Research On Performance Measurements And Evaluation As Well As Numerical Simulation Methods Of Automotive Gasoline Engines Under Transient Conditions

Under double pressures from the global energy crisis and environmental pollution reduction, automotive gasoline engines are constantly requested for energy conservation and emissions reduction. Nowadays, gasoline engines are still the major power sources for automobiles, especially for passenger cars. However, the engine performance development, as well as the calibration of operational parameters of gasoline engines are mostly conducted at steady-state conditions in engine labs. Up to now, the research on the engine performance characteristics under transient operation conditions are not deep nor through enough. However, the automotive gasoline engines work much more often under transient conditions than steady state. Therefore engine behavior under transient conditions determines its actual performance, fuel economy and the emission level. So, study on the performances of gasoline engines under the transient conditions has a huge potential for energy saving and emission reduction.At present, there exist many difficulties in detecting and analyzing the transient performance of gasoline engines:some parameters could only be measured utilizing special equipments and with the help of the calibration companies who usually are not willing to, such as the spark timing, the VVT signal, as well as the fuel injection parameters; some parameters could only be measured after large retrofit to the powertrain components, such as installing the torque meter, the dynamical fuel mass flow meter; some parameters can’t be measured directly at all, such as the in-cylinder residual gas fraction. In order to comprehensively test and analyze the performance of the gasoline engine under transient operating conditions, a large number of engine steady-state bench tests were carried out. The general trends were then summarized between the operational parameters with performance parameters of the gasoline engines. A turbocharged direct-injection gasoline engine was then tested for various transient operations and the transient performance parameters were analyzed and compared to the steady state parameters. A study on the numerical simulation of the transient process was then conducted, in this thesis four kinds of gasoline engine simulation models were built, calibrated to the experimental data and cross-over comparison conducted:namely the neural network engine model, the mean value engine model, the finite difference method based engine model and the finite volume method based engine model. In the end, both the vehicle chassis dyno. tests and actual road condition tests were simulated through modeling. The major results and conclusions of this research are summarized as following:(1) Utilizing the advanced AVL dynamic engine dyno. control system, a detailed experimental research is carried out for six gasoline engines, and comprehensive operational and performance parameters were acquired. Then, the general and common trends across the different kinds of gasoline engines were acquired and analyzed for the key performance parameters, i.e. the volumetric efficiency, the relative air fuel ratio, PMEP, FMEP, IMEP and so on, great similarity is shown among the different engines, with engine speed and torque as the dominant influencing factors.(2) The injection timings, ignition timings, VVT timings, as well as the relative air/fuel ratios of the gasoline engines were continuously measured for the transient processes and analyzed online. The transient speed, load, as well as the turbocharger’s transient response characteristics were analyzed and then compared to the steady-state measurements. The results show that the more severely the transient process is, the faster the operation parameters and performance parameters change, the greater the deviation from the steady-state process of the performance parameters would be likely.(3) The theories of the neural network engine model, the mean value engine model, the finite difference method based engine model, as well as the finite volume method based engine model were studied, and all the models were built and calibrated based on the experimental data. The advantages and disadvantages of each of the four different kinds of models were compared. The neural network engine model is shown the fastest, but its accuracy largely relies on experimental data. The mean value engine model can meet the speed requirement for real-time applications, but can’t meet the simulation accuracy requirement of the gasoline engine under transient conditions. The difference method based engine model has a certain capacity of predicting, but its computation speed could hardly catch-up with the real-time response. The GT-Power model can predict the transient engine performance with high accuracy, but the computation speed has no chance to apply real-time. From this comparison, it is shown that the computational speed and the accuracy of the performance prediction have a trade-off relationship.(4) Utilizing the combination method of the transient measurement technique and simulation, the transient performances of a gasoline engine were measured under vehicle road conditions. Engine performance parameters under both the chassis dyno. tests and the actual vehicle road tests were obtained. By analyzing the transient performance of the gasoline engine, one can see that the transient performance of the gasoline engine fluctuates up and down under the sharply changing transient conditions, such as the sharp acceleration, the sharp deceleration, the shifting process, the bypass valve opening process and so on. The main reason is that the spark ignition and fuel injection control strategy of the tested engine under those conditions is questionable. The comparison of the steady-state and transient conditions performances shows that there is a big difference between them under engine full load, as well as the exact value of deviation. In addition, The key influence parameters that lead to those deviations are identified.Through this study, some of the difficult technical problems in gasoline engine performance testing and simulation under transient condition were solved; the general and common trends and influence mechanism of gasoline engine performance under transient condition were obtained; the deviation and the influencing parameters of gasoline engine performances between the steady-state and transient conditions were revealed; some optimization methods of transient performance of gasoline engine are provided. All of them provide the theoretical guidance and technical support to optimize for the transient performance of the gasoline engine under actual road conditions.