【作者】 童宝宏；

【导师】 桂长林； 陈华；

【作者基本信息】 合肥工业大学 ， 机械设计及理论， 2007， 博士

【摘要】 从传统上来说，在内燃机设计和使用中对润滑系统的基本要求是供给摩擦副充分的润滑油和润滑系统不出故障，以保障内燃机可靠工作。然而，当代社会的经济发展对内燃机润滑系统设计提出了更新的要求：在继续满足和进一步改善传统要求的同时，还需满足不断提高的排放限制等环保性、节能性要求。环保性方面，有关试验研究证实，润滑油颗粒排放的形成与内燃机润滑油消耗有很大的关系。节能性方面，试验表明，一台中速内燃机每年的机油费用约占燃油费用的4％。同时，润滑系统中机油泵的驱动功率也具备有降低燃油耗的一种潜力。传统的经验方法和单一工况下的研究已经不能满足对现代内燃机润滑系统设计的要求，需要有一个更完善的理论分析方法用于对润滑系统的工作特性进行全面考察。论文针对目前国际上采用的内燃机润滑系统网络法设计中的几个关键问题进行了探讨，并结合样机，利用改进的网络化分析方法和试验研究对润滑系统多工况下的工作特性进行了考察。论文首先将润滑系统网络法分析中各种简化的轴承流量特性计算方法分别作了阐述。在此基础上，提出了一种新的表达方法用于描述轴承润滑油流量与压降关系。通过实例计算将这些方法得到的结果同流体动力润滑分析得到的结果进行了比较研究。同时，还以某内燃机主轴承为例，阐述了一种计入热变形影响时的内燃机主轴承热流体动力润滑分析方法，并进行数值仿真计算。通过对考虑和不考虑热变形因素影响时得到的分析结果进行比较研究，探讨了热变形对内燃机主轴承热流体动力润滑性能的影响。论文还对径向滑动轴承在启动和运行过程中加速、加载时轴承温度变化的历程特性进行了试验研究。结合润滑油在内燃机管道中流动的实际情况，研制了用于测量管道内润滑油流动特性的专用试验装置。并根据相应的试验结果数据，对目前润滑系统网络法分析中用于描述润滑油在内燃机管道内流动特性的几种计算方法进行了对比分析。同时，还通过试验考察了管径、管长和油温等因素对润滑油在管道中流动特性的影响情况。分别对样机机油泵、机油滤清器和冷却器等典型部件的流量特性进行了试验研究，并利用BP神经网络的非线性映射建模功能，结合相关试验数据建立了表达内燃机机油泵供油特性以及滤清器、冷却器流量特性的仿真模型，探讨了神经网络模型建立过程中训练样本的选择等问题，并通过试验验证了所建模型的正确性。对于所建立的内燃机润滑系统网络模型，利用质量守恒、能量守恒原理，建立了各网络节点上的流压关系：提出了利用基于最小二乘算法的最优化方法来求解网络法润滑系统分析中建立的非线性方程组，并结合样机进行实例仿真分析，重点考察了内燃机工作过程中润滑系统流量分配以及压力分布状况。论文最后，结合内燃机多工况下的试验研究对润滑系统网络模拟法的应用情况进行了检验。论文对样机的磨合试验和多工况下的示功图测试工作进行了阐述，并对测试结果进行了分析；结合示功图测试数据，完成了多工况下内燃机轴承载荷的计算和分析工作。论文还将几种轴承流量计算方法分别应用于润滑系统的网络法模拟分析过程中，并结合试验对各个模拟分析结果进行了对比研究，分析了各种轴承流量计算方法对润滑系统设计中机油泵选择的影响；通过网络模拟分析，结合试验测试结果，论文重点考察了多工况下内燃机转速、负荷和润滑油温度等因素对润滑系统主油道压力特性的影响情况，以及多工况下内燃机缸内润滑油消耗情况。论文的创新点主要有，将目前网络法润滑系统分析中各种轴承流量特性的计算方法进行了对比分析，提出了一种计入热变形因素影响的轴承润滑分析方法，并分析了各种轴承流量计算方法对润滑系统设计中机油泵选择的影响。研制了用于测量管道内润滑油流动特性的试验装置，对目前润滑系统网络法分析中的各种管道内润滑油流动特性计算方法进行了试验检测。建立了表达内燃机机油泵供油特性以及滤清器、冷却器流量特性的BP神经网络仿真模型，并提出可以通过采用正交设计法选取机油泵神经网络模型的训练样本来提高部件建模效率。提出一种利用Pro／E提供的Mechanism模块和Matlab软件对内燃机轴承载荷进行联合仿真求解的简便有效方法。提出了利用基于最小二乘算法的优化方法来求解网络法分析中表达样机润滑系统网络模型的非线性方程组，实现了对润滑系统网络化仿真分析方法的改进应用。针对样机，对润滑系统在全负荷、部分负荷等多工况下的工作特性进行了网络法仿真分析，并结合主油道压力特性试验结果对仿真结果进行了检验，证明了润滑系统改进网络化分析方法的可行性和正确性。更多还原

【Abstract】 Basic demands for lubrication system in traditional designing process of internal combustion engine (ICE) is to supply sufficient lubricant to friction parts and to work without failure in order to make sure that ICE can operate credibly. However, the development of modern society have put forward more new demands on the design of ICE’ lubrication system. Requirements from environmental protection and energy saving area such as emission control regulations should be resolved when traditional one continues to be fulfilled and improved. On environmental protection, results from experimental study show that there’s strong connection between the formation of particle emission and oil consumption from lubrication system. In the energy saving area, there’s experimental study which show that the oil fee consumed by a mean ICE in one year is about 4 percent of the fuel fee. And same time, there’s potential to reduce fuel consumption within pump’s driving power in lubrication system. Traditional method with experiences and research in single operating condition can not satisfy the demands of modern design of ICE’s lubrication system. There should have a more perfect theory used to review the operating property of ICE’s lubrication system. Several key points in the process of network analyzing with lubrication system were discussed in this paper. Combining with a model engine, a improved network analysis method was used to analyze the property of lubrication system in multiple operating conditions.Several kinds of simplified methods for flow characteristic of oil in bearing are described firstly in this paper. And a new idea was also put forward. By analyzing with an example, all those methods have been compared with the method by hydrodynamic lubricating analysis. A research on the lubrication of ICE’s main bearing considering thermal deformation effects are presented combining with a model engine’s main bearing. Through comparing the results from analysis by with and without considering thermal deformation effects, effects from thermal deformation factors on the thermal hydrodynamic analysis of ICE’s main bearing were discussed. An experimental study on the temperature of journal bearing during start-up and operation process with variation of working condition was also carried out in this paper.Combining with the true state of oil that flowing in lubrication system’s pipe, a special equipment used to test the character of oil flowing in pipe was constructed. With the results from test, several measures that currently used to describe the character of oil flowing in pipe were analyzed by comparing. And effects on the character of oil flowing in pipe from many factors, such as diameter of pipe, length of pipe and oil temperature, were studied by this test.Working characters of oil pump, oil filter and cooler were investigated separately by experiments. And these characters were modeled by using the function of modeling nonlinearly of BP neural net combining with the results from test. How to choose learning samples in the process of modeling with neural network was discussed here and was also examined by the test.With theories of conversation of energy and mass, relationships of drop-flowing on each net node were established for network model of sample engine’s lubrication system. Optimization method based on least-squares algorithm was put forwarded to solve the nonlinearly equations that constructed in the network analysis of lubrication system. And a simulation analysis on model engine was presented to review emphatically the distributions of oil pressure and flowing of lubrication system in ICE’s operation conditions.In the end of this paper, application of network simulation method on lubrication system was examined combining with the experimental research under multiple operating condition of ICE. Running-in tests and indicator diagram tests under multiple operating condition of ICE. were presented, and all results of these tests have been analyzed. Calculations and analysis of bearing loads under multiple operating condition of ICE were done with the results of indicator diagram tests. Simplified methods for flow characteristic of oil in bearing were applied in the process of network analysis of lubrication system. Through comparing with test results, influences of these simplified methods on the selection of oil pump in the process of ICE’s lubrication system design process were analyzed. Effects on the pressure property of main oil gallery from those factors such as rotate speed, load and oil temperature were examined emphatically under multiple operating condition of ICE. And oil consumption under multiple operating condition of ICE was also discussed here.Innovative points of this paper can be summarized as following. Several kinds of simplified methods used to describe flow characteristic of oil in bearing are compared firstly. And influences of these simplified methods on the selection of oil pump in the design process of ICE’s lubrication system were investigated. Research on the lubrication of ICE’s main bearing considering thermal deformation effects are presented. A special equipment used to test the character of oil flowing in pipe was constructed. Several measures that currently used to describe the character of oil flowing in pipe were analyzed by comparing test results. Working characters of oil pump, oil filter and cooler were modeled by using the function of modeling nonlinearly of BP neural net. An efficient way to calculate bearing loads by union application of mechanism module in Pro/E and Matlab software was introduced. Optimization method based on least-squares algorithm was put forwarded to solve the nonlinearly equations that constructed in the network analysis of lubrication system. Network analysis on property of lubrication system of sample engine was carried out under both full load condition and part load condition. And by the experimental study on the pressure property of main oil gallery, feasibility of improved network method was testified.更多还原