Deep learning based model predictive control for compression ignition engines

نویسندگان

چکیده

Machine learning (ML) and a nonlinear model predictive controller (NMPC) are used in this paper to minimize the emissions fuel consumption of compression ignition engine. In work machine is applied two methods. first application, ML identify for implementation control optimization problems. second as replacement NMPC where learns optimal action by imitating or mimicking behavior controller. study, deep recurrent neural network including long–short term memory (LSTM) layers performance an industrial 4.5 liter 4-cylinder Cummins diesel This then implementation. Then, scheme deployed clone developed LSTM integration, novel augmenting hidden cell states problem. The LSTM-NMPC imitative compared with calibrated Engine Control Unit (ECU) experimentally validated engine simulation platform. Results show significant reduction Nitrogen Oxides (NO x ) slight decrease injected quantity while maintaining same load. addition, has similar but orders magnitude computation time.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Emissions Control Challenges for Compression Ignition Engines

As emission standards continue to evolve, it is clear that future engine control strategies will involve the integration of combustion optimization, fuel refinement and advanced exhaust after-treatment technologies. The West Virginia University (WVU) Center for Alternative Fuels Engines and Emissions (CAFEE) continues to engage the challenge of future regulation in a multi-pronged approach, inv...

متن کامل

DeepMPC: Learning Deep Latent Features for Model Predictive Control

Designing controllers for tasks with complex nonlinear dynamics is extremely challenging, time-consuming, and in many cases, infeasible. This difficulty is exacerbated in tasks such as robotic food-cutting, in which dynamics might vary both with environmental properties, such as material and tool class, and with time while acting. In this work, we present DeepMPC, an online real-time model-pred...

متن کامل

Model-based cold-start speed control scheme for spark ignition engines

This paper presents a model-based control scheme to the cold-start speed control in spark ignition (SI) engines. The multi-variable control algorithm is developed with the purpose of improving the transient performance of the starting engine speed: the control inputs are the fuel injection, the throttle and the spark advance (SA), while the engine speed and the air mass flow rate are the measur...

متن کامل

Development of a Flame Propagation Model for Dual-Fuel Partially Premixed Compression Ignition Engines

A new combustion model is developed and applied to simulate combustion in dual-fuel engines in which the premixed natural gas is ignited by the combustion flame initiated by a diesel spray. The model consists of a diesel auto-ignition model and a flame propagation model. A G-equation model previously developed to simulate SI engine combustion was incorporated with an auto-ignition model to simu...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Control Engineering Practice

سال: 2022

ISSN: ['1873-6939', '0967-0661']

DOI: https://doi.org/10.1016/j.conengprac.2022.105299