Search Results

This paper presents a novel model-based algorithm which is able to detect and isolate major faults assigned to the gas exchange path of a gasoline engine both in the intake and exhaust sides. The diagnostics system is developed for detection and isolation of these faults: air leakage fault between the compressor and the air throttle, exhaust manifold pressure sensor fault, wastegate stuck-closed fault and wastegate stuck-open fault. Sliding mode observers (SMOs) are the core detection algorithms utilized in this work. A first order SMO is designed to estimate the turbocharger rotational dynamics. The wastegate displacement dynamics coupled to the exhaust manifold pressure dynamics is estimated using a second order SMO. Verified with experimental data from a modern TC gasoline engine running in a test cell, the two sliding mode observers are then used in a strategy to detect the faults in the gas exchange path.

Proper operation of an internal combustion engine is required by demands of a vehicle driver and governmental legislations. Therefore it is necessary to monitor, within an online technique, the engine and detect any fault which disrupts its normal operation. In this paper, the air-charge path, as a key element in a turbocharged engine, is monitored for an air leakage fault. At first, a robust algorithm to estimate unmeasured turbocharger rotational speed is presented. The sliding mode methodology is used to design the estimator which is shown to be robust to the compressor modeling uncertainties. The estimation error from the sliding mode observer (SMO) is then used to detect abnormal behavior of the turbocharger along with the engine due to a leakage fault in the air-charge path. Experimental results from a modern turbocharged SI engine indicate the designed monitoring technique is able to detect a leakage fault, of 7 mm or higher sizes, in the air-charge path.

One of the main problems associated with design of fault detection (FD) strategy is availability of first generation engine. To solve this problem a methodology based on a virtual engine platform is proposed in this paper. This approach allows designing an FD algorithm in early stages of engine development. The application of this methodology is illustrated on a modern turbocharged gasoline engine by investigating the effect of a leakage in the exhaust manifold. Experimental results show good ability of the virtual engine platform to predict the effect of the leakage fault on the engine performance. Moreover unexpected results of exhaust manifold leakage effects are presented which are very useful for designing a leak detection strategy.

In this paper, development of a MATLAB-based computer environment for optimization of EMS calibration is presented. The objective is to eliminate the complicated and tedious calibration process on chassis dynamometer or at least reduce the time required. In this way, first a black-box model of engine is developed. Then this model is linked to ADVISOR -the renowned vehicle simulation software- to obtain an integrated engine-driveline model. This model is used in the optimization process, which includes different optimization techniques.