【作者】 黄豪中；

【导师】 苏万华；

【作者基本信息】 天津大学 ， 动力机械及工程， 2007， 博士

【摘要】 均质压燃着火(HCCI)燃烧概念给出了实现内燃机高效、低污染的新途径,是目前国际内燃机界研究的热点。但是在柴油机上实现HCCI燃烧仍有许多理论和技术上的难题。本文以MULINBUMP-HCCI复合燃烧系统为研究对象,通过实验和化学动力学数值模拟相结合的方法,开展柴油HCCI燃烧基础理论及其控制技术的研究。为解决三维CFD与化学反应动力学模型耦合导致计算量太大和计算时间太过冗长的问题,本文首先总结前人工作,从高分子烃在低温和高温化学反应的机理出发,利用敏感性分析、主要组分分析以及准稳态假定三种方法,选出关键地适用于HCCI发动机燃烧模拟研究的正庚烷化学反应,然后,基于微种群遗传算法和化学动力学计算软件CHEMKIN中的SENKIN程序,开发出一个通用性很强的动力学参数自动优化标定软件,利用该软件优化正庚烷化学反应中的关键动力学参数,最后将这两部分合并构筑成一个新的包含40种组分和56个反应的正庚烷化学反应动力学简化模型,称为“SKLE简化模型”。该模型可以较准确预报柴油HCCI发动机的CO、HC和NOx排放,对着火时刻和主要中间组分浓度的预报精度与美国劳伦斯国家实验室的详细模型相当吻合,而计算机时间只是后者的1/1000,为耦合化学动力学模型的多维模型计算创造了必要条件。建立了一整套从零维单区、多区到三维耦合化学反应动力学并能反映HCCI燃烧特征的数值模拟模型。应用这些模型开展了柴油HCCI燃烧边界条件控制规律的理论研究,并绘制出CO的φ-T图。结果发现,当量比强烈影响燃烧温度;CO快速反应的起始温度在1400～1500K;温度和浓度分层可以控制燃烧速率,但浓度分层可能使NO排放增加;EGR中N2主要影响燃烧最高温度,CO2对着火时刻延迟的作用最显著;MULINBUMP-HCCI燃烧产生的NOx、CO和HC排放主要来源于靠近缸壁余隙内,要提高其燃烧效率,减少燃油附壁是关键;低温燃烧要获得较高的燃烧效率和热效率,同时保持较低的NOx排放,混合速率应当随EGR率增加而适当增加。开展了基于调制喷油模式的预混燃烧实验研究,发现通过调制喷油参数可以实现对预混燃烧速率的控制,实现接近于零的NOx和碳烟排放。为进一步提高发动机的热效率,针对其控制参数多的特点,在完善三维发动机CFD程序的基础上,应用微种群遗传算法,建立了发动机控制参数优化软件平台。在该平台上优化多脉冲喷油参数,发现脉冲油量的分布影响燃料喷雾的蒸发、混合以及碰壁,优化多脉冲喷油过程可以形成浓度和温度分层适当的混合气,从而提高热效率。更多还原

【Abstract】 Homogeneous Charge Compression Ignition (HCCI) engines are being paid much attention and widely investigated due to their potential of high thermal efficiency and very low emissions of NOx and particulate matter (PM). However, for the diesel-fuel HCCI, there are still some critical problems that need to be solved. In this study, both the essential characteristics and control approaches of diesel-fuel HCCI combustion were investigated by experiment and simulation in a diesel engine with a MULINBUMP-HCCI combustion system.Since detailed chemical kinetic models can increase the computational burden so greatly that they are beyond the current ordinary PC capabilities, a new reduced chemical kinetic model of n-heptane named as“SKLE model”was developed in this paper. The model is based on two previous reduced kinetic models for alkane oxidation, from which some reactions have been eliminated and with enhanced treatment of the oxidization of CO and CH3O by using a combination of sensitivity analysis techniques, principal components analysis and steady-state approximation for intermediate species. In order to improve ignition timing predicted by the SKLE model, the key kinetic parameters of the model were optimized by using a micro-genetic algorithm coupled with the SENKIN program. The final model contains 40 species and 56 reactions, and it can predict CO, HC and NOx emissions of diesel HCCI engines. The simulations showed that the SKLE model generally agrees well with those of the detailed chemical kinetic model (544 species and 2446 reactions); the computational time of using the former is less 1/1,000 that of the latter. Thus, the highly efficient HCCI engine simulations using chemistry with multi-dimensional CFD are attainable by using the present model.Three categories of models have been established and applied for HCCI engine simulation in this paper. These models include a single-zone model with detailed chemistry, a multi-zone model with detailed chemistry and a three-dimensional (3D) CFD model with reduced chemistry. In addition, a quantitative“Φ(equivalence ratio)-T(temperature)”for CO formation has been created by performing the single-zone calculations using a detailed chemistry of n-heptane. The results show as follows: fuel/air equivalence ratio has significantly effect on burning temperature; CO oxidation rate become very fast when in-cyliner temperature reach 1400~1500K; charge stratification can control burn rate, but lead to an increase in NOx level; N2 strongly affects the peak combustion temperature, while CO2 has the highest impact on ignition delay among all gases in EGR; in the diesel engine with the MULINBUMP-HCCI combustion system, NOx, CO and HC emissions primarily arise from the crevices and liner, so it is the most important for the improvement of the thermal efficiency to avoid cylinder wall wetting; for low-temperature diesel combustion, as EGR rate increases, mixing rate must properly increase in order to keep high thermal efficiency and low NOx emission simultaneously.Diesel HCCI Combustion organized by the modulated injection mode was experimentally studied. It was found that this injection strategy can control combustion phasing and get very low NOx and smoke emissions. At the same time, a micro-genetic algorithm coupled with a modified 3D engine simulation code is utilized to optimize the injection parameters including the injection pressure, start-of-first-injection timing (SOI), fuel mass in each pulse injection and dwell time between consecutive pulse injections. The results showed that the pulse fuel distribution strongly influences the behavior of atomization, mixing, and wall wetting. The optimized injection parameters can provide the desired stratification of both fuel and in-cylinder temperature, resulting in high thermal efficiency.更多还原