An Application of Sliding Mode Control to Model-Based Reinforcement Learning

2019 ◽  
Author(s):  
Aaron Thomas Parisi
Keyword(s):  
Reinforcement Learning ◽  
Sliding Mode Control ◽  
Sliding Mode ◽  
Model Based ◽  
Mode Control

Nonlinear observer based sliding mode control and oxygen fraction estimation for diesel engine

2017 ◽  
Vol 40 (7) ◽  
pp. 2227-2239 ◽  
Author(s):  
Haoping Wang ◽  
Qiankun Qu ◽  
Yang Tian
Keyword(s):  
Diesel Engine ◽  
Sliding Mode Control ◽  
Oxygen Concentration ◽  
Control Method ◽  
Sliding Mode ◽  
Observer Design ◽  
Nonlinear Observer ◽  
Linear Matrix ◽  
Model Based ◽  
Mode Control

In this paper, a nonlinear observer based sliding mode control (NOSMC) approach for air-path and a model-based observer for oxygen concentration in the diesel engine equipped with a variable geometry turbocharger and exhaust gas recirculation is introduced. We propose a less conservative observer design technique for Lipschitz nonlinear systems using Ricatti equations. The observer gains are obtained by solving the linear matrix inequality (LMI). Then a robust nonlinear control method, sliding mode control is applied for the states of intake and exhaust manifold pressure and compressor mass flow rate for the sake of the minimization of emissions. The proposed NOSMC controller is applied on a mean value model of turbocharged diesel engine. Besides this, a model-based observer is developed to estimate the oxygen concentration in the intake and exhaust manifolds owing to its significance in reducing emissions of diesel engines. The validation and efficiency of the proposed method are demonstrated by AMESim and Matlab/Simulink co-simulation results.

Sign in / Sign up

Export Citation Format

Share Document