Data Driven Nonlinear Dynamic Models for Predicting Heavy-Duty Diesel Engine Torque and Combustion Emissions

Speaker: Gökhan Alcan

Robotic Seminar Series. Second Session – 21st February 2020, 15:00-16:00, room 2103, TUAS Building, Aalto University.


Diesel engines’ reliable and durable structures, high torque generation capabilities at low speeds, and fuel consumption efficiencies make them irreplaceable for heavy-duty vehicles in the market. However, inefficiencies in the combustion process result in the release of emissions to the environment. In addition to the restrictive international regulations for emissions, the competitive demands for more powerful engines and increasing fuel prices obligate heavy-duty engine and vehicle manufacturers to seek for solutions to reduce the emissions while meeting the performance requirements. In line with these objectives, remarkable progress has been made in modern diesel engine systems such as air handling, fuel injection, combustion, and after-treatment. However, such systems utilize quite sophisticated equipment with a large number of calibratable parameters that increases the experimentation time and effort to find the optimal operating points. Therefore, a dynamic model-based transient calibration is required for an efficient combustion optimization which obeys the emission limits, and meets the desired power and efficiency requirements.
This study is about developing optimization-oriented high fidelity nonlinear dynamic models for predicting heavy-duty diesel engine torque and combustion emissions. Combustion process in diesel engines is a challenging MIMO dynamical system which can be modeled using a data driven approach where the air path and the fuel path input channels can be utilized to estimate the indicated engine torque and the exhaust emissions. Once a reliable model of the process is obtained, it can be employed in powertrain development for the dynamic model-based transient calibration of combustion and aftertreatment control. For data generation, a new dynamic DoE (design of experiments) is proposed where chirp and ramp-hold signals are used to excite input channels. The proposed approach is a strong alternative to the steady-state experiment based approaches to reduce the testing time considerably and improve the modeling accuracy in both steady-state and transient conditions. Modeling of the indicated torque and NOx emissions of a diesel engine by utilizing the Nonlinear Finite Impulse Response (NFIR) and Nonlinear Autoregressive with Exogenous Inputs (NARX) models will then be presented. Furthermore, it will be shown that soot emission can be quite successfully modeled using modern network structures such as Gated Recurrent Unit (GRU) and Long Short Term Memory (LSTM) networks. These networks outperform the classical NARX type recurrent structures in terms of generating more accurate and smoother soot predictions.

  • Alcan, G., Aran, V., Unel, M., Yilmaz, M., Gurel, C., Koprubasi, K., and Otosan, F. (2020). Optimization-Oriented High Fidelity NFIR Models for Estimating Indicated Torque in Diesel Engines. International Journal of Automotive Technology, 21(3), 729-737.
  • Alcan, G., Yilmaz, E., Unel, M., Aran, V., Yilmaz, M., Gurel, C., and Koprubasi, K. (2019). Estimating soot emission in diesel engines using gated recurrent unit networks. IFAC-PapersOnLine, 52(5), 544-549.
  • Alcan, G., Unel, M., Aran, V., Yilmaz, M., Gurel, C., and Koprubasi, K. (2019). Predicting NOx emissions in diesel engines via sigmoid NARX models using a new experiment design for combustion identification. Measurement, 137, 71-81.
  • Alcan, G., Unel, M., Aran, V., Yilmaz, M., Gurel, C., and Koprubasi, K. (2018). Diesel engine NOx emission modeling using a new experiment design and reduced set of regressors. IFAC-PapersOnLine, 51(15), 168-173.