Search Results

This paper presents details of the development of, and experimental results from, an internal combustion engine dynamic cylinder heat transfer control device for use on single-cylinder research engines. This device replicates the varying temperature profile and heat transfer distribution circumferentially around a cylinder in a multicylinder engine. This circumferential temperature distribution varies around a cylinder because of the location of, or lack of coolant passages around the cylinders, and varies from cylinder to cylinder as a result of the flow of the coolant through these passages as it accumulates thermal energy and increases in temperature. This temperature distribution is important because it directly affects the NO emissions from each cylinder, as will be seen in the experimental results.

A new method for instantaneous friction estimation in firing internal combustion engines has been developed in the Powertrain Control Research Laboratory (PCRL) at the University of Wisconsin - Madison. This Synthetic Variable approach, which has previously been used for combustion quality diagnostics, focuses on carefully measuring instantaneous engine speed and other easily measurable engine variables and combining them with dynamic models of other engine processes. This approach numerically strips away the dynamic effects that mask friction effects on engine speed and reveals friction estimates with clarity. This information could be useful for engine designers and developers to assist in accurately understanding the sources of instantaneous friction within the running engine. The friction results from these studies have been very encouraging.

A new high-bandwidth transient test system is being developed that allows a single cylinder research engine to be tested under conditions nearly identical to those experienced by individual cylinders of a multi-cylinder engine. The system consists of two unique test components: a high bandwidth transient hydrostatic dynamometer capable of simulating the combustion and rotational dynamics of a multi-cylinder engine, and an air intake simulator that pulls air from the intake manifold plenum to simulate air induction characteristics of the multi-cylinder engine. The system makes it possible to evaluate preliminary engine control strategies and perform more detailed hardware development early in a development program when representative multi-cylinder engines may not be available. This reduces engine development time and allows the transition to multi-cylinder engine hardware to proceed with fewer design changes and less cost.

A new dynamometer system has been developed to improve the accuracy of tests that are run with a single cylinder version of a multi-cylinder engine. The dynamometer control system calculates the inertial torque and combustion torque that would normally be generated in a multi-cylinder engine. The system then applies the torque from the missing cylinders of the engine with the dynamometer. A unique high bandwidth hydraulic system is utilized to accurately apply these torque pulses. This allows the single-cylinder engine to have the identical instantaneous speed trajectory as the multi-cylinder engine, to test the single-cylinder engine at all engine speeds including very low speed operation, and to now do transient speed and load testing. Not only will this dramatically extend the capabilities of current single-cylinder engine test systems, but may open up new areas of research due to its transient testing capabilities.

This paper presents a diagnostics methodology that has applications to internal combustion engines as well as other dynamic devices. Included is an overview of the theoretical foundation of the approach, discussions on its application to internal combustion engine diagnostics, and experimental engine data showing the application of this methodology. Also included are the recent developments addressing issues of the effect of motoring compression and expansion work on crankshaft speed fluctuations and the resulting torque estimation. The methodology consists of a hard-wired nonlinear to linear transformation of engine variables that allow all subsequent diagnostics and control calculations to use linear mathematics, which significantly simplifies the size and complexity of the engine control and diagnostics strategy and code.

A nonlinear model-based method for engine misfire detection has been proposed in the earlier work [2]. Many possible reasons for persistent cylinder misfire (e.g., a burned inlet valve or other faults), however, still need to be identified. Identification of engine misfire enables engineers, vehicle operators or technicians to trace the cause of misfire and to identify the faulty components in the engine. Also, a cylinder-by-cylinder manifold model will provide a considerably more accurate estimate of individual cylinder air flows. This paper develops a model-based nonlinear intake manifold pressure observer and an algorithm to identify a burned inlet valve. The proposed manifold model is a runner-by-runner model. A nonlinear observer using this runner-by-runner model can estimate the plenum pressure and all individual runner pressures and their subsequent flows. The estimate is then used as an indication of a faulty inlet valve, one of the possible causes of engine misfire.

Many researchers have studied and developed methods for on-board engine combustion misfire detection in production vehicles. Misfiring can damage the catalytic converter within a short time and can lead to increased emission levels. For that reason, the on-board detection of engine misfire is one requirement of the On Board Diagnosis II (OBDII) Regulation and a recent interest for many researchers. One object in this paper is to propose a misfire detection method for multi-cylinder SI engines. The detection is achieved by examining the estimated cylinder pressures and combustion heat release rates in engine cylinders. The Sliding Observer methodology is applied in these estimations. This detection method provides a reliable and low-cost way to diagnose engine misfires. The other object of the paper is to eliminate large estimation errors due to system unobservability and reconstruct cylinder pressures.

Nonlinear observer theories are applied to the engine estimation problem in order to reconstruct engine states based on the measured engine variables, and dynamic mean torque production and cylinder-by-cylinder engine models. Engine cylinder and intake manifold pressures are two important factors in engine control and diagnostics. This paper discusses how to design nonlinear engine cylinder pressure and intake manifold pressure observers that have good robustness and estimation accuracy. Sliding mode theory in Variable Structure Systems (VSS) have shown good performance and been successfully applied to many nonlinear systems. Accordingly, sliding observers are selected for this study.

Stoichiometric air-fuel ratio control during transient operation requires an accurate estimate or measurement of the instantaneous air flow rate in an engine. Two methods are commonly used for determining engine air flow rate: air-mass sensing and conventional “speed-density.” The lead air flow information provided by air-mass sensors helps compensate for manifold filling and other fuel system delays. However, the high cost (and sometimes lower reliability) of air-mass sensors has led many manufacturers to continue to use the less accurate speed-density method for determining air flow rate. This paper develops a model-based nonlinear manifold pressure observer that estimates the flow rates at the throttle and the intake ports of an engine using speed-density type sensors. The throttle flow rate estimate can be used instead of an air-mass sensor to provide the lead information necessary for accurate transient air-fuel control on TBI engines.

This paper investigates the dynamic errors between the commonly used two-lump mass connecting rod model and the actual connecting rod model for the internal combustion engine. Because of the errors between the actual rod inertia and this simplified two-lump mass model, incorrect engine dynamics and internal forces are often predicted. In this paper, the magnitudes of force differences related to errors of connecting rod inertia are presented for various engines at different engine operating speeds. A method to predict the maximum side force and its maximum deviation is presented. And the technique to minimize variability in connecting rod mass and moment of inertia, as well as minimizing errors in the lumped mass model commonly used in industry are also introduced to avoid incorrect engine dynamics and internal forces.