Hybrid adaptive control used in diesel engine speed regulation

Abstract

In this paper, a hybrid adaptive control to realize an entire enveloppe speed regulation with fast operating condition changes and input saturation constraint is proposed. It is based on a simplified nonlinear parameter varying diesel engine model which includes variable transport delays in fueling input. The proposed controller which consists of a derived piecewise continuous controller based jumping-up sub-controller and a friction and load torque observer based Lyapunov function stabilization sub-controller guarantees the closed-loop system stability and demonstrates perfectly unknown friction and load torque disturbance rejection capacities. Comparing with a linear parameter varying control, the proposed hybrid adaptive control has better performance in terms of settling time, steady state error and it allows no static state fueling input oscillation and nearly zero speed drop in presence of significant external load torque changes. Moreover, the proposed control has a very simple structure and can be implemented easily for real-time system control. I. INTRODUCTIONDiesel engines are recognized as the most common and preferred solution for the distributed power generating systems in numerous applications, such as automotive vehicles, ships, cranes, and electric power generators due to theirs superior fuel efficiency and durability compared to the other types of internal combustion engines [26], [8], [9]. The diesel engine design objectives are to obtain the best possible performance in terms of speed regulation (such as maintaining good transient and steady-state performance and ensuring good external friction and load disturbance rejection capabilities), fuel consumption economy and exhaust gas emissions. Often, these are conflicting and coupling objectives, e.g. engine speed control has significant impact on engine performance, fuel consumption economy and exhaust gas emission [2]. The nonlinear and open loop possible unstable nature (which is most probably due to the positive feedback from the fueling input pumped from the engine output shaft), time varying delays (due to the interval between the cylinder firings in forward-loop) and multiple operating modes makes the engine speed control a very challenging problem.Until now, different approaches have been tested for engine speed control:• methods based on linearization model which are tuned and scheduled controllers at different operating points: optimal gain scheduling controller [3]

Real-Time Self-Tuning Speed Controller: Performance Comparison between Engine Fuelled with Palm Methyl Esters and Petroleum Diesel

“…The success of real-time implementation of self-tuning speed controller developed in this study proved that the controller algorithm developed based on the simple linear dynamic model is valid and reliable. Compared to other recent studies [11,12,13] which investigate the effective engine speed regulation only in simulation, this research give more thorough testing of the controller capability and reliability. The performance engine speed control strategy adopted to regulate at set-point speed and the ability for disturbance rejection is comparable to the fuzzy self-tuning PID controller designed for diesel generator studied by [14].…”

Robust model-based switching MIMO air handling control of turbocharged lean-burn SI natural gas variable speed engines

“…These include model-independent control strategies, such as classical Proportional-Integral-Derivative (PID) control, sliding mode control (SMC) and model-based control approaches, such as H ∞ control, adaptive control and model predictive control (MPC). In Wang et al, 7 the authors propose an adaptive controller consisting of a sub-controller to guarantee performance in terms of settling time and a sub-controller to stabilize the speed at steady-state and reduce steady-state error. In Chamsai et al, 8 an adaptive PID controller using SMC is developed.…”