Modelling and experimental study of automotive fuel droplets and sprays heating and evaporation

研究问题1.是否有可能使用修改后的商业计算流体动力学（CFD）代码，例如通过在ANSYS Fluent中实现用户定义函数（UDF），对真实的汽车燃油液滴加热和蒸发进行准确和CPU高效的建模？2.该模型是否可以推广到各种形式的多组分燃料，并自动选择合适的准组分进行准确有效的分析？3.我们可以使用现有的实验数据来验证所开发的模型吗？4.这些经过验证的模型是否能够为实际的工程应用提供准确的加热、蒸发和自燃模拟？目的和目标的调查：开发一个新的准确和CPU效率的模型，用于分析现实的汽车燃油液滴加热和蒸发，使用修改后的商业计算流体动力学（CFD）代码ANSYS Fluent和验证此模型使用现有的实验data.Objectives：1-实现先前开发的离散组件模型和多维准离散模型到ANSYS Fluent，通过利用用户定义的功能，并比较结果与先前开发的零维代码和实验数据预测在适当的情况下; 2.对多维准离散模型中准成分的选择进行优化分析;进一步开发该模型，使其能够根据预定标准自动选择这些准成分; 3.研究考虑液滴非球形度的离散组分模型和多维准离散模型的推广可行性;研究新模型与自燃模型相结合的可行性（自燃过程将使用Shell自燃模型来近似）;5-使用先前分析中未使用的柴油和汽油燃料液滴来研究模型的鲁棒性; 6-利用现有的或内部的实验数据验证模型的预测，该项目将侧重于开发汽车燃料液滴加热和蒸发的新模型，并根据现有的实验数据验证该模型。待开发的模型将基于首席主管与Al Qubeissi博士（前研究生，以学生身份获得布莱顿大学的支持）合作开发的先前工作（Sazhin，2014; Sazhin等人，2014 a，B; Al Qubeissi等人，2015 a，B;这些模型基于先前开发的传热和物质扩散方程的分析解考虑了液滴内的温度梯度和物质扩散。基于有效导热系数和有效扩散系数模型，考虑了液滴内部再循环的影响。对于燃料组分数量相对较少的汽车燃料（如生物柴油），重点是利用离散组分模型（Sazhin等人，2014 a; Al Qubeissi等人，2015年a）。在具有大量组分的汽车燃料如柴油和汽油燃料的情况下，焦点转移到准组分在液相中的扩散，如多维准离散模型（Sazhin等人，2014 b; Al Qubeissi等人，2015年b; 2017年）。后一种模型允许考虑柴油和汽油燃料成分的复杂性的解决方案，其中包括柴油燃料中多达98种不同成分的存在。

Research questions1. Is it possible to perform accurate and CPU efficient modelling of realistic automotive fuel droplet heating and evaporation using modified commercial Computational Fluid Dynamics (CFD) codes, for example via the implementation user defined functions (UDF) in ANSYS Fluent?2. Can the model be generalised for various forms of multi-component fuels and automatically select suitable quasi-components for accurate and efficient analysis? 3. Can we validate the developed models using available experimental data?4. Do these validated models then enable accurate simulation of heating, evaporation and auto-ignition for realistic engineering applications?Aims and objectives of the investigation:To develop a new accurate and CPU efficient model for the analysis of realistic automotive fuel droplet heating and evaporation using the modified commercial Computational Fluid Dynamics (CFD) code ANSYS Fluent and to validate this model using available experimental data.Objectives:1- To implement the previously developed Model of Discretised Components and the Multidimensional Quasi-discrete Model into ANSYS Fluent, via the utilisation of User Defined Functions, and comparing the results with those predicted by the previously developed zero-dimensional code and experimental data where appropriate;2- To perform optimisation analyses for the selection of quasi-components in the Multidimensional Quasi-discrete Model; further development of the model to enable it to select these quasi-components automatically based on pre-determined criteria;3- To investigate feasibility of generalising the Discrete Component Model and Multidimensional Quasi-discrete Model taking into account non-sphericity of droplets; and applicability for fuel spray modelling.4- To investigate the feasibility of combining the new model with an auto-ignition model (the auto-ignition process will be approximated using the Shell auto-ignition model);5- To investigate the robustness of the model using Diesel and gasoline fuel droplets not used in the previous analysis;6- To validate the predictions of the model using available or in-house experimental data.The project will focus on the development of a new model of automotive fuel droplet heating and evaporation and validation of this model against available experimental data. The model to be developed will be based on the previous works developed by the lead supervisor in collaboration with Dr Al Qubeissi (Former research student, supported in the form of studentship by the University of Brighton) (Sazhin, 2014; Sazhin et al., 2014a,b; Al Qubeissi et al., 2015a,b; 2017).These models took into account temperature gradient and species diffusion inside droplets based upon previously developed analytical solutions to the heat transfer and species diffusion equations. The effects of recirculation inside droplets was taken into account based on the Effective Thermal Conductivity and Effective Diffusivity models.In the case of automotive fuels with relatively small number of components for fuels such as biodiesel the focus was on the diffusion of the individual components in the liquid phase by utilizing the Discrete Component Model (Sazhin et al., 2014a; Al Qubeissi et al., 2015a). In the case automotive fuels with large number of components like that of Diesel and gasoline fuels, the focus shifted to the diffusion of the quasi-components in the liquid phase as described in the Multidimensional Quasi-discrete Model (Sazhin et al., 2014b; Al Qubeissi et al., 2015b; 2017). The latter model had allowed a solution to take into account the complexity of Diesel and gasoline fuel composition which includes the presence of up to 98 various components in Diesel fuel.