Research On Dynamic Match Of Engine And Hydrodynamic Torque Converter

In this dissertation, the dynamic matching problem of diesel engine and hydrodynamictorque converter mounted on the armored vehicles was studied. The dynamic matchingtarget is to ensure that vehicles get the optimal power performance, economy orcomprehensive performance during the whole vehicles’ running process or for specificrequirements. This study can also provide a theoretical basis for the follow-up dynamicanalysis and dynamic control of drive systems.Using the analytic hierarchy process (AHP) with fuzzy quantification for qualitativeindexes and the improved radar chart method with uniqueness of evaluation results, thestatic and dynamic matching performance evaluation system was constructed. Matchingperformmce evaluation indexes were developed, including the dynamic and economicindicators. The dynamic indicator consists of torque output coefficient and power outputcoefficient with corresponding weighing factor, and economic indicator consists of the fuelconsumption rate and its weighing factor. Through the expert decision judgment matrix inAHP, the relative weight of each indicator was determined.The static and dynamic characteristic test of a diesel engine was conducted. A methodof neural networks combination adopting multi-step linear interpolation was put forward toidentify and fit the diesel engine performance, by which the diesel engine neural networkmodel was constructed based on the limited test data. This diesel model can obtain theacceleration and deceleration dynamic diesel engine output characteristics during differentwork conditions with varing throttle or varing load. Model predictions compared with thetest data showed that, the model’s prediction accuracy reached above85.9%.Combining shear stress transport model and single flow channel with periodicboundary model, a hydrodynamic torque converter’s mixed flow channel CFD simulationmodel was set up. The mixed flow channel model can identify whether stator wheel isidling according to the working wheels’ force status, and effectively calculate torqueconverter performance in its torque-converting, coupling and power-reversing workconditions. The static and dynamic characteristics of torque converter test data proved thecredibility of the mixed flow channel CFD model. This CFD model’s prediction accuracy is above87.3%for steady performance prediction and above82.9%for dynamic performanceprediction.The dynmiic matching equations covering the diesel engine’s dynamic input-outputneural network fitting functions and the torque converter’s dynamic original characteristicsequations were analyzed. The diesel engine and torque converter’s dynamic matchingperformance simulation model was established, and the dynamic matching simulationcalculation and analysis was carried on. The dynamic matching test was conducted,including external characteristic test, universal characteristics test and two kinds of typicaldynamic work conditions test. Test data proved the credibility of the dynamic matchingperformance simulation model. Model predictions compared with the test data showed thatthe moders prediction accuracy was above79.2%.The influencing factors of dynamic matching performance were analyzed by design ofexperiments(DOE), through which the main effect and interaction effect of the torqueconverter effective diameter md the intermediate transmission ratio on the optimal dynamicmatching performance was obtained. Under the ETC test cycle, the dynamic matchingbetween diesel engine and torque converter was carried on according to the requirement ofthe optimal dynamic power performance, economy and comprehensive performance. Effectof both the throttle position and torque converter lock-up ratio on the dynamic matchingwas also studied. Dynamic matching optimization model for hydraulic transmission systemwas constructed. Using this optimization model, under a series of torque converter lock-upspeed ratios, the intermediate transmission ratio was optimized through multi-objectiveoptimization design with archive-based micro genetic algrithm(AMGA).