Machine Learning Integrated with Model Predictive Control for Imitative Optimal Control of Compression Ignition Engines

Norouzi, Armin; Shahpouri, Saeid; Gordon, David C.; Winkler, Alexander; Nuss, Eugen; Abel, Dirk; Andert, Jakob; Shahbakhti, Mahdi; Koch, Charles Robert

doi:10.1016/j.ifacol.2022.10.256

Cited by 9 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where f ( x ( k ) ) and g ( x ( k ) ) are coefficients of the constraint function which depend on the modeled plant states x ( k ) . Linear plant dynamics developed in our previous study are used in the optimization 11…”

Section: Deep Rlmentioning

confidence: 99%

“…where A and B are state-space matrices developed using a autoregressive with extra input (ARX) model 11 as follows…”

Section: Deep Rlmentioning

confidence: 99%

“…where A and B are state-space matrices developed using a autoregressive with extra input (ARX) model 11 as follows where the constrained output y(k), states x(k), and control actions u(k) are defined as follows…”

Section: Ddpg Agents Algorithmmentioning

confidence: 99%

“… 1 The use of a feedback controller, especially a model-based optimal controller, is a promising method to help solve the ever-increasing calibration efforts. Model-based methods such as linear quadratic regulator (LQR), 4 sliding mode controller (SMC), 5 , 6 adaptive, 7 , 8 and model predictive control (MPC) 9 – 11 have been previously investigated for engine applications. The two main drawbacks of these model-based controllers are their sensitivity to model accuracy and the required runtime especially for online optimization.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Safe deep reinforcement learning in diesel engine emission control

Norouzi

Shahpouri

Gordon

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part I:

Self Cite

View full text Add to dashboard Cite

A deep reinforcement learning application is investigated to control the emissions of a compression ignition diesel engine. The main purpose of this study is to reduce the engine-out nitrogen oxide [Formula: see text] emissions and to minimize fuel consumption while tracking a reference engine load. First, a physics-based engine simulation model is developed in GT-Power and calibrated using experimental data. Using this model and a GT-Power/Simulink co-simulation, a deep deterministic policy gradient is developed. To reduce the risk of an unwanted output, a safety filter is added to the deep reinforcement learning. Based on the simulation results, this filter has no effect on the final trained deep reinforcement learning; however, during the training process, it is crucial to enforce constraints on the controller output. The developed safe reinforcement learning is then compared with an iterative learning controller and a deep neural network–based nonlinear model predictive controller. This comparison shows that the safe reinforcement learning is capable of accurately tracking an arbitrary reference input while the iterative learning controller is limited to a repetitive reference. The comparison between the nonlinear model predictive control and reinforcement learning indicates that for this case reinforcement learning is able to learn the optimal control output directly from the experiment without the need for a model. However, to enforce output constraint for safe learning reinforcement learning, a simple model of system is required. In this work, reinforcement learning was able to reduce [Formula: see text] emissions more than the nonlinear model predictive control; however, it suffered from slightly higher error in load tracking and a higher fuel consumption.

show abstract

Section: Deep Rlmentioning

confidence: 99%

“…where A and B are state-space matrices developed using a autoregressive with extra input (ARX) model 11 as follows…”

Section: Deep Rlmentioning

confidence: 99%

Section: Ddpg Agents Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Safe deep reinforcement learning in diesel engine emission control

Norouzi

Shahpouri

Gordon

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part I:

Self Cite

View full text Add to dashboard Cite

show abstract

“…the crank angle at which 50% of the fuel's energy is released), indicated mean effective pressure (IMEP), 20 torque, 21 combustion phasing, 22 cyclic variations 23 and NOx emissions. 24 Machine Learning (ML)-based surrogate modeling of internal combustion engines (ICE) has been widely used for a broad range of applications. [25][26][27][28][29] Data-driven ML approaches, in particular, are popular for building ICE surrogate models; such approaches include neural networks (NN), [30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45] Support Vector Machines (SVM), [46][47][48][49] Gaussian Processes (GPs, [50][51][52][53][54][55][56][57][58][59][60] also known as kriging 61 ), and other learning models.…”

Section: Introductionmentioning

confidence: 99%

A misfire-integrated Gaussian process (MInt-GP) emulator for energy-assisted compression ignition (EACI) engines with varying cetane number jet fuels

Narayanan,

Ji,

Sapra

et al. 2024

International Journal of Engine Research

View full text Add to dashboard Cite

For energy-assisted compression ignition (EACI) engine propulsion at high-altitude operating conditions using sustainable jet fuels with varying cetane numbers, it is essential to develop an efficient engine control system for robust and optimal operation. Control systems are typically trained using experimental data, which can be costly and time consuming to generate due to setup time of experiments, unforeseen delays/issues with manufacturing, mishaps/engine failures and the consequent repairs (which can take weeks), and errors in measurements. Computational fluid dynamics (CFD) simulations can overcome such burdens by complementing experiments with simulated data for control system training. Such simulations, however, can be computationally expensive. Existing data-driven machine learning (ML) models have shown promise for emulating the expensive CFD simulator, but encounter key limitations here due to the expensive nature of the training data and the range of differing combustion behaviors (e.g. misfires and partial/delayed ignition) observed at such broad operating conditions. We thus develop a novel physics-integrated emulator, called the Misfire-Integrated GP (MInt-GP), which integrates important auxiliary information on engine misfires within a Gaussian process surrogate model. With limited CFD training data, we show the MInt-GP model can yield reliable predictions of in-cylinder pressure evolution profiles and subsequent heat release profiles and engine CA50 predictions at a broad range of input conditions. We further demonstrate much better prediction capabilities of the MInt-GP at different combustion behaviors compared to existing data-driven ML models such as kriging and neural networks, while also observing up to 80 times computational speed-up over CFD, thus establishing its effectiveness as a tool to assist CFD for fast data generation in control system training.

show abstract