Takehiko Kowatari

Throttle-control algorithm for improving engine response based on air-intake model and throttle-response model

An electric-throttle-control actuator (ETC) is a device for control of the air mass flow to an engine cylinder. As adaptive-cruise-control and direct-fuel-injection systems become popular, the market of ETC has become larger. The ETC is controlled so that the engine torque follows the target value. Between the change of the control signal to the ETC and the engine-torque response, two delays exist-the delay in the throttle response and manifold filling. These delays must be compensated to improve the engine response. In this paper, a throttle-control algorithm for improving engine response is proposed. This algorithm compensates these two delays based on the response model. The response of the manifold pressure was experimented in two cases, when the ETC was controlled by a step input and when the throttle was controlled by the developed algorithm. The experimental results show that the rise time of the manifold pressure response decreased to one-tenth by the developed algorithm. Because the engine torque is proportional to the manifold pressure, it can be concluded that the torque response improved by compensating the two delay factors.

A throttle control algorithm for improving engine response based on the characteristics of electronic-throttle-control actuator

A throttle control algorithm which compensates for the delay in manifold filling is proposed. This algorithm takes into account the control characteristics of an electronic-throttle-control actuator and improves the engine output response. The performance of the algorithm was evaluated in terms of rise time of the manifold pressure and duration of switching control. The results showed that the rise time and the duration of switching control was less than about 100 ms.

Method of Predicting Hill Climbing Performance in Batteryless Motorized-Four-Wheel-Drive System

We have predicted the hill climbing performance of a batteryless motorized-four-wheel-drive (M4WD) system. With this type of M4WD system, the engine drives the front wheels, and an electric motor drives the rear wheels. The electrical power to the motor is supplied directly by the M4WD generator (water-cooled alternator for M4WD use only) driven by the engine. The system is small and simple due to this batteryless powertrain. However, predicting hill climbing performance is complex due to the power tradeoff between the front and rear wheels. We clarify nonlinear mechanical and electrical power flow models, and discuss how our M4WD vehicle simulator using these power flow models is able to predict hill climbing performance. The performance we predicted agreed well with the actual performance of a prototype vehicle.

A signal processing algorithm for compensating for back flow effect in intake air mass measurement in internal combustion engines

We developed a method to compensate for back flow occurring during intake air mass measurement in internal combustion engines using frequency characteristics of the signal from an air flow sensor (AFS). The parameters of the AFS signal, which has close correlation with back flow ratio, were explored. The parameters were searched in terms of simulation, mathematical analysis and experiments. The simulation results showed that the normalized intensities f/sub 2//f/sub 1/ and f/sub 3//f/sub 1/ have close correlation with back flow ratio, where f/sub 1/ is the intensity of the base frequency of the intake air pulsation, f/sub 2/ is the intensity of twice of the frequency, and f/sub 3/ is the intensity of three times of the frequency. These two correlations were robust to the specification of the induction system, engine revolution speed, and load. The back flow ratio could be estimated using normalized frequency intensity f/sub 2//f/sub 1/ and f/sub 3//f/sub 1/ within an accuracy of 5%. The correlations between f/sub 2//f/sub 1/, f/sub 3//f/sub 1/ and the back flow ratio were validated by mathematical analysis. The correlations were also evaluated by experiments using a four cycle engine. The experimental results showed that the back flow ratio was estimated using f/sub 2//f/sub 1/, f/sub 3//f/sub 1/ within an accuracy of 5%.

Signal Processing for Error Reduction of Automotive Hot-Wire Air Flow Sensor. Reducing Understimation-Error of Bobbin Type Hot-Wire under Pulsatile Flow.

The bobbin type hot-wire (H/W) element of an air flow sensor has a tend ency to underestimate the actual flow rate when subjected to large-amplitude pulsatile flow. We investigated the causes of this measurement error and propose a new dynamic model for the bobbin type H/W element, which dynamically combines its static nonlinear output characteristic and its dynamic response. By inverting our bobbin type H/W dynamic model, we have developed a signal processing method to correct the measurement error of the bobbin type H/W. Measurements using this signal processing method show that it reduces the bobbin type H/W measurement error.