Development of a Parametric Model for Burn Rate Estimation in Direct Injection Diesel Engine

Kamat, Prakash; Schneemann, Arne; Patel, Sanjay; Kasper, Werner; Flohr, Andreas; Rajasekar, Gurusathya; Patne, Mayur

doi:10.4271/2019-26-0035

Cited by 2 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The physics-based BD model was developed by parameterizing the physical contribution of the laminar flame speed and the rate of diffusion and late combustion in the burn profile [31][32][33]. A modified form of the van Tiggelen and Tabaczynski model capturing the laminar flame speed's contribution was utilized, and is shown in Equation (8).…”

Section: Physics-based Bd Modelmentioning

confidence: 99%

“…Figure 11 shows the model structure of the physics-based CA50 and IMEP models. The Wiebe function is a simple and widely accepted 0D combustion model, which is highly amenable for control; however, it generally lacks high predictive capability [31], and cannot adapt to changes that may occur as an engine ages [8]. Consequently, the Wiebe function becomes less effective if it is globally parameterized for the ECU [14,31].…”

Section: Physics-based Bd Modelmentioning

confidence: 99%

“…The Wiebe function is a simple and widely accepted 0D combustion model, which is highly amenable for control; however, it generally lacks high predictive capability [31], and cannot adapt to changes that may occur as an engine ages [8]. Consequently, the Wiebe function becomes less effective if it is globally parameterized for the ECU [14,31]. In order to improve its predictive robustness for wider engine operating conditions, the Wiebe function was made adaptable by the use of a recursive least-squares model parameter optimization.…”

Section: Physics-based Bd Modelmentioning

confidence: 99%

“…The hybridized model framework's outputs from the adaptive Wiebe model were treated as transitional dummy output steps and then passed on to the TLM as a subroutine for a final model fine-tuning step. This could be achieved by minimizing the error between the Wiebe model predictions and the target optimal trajectory, as shown in Equation (31). This realization was based on the neural network transfer learning model's premise that, given stochastic variations in the distributions in the oracle of an ANN, the stored knowledge from previous distributions can be utilized to enhance the predictions in the presence of stochastic perturbations [22].…”

Section: Ca50 Brann Transfer Learning Model (Tlm)mentioning

confidence: 99%

“…In Equation (31), ϑ Gaussian represents the attributed external stochastic changes in the environment, and Θ final becomes the new overall objective function that accounts for stochastic variations in the modeling environment. Computational tractability for real-time implementation was ensured in this step by utilizing the BRANN, which circumvents unnecessary model validation steps.…”

Section: Ca50 Brann Transfer Learning Model (Tlm)mentioning

confidence: 99%

See 4 more Smart Citations

A Hybrid Physics-Based and Stochastic Neural Network Model Structure for Diesel Engine Combustion Events

Ankobea-Ansah

Hall

2022

Vehicles

View full text Add to dashboard Cite

Estimation of combustion phasing and power production is essential to ensuring proper combustion and load control. However, archetypal control-oriented physics-based combustion models can become computationally expensive if highly accurate predictive capabilities are achieved. Artificial neural network (ANN) models, on the other hand, may provide superior predictive and computational capabilities. However, using classical ANNs for model-based prediction and control can be challenging, since their heuristic and deterministic black-box nature may make them intractable or create instabilities. In this paper, a hybridized modeling framework that leverages the advantages of both physics-based and stochastic neural network modeling approaches is utilized to capture CA50 (the timing when 50% of the fuel energy has been released) along with indicated mean effective pressure (IMEP). The performance of the hybridized framework is compared to a classical ANN and a physics-based-only framework in a stochastic environment. To ensure high robustness and low computational burden in the hybrid framework, the CA50 input parameters along with IMEP are captured with a Bayesian regularized ANN (BRANN) and then integrated into an overall physics-based 0D Wiebe model. The outputs of the hybridized CA50 and IMEP models are then successively fine-tuned with BRANN transfer learning models (TLMs). The study shows that in the presence of a Gaussian-distributed model uncertainty, the proposed hybridized model framework can achieve an RMSE of 1.3 × 10−5 CAD and 4.37 kPa with a 45.4 and 3.6 s total model runtime for CA50 and IMEP, respectively, for over 200 steady-state engine operating conditions. As such, this model framework may be a useful tool for real-time combustion control where in-cylinder feedback is limited.

show abstract