light and medium-duty engines, the need for cost reduction pushes toward solutions based on low-cost or virtual sensors [25,26], which also allow the estimation of the engine-out emissions [27] and the combustion indexes [28][29][30] or predicting failures [29]. All these approaches are becoming increasingly common thanks to the increase in digitalization [5,29] and the advanced data analysis made available by the cloud technologies [16].…”

mentioning

confidence: 99%

“…Better results, however, can be obtained by monitoring the combustion process using the in-cylinder pressure information. Both indirect [30] and direct [31,32] sensing is achievable, and the latter guarantees better accuracy. The efficiency increase for a production heavy-duty engine by means of a proper combustion control system needs an accurate evaluation of the combustion indexes, which is also useful for the implementation of predictive maintenance strategies [33].…”

mentioning

confidence: 99%

Enhancement of Heavy-Duty Engines Performance and Reliability Using Cylinder Pressure Information

et al. 2023

Self Cite

View full text Add to dashboard Cite

Sustainability issues are becoming increasingly prominent in applications requiring the use of heavy-duty engines. Therefore, it is important to cut the emissions and costs of such engines to reduce the carbon footprint and keep the operating expenses under control. Even if for some applications a battery electric equipment is introduced, the diesel-equipped machinery is still popular thanks to the longer operating range. In this field, the open pit mines are a good example. In fact, the Total Cost of Ownership (TCO) of the mining equipment is highly impacted by fuel consumption (engine efficiency) and reliability (service interval and engine life). The present work is focused on efficiency enhancements achievable through the application of a combustion control strategy based on the in-cylinder pressure information. The benefits are mainly due to two factors. First, the negative effects of injectors aging can be compensated. Second, cylindrical online calibration of the control parameters enables the combustion system optimization. The article is divided into two parts. The first part describes the toolchain that is designed for the real-time application of the combustion control system, while the second part concerns the algorithm that would be implemented on the Engine Control Unit (ECU) to leverage the in-cylinder pressure information. The assessment of the potential benefits and feasibility of the combustion control algorithm is carried out in a Software in the Loop (SiL) environment, simulating both the developed control strategy and the engine behavior (Liebherr D98). Our goal is to validate the control algorithm through SiL simulations. The results of the validation process demonstrate the effectiveness of the control strategy: firstly, cylinder disparity on IMEP (+/−2.5% in reference conditions) is virtually canceled. Secondly, MFB50 is individually optimized, equalizing Pmax among the cylinders (+/−4% for the standard calibration) without exceeding the reliability threshold. In addition to this, BSFC is reduced by 1% thanks to the accurate cylinder-by-cylinder calibration. Finally, aging effects or fuel variations can be implicitly compensated, keeping optimal performance thorough the engine life.

show abstract

Development and Validation of a Virtual Sensor for Estimating the Maximum in-Cylinder Pressure of SI and GCI Engines

Cited by 9 publications

References 21 publications

Accelerometer-based SOC estimation methodology for combustion control applied to Gasoline Compression Ignition

Accelerometer-based SOC estimation methodology for combustion control applied to Gasoline Compression Ignition

Comparative Evaluation of Data-Driven Approaches to Develop an Engine Surrogate Model for NOx Engine-Out Emissions under Steady-State and Transient Conditions

Enhancement of Heavy-Duty Engines Performance and Reliability Using Cylinder Pressure Information

Contact Info

Product

Resources

About