Junichi Kako

Benchmark problem for automotive engine control

The SICE research committee on advanced powertrain control theory provides a V6 engine model for benchmark of advanced control methodologies. The application problem is to design the control of starting the engine model. Currently, over ten groups are tackling the problem. In this paper, the model, the distinctive features of the problem and their midterm results are briefly summarized. Through this activity, we have the confidence that the activity is effective to enhance the collaboration between the control engineering academic society and the automotive industry.

Input Observer-Based Individual Cylinder Air-Fuel Ratio Control: Modelling, Design and Validation

A general approach is proposed to address the cylinder-to-cylinder air-fuel ratio (AFR) balancing control problem using a single universal exhaust gas oxygen (UEGO) sensor. Using the multirate sampling technique, a control-oriented engine model, well combining the fuel delivery dynamics and exhaust gas dynamics, is established. Gas mixing phenomena is characterized by a gas mixing matrix, which is valid both for equal-length and unequal-length exhaust runners. Based on the developed model, the observability of the individual cylinder AFR estimation problem is analyzed using the input observability concept. An input observer using the one-step forward feedback is proposed to design the estimator. Using the estimated AFR, a decoupled PI compensator is designed for each cylinder to compensate the cylinder-by-cylinder variations. Finally, the proposed approach is verified by simulation and experimental results.

Torque Observers Design for Spark Ignition Engines With Different Intake Air Measurement Sensors

This paper presents a design approach and an experimental evaluation of torque observers for Spark Ignition (SI) engines. An engine model simplified from the generally used mean value model is proposed. Based on the developed model, torque observers are designed for engine systems with different intake air measurement sensors. The torque production model is introduced first to give a nominal value of torque estimation. Then, using the rotational dynamic equation of the crankshaft, an online adjusting term is added to modify the nominal estimation. The intake pressure, as an input of the torque production model, is directly measured if a pressure sensor is used. When an air mass flow sensor is adopted instead, an additional intake pressure observer is included in the torque observer. Simulation and experimental results are shown finally to demonstrate the proposed observers.

Individual A/F Estimation and Control With the Fuel–Gas Ratio for Multicylinder IC Engines

In internal combustion engines with multicylinders, balancing the air-fuel ratio (A/F) for individual cylinders is an important issue for providing high-quality torque generation and reducing emissions. This paper addresses the individual A/F control problem for a six-cylinder engine with a single sensor. First, a modeling method to estimate the individual fuel-gas ratio, which immediately provides the A/F, is proposed, and using the estimation model, an adaptive generalized predictive control approach to balancing the individual cylinder characteristics is presented, which targets the static engine-operation mode. The effectiveness of the presented model is validated based on the experimental results, and a comparison with existing models is given based on the experiments. Finally, control performance with the proposed individual A/F control approach is demonstrated with the experiments conducted on a six-cylinder engine test bench.

Introduction to the Benchmark Challenge on SICE Engine Start Control Problem

The speed control during cold start for SI engines is a challenging topic due to the difficulties in handling the dramatic change of the engine dynamics. The SICE Research Committee on Advanced Powertrain Control Theory provides a V6 SI engine model and the benchmark problem of cold start control focusing on the engine speed behavior when the engine model starts. The control design specification is explained in detail and a traditional control is also shown in this paper. Finally, a brief review is given on intermediate results of the challengers.

Development of a New Model Based Air-Fuel Ratio Control System

The second-generation air-fuel ratio control method has been developed to reduce exhaust gas emissions in accordance with the improvements in catalysts. The control system consists of a feedforward control using a fuel behavior model, a feedback control using an universal exhaust gas oxygen (UEGO) sensor and a feedback control utilizing the heated exhaust gas oxygen (HEGO) sensor. Experimental results on actual vehicles show the effectiveness of presented method.

Coordinated Nonlinear Speed Control Approach for SI Engine With Alternator

In the paper, the speed control problem for spark-ignition (SI) engine with alternator is investigated. The problem is addressed from the engine side and the alternator side, respectively, and two coordinated control schemes are proposed for two different operating modes. From the engine side, the torque produced by the alternator is treated as the external disturbance, and a state feedback controller with the magnetic torque feedforward is designed for regulating the engine speed. And from the alternator side, the engine torque ripple is considered as the external disturbance of the alternator rotational dynamics. A disturbance rejection method is proposed to achieve the L 2-gain performance for the speed servo error. Finally, to demonstrate the validity of the proposed approaches, some simulation results will be shown

Individual A/F Control with Fuel-Gas Ratio Estimation for Multi-cylinder IC Engines

In IC engine with multi-cylinder, balancing the air-fuel ratio for individual cylinder is an important issue for providing high quality torque generation and reducing emission. This paper addresses the individual air-fuel ratio control problem for a six-cylinder engine with single sensor. First, a modeling to estimate the individual fuel-gas ratio, which provides the air-fuel ratio immediately, is proposed, and with the estimation model, an adaptive control approach to balancing the individual cylinder characteristics is presented. Under a given operating condition, parameter identification results for the estimation model are obtained by experiment conducted on a real engine test bench. Finally, experimental result with the proposed controller is demonstrated.

Efficient control system development using Model Based Development

Currently, almost all automotive and supplier companies are trying to expand model based development (MBD) for control system development. To shorten the development period, concurrent development of hardware and software is effective. Moreover, the control logic complexity is becoming higher in accordance with the implementation of new sensors, actuators, improved control system performance, and increased the interface with another ECU. One significant process of the MBD is the ECU verification with the plant model installed in HILS. We describe some key points for HILS in hybrid system development.

Individual A/F Estimation and Control for Multi-cylinder IC Engines

In the IC engine with multi-cylinders, estimation of cylinder imbalance characteristics is important for closed loop torque balancing control, in order to obtain high quality torque generation. This paper addressed the individual cylinder A/F estimation and control problem. First, a calculating model for the individual A/F based on multiple sampling data of the single sensor is proposed. Then, using the estimated value of the individual cylinder A/F, a feedback control approach with fuel injection for each cylinder is proposed. Finally, a case study of feedback control is demonstrated with experiment