Model-based decoupling and disturbance-rejection control method for homogeneous charge compression ignition (HCCI)

Abstract

The invention discloses a model-based decoupling and disturbance-rejection control method for HCCI. The method comprises model-based decoupling control, active disturbance rejection control (ADRC), feed-forward control and self-adaptation compensation control for valve mechanism action delay. The method comprises the steps of designing a decoupling compensator on the basis of a control model of the HCCI, and converting an HCCI system into a plurality of independent single-input single-output (SISO) systems; conducting inversion on transfer functions of all SISO systems to obtain a feed-forward controller; designing ADRC controllers for all SISO systems respectively, and observing and compensating deviations of the model and random disturbance of the outside in real time; and estimating the action delay of a valve mechanism in real time, actively delaying an oil injection action, and obtaining the response speed which is the same as that of the valve mechanism. By the aid of the method, the modeling burden of the HCCI control model and the standard workload of controller parameters can be reduced greatly, the robustness for engine working condition variations is high, and the ignition process control is stable.

Description

technical field [0001] The invention relates to the field of homogeneous compression ignition (HCCI) combustion control, in particular to a control of HCCI combustion realized by using a fully variable valve train (VVA) and a negative valve overlap angle (NVO) strategy. Background technique [0002] Homogeneous compression ignition (HCCI) combustion can reduce engine fuel consumption and emissions simultaneously, and is a potential combustion method for future engines. However, since there is no direct ignition control method and is extremely sensitive to boundary conditions, HCCI engines are prone to misfire or detonation during operation, making its engineering applications face control difficulties. [0003] In order to realize the effective control of HCCI, two types of control methods are usually adopted in the world: the traditional model-free PID control method and the model-based control method. [0004] PID control is the earliest control method adopted by people, ...

AI Technical Summary

Problems solved by technology

However, the amount of calculation of MPC is usually large. Due to the limitation of real-time calculation, MPC is still difficult to be applied in mass-produced real-vehicle ECUs.

[0006] Therefore, a major difficulty in HCCI control lies in the contradiction between the control model and the model-based control method: the existing control methods usually rely on accurate models, while the complex and accurate modeling of the HCCI combustion process is difficult

[0007] In response to the above problems, Chiang et al. (Chia-Jui Chiang; Stefanopoulou, A.G.; Jankovic, M., "Nonlinear Observer-Based Control of Load Transitions in Homogeneous Charge Compression Ignition Engines," Control Systems Technology, IEEE Transactions on, vol.15 , no.3, pp.438, 448, May 2007, doi:10.1109 / TCST.2007.894637) designed a nonlinear feedback controller, the results show that although this control method has a good anti-disturbance effect, but for some unknown The disturbance (such as the random disturbance of the intake air temperature) still does not have a good suppression ability

Widd (Widd, A.; Hsien-Hsin Liao; Gerdes, J.C.; Tunestal, P.; Johansson, R.," Control of exhaust recompression HCCI using hybrid model predictive control," American Control Conference (ACC), 2011, vol. ,no.,pp.420,425,June292011-July12011) and Erlien (Erlien,S.M.,A.F.Jungkunz and J.C.Gerdes,MULTI-CYLINDER HCCI CONTROL WITH CAM PHASER VARIABLE VALVE ACTUATION USING MODEL PREDICTIVE CONTROL.) proposed an integrated output observation deviation The composite MPC controller of the controller reduces the fluctuation range of CA50 under steady-state conditions, but the huge computational burden of MPC is still an unavoidable problem

LQG or MPC based on the combination of multiple sub-control models (A.Widd, "Predictive control of hcci engines using physical models," Department of Automatic Control, Lund University, Sweden, Licentiate Thesis TFRT--3246--SE, May2009; Hsien- Hsin Liao, Ravi, N., Jungkunz, A.F., Jun-Mo Kang, Gerdes, J.C., "Representing recompression HCCI dynamics with a switching linear model," American Control Conference (ACC), 2010, vol., no., pp. 3803-3808, June 302010-July22010.) Control is a possible solution, but the required sub-control model requires a high amount of experimental data and is difficult to obtain

Robust control (Souder, Jason Scott, Ph.D, "Closed-Loop Control of a Multi-Cylinder HCCI Engine," UNIVERSITY OF CALIFORNIA, BERKELEY, 2004, 186 pages; 3165567.) has also been tried by numerous researchers, but for Guaranteed robustness, robust controller design is usually very conservative, the regulation speed of the transient process is not fast enough

In addition, adaptive control (Chia-Jui Chiang; Chih-Cheng Chou; Jian-Hong Lin, "Adaptive control of homogeneous charge compression ignition (HCCI) engines," American Control Conference (ACC), 2012, vol., no., pp.2066,2071,27-29June2012.) have also been tried by researchers, but for multi-input and multi-output systems, the calculation of adaptive control is very cumbersome, and it is not realistic to apply it in the current mass-produced vehicle ECU

Images