Shiwei Wang

Adaptive neural network model based predictive control for air-fuel ratio of SI engines

The dynamics of air manifold and fuel injection of the spark ignition engines are severely nonlinear. This is reflected in nonlinearities of the model parameters in different regions of the operating space. Control of the engines has been investigated using observer-based methods or sliding-mode methods. In this paper, the model predictive control (MPC) based on a neural network model is attempted for air-fuel ratio, in which the model is adapted on-line to cope with nonlinear dynamics and parameter uncertainties. A radial basis function (RBF) network is employed and the recursive least-squares (RLS) algorithm is used for weight updating. Based on the adaptive model, a MPC strategy for controlling air-fuel ratio is realised to a nonlinear simulation of the engines, and its control performance is compared with that of a conventional PI controller. A reduced Hessian method, a new developed sequential quadratic programming (SQP) method for solving nonlinear programming (NLP) problems, is implemented to speed up the nonlinear optimisation in MPC.

Adaptive RBF network for parameter estimation and stable air-fuel ratio control

In the application of variable structure control to engine air-fuel ratio, the ratio is subjected to chattering due to system uncertainty, such as unknown parameters or time varying dynamics. This paper proposes an adaptive neural network method to estimate two immeasurable physical parameters on-line and to compensate for the model uncertainty and engine time varying dynamics, so that the chattering is substantially reduced and the air-fuel ratio is regulated within the desired range of the stoichiometric value. The adaptive law of the neural network is derived using the Lyapunov method, so that the stability of the whole system and the convergence of the networks are guaranteed. Computer simulations based on a mean value engine model demonstrate the effectiveness of the technique.

A new development of internal combustion engine air-fuel ratio control with second-order sliding mode

A novel application of a second-order sliding mode control (SMC) scheme to the air-fuel ratio (AFR) control of automobile internal combustion engines is developed in this paper. In this scheme, the sliding surface S[x(t)] is steered to zero in finite time by using the discontinuous first-order derivative of a control variable u c (t), and the corresponding actual control variable u c (t) turns out to be continuous, which significantly reduces the undesired chattering. Its sliding gain is adjusted by a novel radial basis function network based adaptation method derived using the Lyapunov theory. It not only avoids handling the unavailable parameters and variables, but also saves the unnecessary manual adjusting time of the second-order SMC. The proposed method is applied to a widely used engine benchmark, the mean value engine model for evaluation. The simulation results show substantially improved AFR control performance compared with the conventional SMC.

Adaptive neural network model based predictive control of an internal combustion engine with a new optimization algorithm

Abstract This paper presents an application of adaptive neural network modelling and model-based predictive control (MPC) for an engine simulation. A radial basis function (RBF) neural network trained by a recursive least-squares (RLS) algorithm is compared with the network with fixed parameters and demonstrated to be more suitable for modelling the crankshaft speed, the intake manifold pressure, and the manifold temperature. Based on the obtained adaptive neural network model, an MPC strategy for controlling the crankshaft speed is realized successfully. A reduced Hessian method, a newly developed sequential quadratic programming (SQP) method for solving non-linear programming (NLP) problems, is implemented to solve the non-linear optimization in MPC. Some important modifications are proposed for the algorithm settings in this research to make the reduced Hessian method more appropriate for the adaptive neural network model based predictive control strategy of internal combustion (IC) engines.

An Application of Second-Order Sliding Mode Control for IC Engine Fuel Injection

A novel application of a second-order sliding mode control (SMC) scheme for the fuel injection control of automobile internal combustion (IC) engines is developed in this paper. In this scheme, the sliding surface, S(x(t)) is steered to zero in finite time by using the discontinuous first-order derivative of a control variable, u(t) and the corresponding actual control variable, u(t) turns out to be continuous, which significantly reduces the undesired chattering of u(t) deduced in the conventional SMC. By using a twisting algorithm to solve an introduced auxiliary problem, a simple and robust control law is derived. It avoids involving some immeasurable parameters and system uncertainties and therefore has strong potential for practical applications. The scheme is applied to a widely used engine benchmark, the mean value engine model for evaluation. The simulation results show substantially improved air fuel ratio characteristics by the proposed scheme compared with the conventional SMC

Neural Network in Stable Adaptive Control Law for Automotive Engines

This paper proposes to use a radial basis function (RBF) neural network in realising an adaptive control law for air/fuel ratio (AFR) regulation of automotive engines. The sliding mode control (SMC) structure is used and a new sliding surface is developed in the paper. The RBF network adaptation and the control law are derived using the Lyapunov method so that the entire system stability and the network convergence are guaranteed. The developed method is evaluated by computer simulation using the well-known mean value engine model (MVEM) and the effectiveness of the method is proved.

ADAPTIVE MODELLING AND PREDICTIVE CONTROL OF AN IC ENGINE

Abstract This paper presents modelling of internal combustion (IC) engine with adaptive neural networks. A radial basis function network model with both centres and weights adapted and a model with only weights adapted are compared with a fixed parameter model. The developed models are used in model based predictive control (MPC) to form an adaptive nonlinear MPC scheme and applied to engine speed tracking control. The modelling and control are based on a generic mean value engine model and consists of three submodels that describe the fuel mass flow dynamics, the intake manifold filling dynamics and the crankshaft speed. Adaptive MPC is shown superior over the fixed parameter model based control.