Dar Gözenekli Kanalların Akustik Ve Akış Davranışlarının Optimizasyonu - Dizel Partikül Filtresine Uygulanması

Abstract

Dar gözenekli kanallar otomobil motorlarından, ısı değiştiricilerine, gaz türbinlerine kadar birçok mühendislik uygulamasında yer alır. Bu nedenle bilimsel ve teknik alanda birçok çalışmaya konu olmuştur. Dar gözenekli kanal uygulamasına en iyi örneklerden biri de Dizel Partikül Filtreleridir (DPF). Hava kirliliğinde önemli bir payı olan dizel egzoz gazlarının kötü etkilerini azaltmak için otomobillerde kullanılan DPF’ler birçok ülkede standart haline gelmektedir. DPF’lerin esas amacının duman partiküllerinin zararlı emisyonunu azaltmak olmasına rağmen, aynı zamanda akustik emisyonu da etkilerler. Dizel motorların araç pazarındaki payının giderek artmasıyla, hava kirliliği ve gürültü gibi çevresel etkilerinin de olmasından dolayı DPF’ler inceleme konusu olmaya başlamıştır. Son 20 yıl içinde, DPF’lerin performansı birçok teorik ve deneysel çalışmaya konu olmuştur. Yapılan detaylı literatür araştırması sonucunda dar gözenekli kanallar ve bunun bir uygulaması olan DPF’ler ile ilgili olarak çok sayıda çalışma olduğu, ancak bunların akustiği ile ilgili çalışmaların sayısının oldukça sınırlı olduğu görülmüştür. Bu tezin amacı dar gözenekli kanalların akustik davranışını, mühendislik uygulamalarından biri olan DPF’ler özelinde yapılan detaylı çalışmalar ile incelemektir. Bu amaçla analitik, deneysel, parametrik ve sayısal çalışmalar gerçekleştirilmiştir. Elde edilen veriler kullanılarak DPF’nin akustiği üzerine bir optimizasyon çalışması yapılmıştır. DPF’nin akustik performansını belirlemek için ses iletim kaybı ve basınç kaybı parametreleri kullanılmıştır. Bu tezde ilk olarak, DPF’ler transfer matris metodu kullanılarak analitik olarak modellenmiştir. Analitik model, literatürde yer alan çalışmalardaki sonuçlar ile doğrulandıktan sonra, model üzerinde bazı parametrelerin DPF’lerin akustik davranışına etkileri ortaya konmaya çalışılmıştır. Benzer çalışma COMSOL programı kullanılarak oluşturulan sayısal model ile de gerçekleştirilmiştir. Tez kapsamında, ayrıca, DPF’lerin akış ve akustik ölçümleri gerçekleştirilmiştir. Bu deneysel çalışmalar ile DPF’lerin akustik performansını değerlendirmek üzere ses iletim kayıpları ve geri basınç kayıpları belirlenmiştir. Elde edilen sonuçlar, önceden oluşturulmuş olan analitik modeli ve sayısal modeli doğrulamak için kullanılmıştır. Tezde son olarak, DPF’lerin akustik performansını geliştirmek için optimizasyon çalışması gerçekleştirilmiştir. İlk defa bu çalışma ile DPF’lerin akustik optimizasyonu, literatürde yer alan birkaç parametrenin etkisinin belirlenerek yeni bir tasarım önerisi olmaktan öteye geçmiş ve hedeflenen sınırlar içerisinde DPF’lerin akustiğini en uygun hale getirecek olan tasarım belirlenmiştir. Bu yapılırken, DPF’ler için önemli olan diğer bir parametre, basınç kaybı parametresi de göz önüne alınarak, çok ölçütlü bir optimizasyon probleminin çözümü şeklinde de kurgulanmıştır. Optimizasyon için birbirinden farklı yöntemler kullanılmıştır. İlk olarak Ardışık Kuadratik Programlama ve Nelder-Mead yöntemleri kullanılmış, ardından genetik algoritmalardan Parçacık Sürü Optimizasyonu ve Hızlı ve Elitist Çok Ölçütlü Genetik Algoritma: NSGA-II yöntemleri kullanılmıştır.

Sound propogation in channels is a fundamental problem for acoustics. Sound wave propogation in channels can be divided into three main groups as “narrow”, “wide” and “very wide” depending on the diameter and frequency. In narrow channels, propogation is isothermal at very small diameters and low frequencies and can be expressed by using viscous forces. In wide channel, sound energy propogate equally within the section. Finally, in the very wide channels, sound energy becomes intense close to the walls. Narrow porous channels are in many engineering applications such as automobile engines, heat exchangers and gas turbines. For this reason, they have been the subject of many studies in the scientific and technical field. One of the best examples of the implementation of narrow porous channels is diesel particulate filters (DPF). Using a diesel particulate filter (DPF) on cars is becoming a standard in many countries to reduce the ill effects of diesel exhaust gases, which have a significant role in air pollution. Although the main purpose of DPFs is to reduce the harmful emission of soot particles, they also affect acoustic emissions. With the increasing share of diesel engines in the vehicle market, air pollution and environmental impacts such as noise have begun to be the subject of examination. Within the last 20 years, the performance of the DPF has been the subject of many theoretical and experimental studies. A detailed literature survey on the subject revealed that there are significant amount of studies addressing the narrow porous channels and the diesel particulate filter which is an application of them. However, studies dealing with acoustics of them are very limited. The main objective of this thesis is to examine the acoustic behaviour of the narrow porous channels by working detailly on the DPF which is an engineering application of it. For this purpose, analytical, experimental, parametric and numerical studies were carried out. An optimization study on the acoustics of DPF was conducted using the data obtained from these studies. Diesel engines are the most important power supply because of their strength and high efficiency. The diesel engine has the highest thermal efficiency of any standard internal or external combustion engine. Therefore they are commonly used power supplies. But the particulate emissions from combustion are a major problem worldwide. In recent years, due to environmental regulations and exhaust emission standards, diesel particulate filters (DPF) capturing emissions have been developed and some of them show quite impressive filtration efficiency up to 90%. Although the main purpose of a DPF is to reduce harmful emissions of soot particles, it also affects the acoustic emission. This has led researchers to study the acoustic performance of the DPF. A ‘particulate’ is a tiny fragment of matter, solid or liquid, suspended in the air. These particles are harmful for the environment and human. They contribute to cancer. Soot particles are called PM (particulate matter). Combustion quality is dependent on how the mixture of oxygen and fuel, is formed. Diesel engines produce significant amounts of particulates in their soot. Diesel particulates are still a problem and there is a special filter fitted to modern diesels to catch them – the DPF. A diesel particulate filter (DPF) is a device designed to remove diesel particulate matter or soot from the exhaust gas of a diesel engine. The first place that the gases coming from the engine exhaust manifold is the particulate filter. The wall-flow monolith DPF is the most common design. Particulate filter looks like a typical silencer and has many small parallel channels running in the axial direction and usually a cylindrical ceramic structure. The walls of the channels are porous, and adjacent channels are plugged at each end in order to cause gas to flow through these porous walls. A large part of unburned fuel institutions in the exhaust gas is accumulated in the closed cells of the DPF to prevent this gas, which is harmful to environment, to exhaust into the air. DPF is generally made from a material that can hold the institution occured after combustion in the diesel exhaust, such as porous metal, silicon carbide and cordierit, and its ceramic body is placed inside a metal enclosure. The materials that have high mechanical strength, the ability to withstand temperature changes, high strength to thermal loading, thermal conduction and abrasion, are prefered for production of the DPFs. Sound transmission loss and back pressure parameters were used to determine the acoustic performance of the DPF. To understand the mechanisms that affect flow and pressure drop in DPF has an important place in the DPF design. A good filter has a low pressure drop. Within the scope of the thesis, a mathematical model was built in order to examine the acoustical behavior of the filter parametrically by using the studies in the litearture and making neccessary assumptions. The accuracy of the model has been checked by using the MATLAB program. The DPF is modeled analytically by using the transfer matrix method. By using the mathematical model of the DPF, the acoustic effects of some parameters on the behavior of the DPF were determined. The examined parameters are the length of the DPF, the width and the wall thickness of the channels, temperetaure, the permeability of the channel walls, fluid density, dynamic viscosity, open area ratio in the DPF inlet and the outlet, end correction and finally the pressure drop factor. As a result of parametric studies, it is aimed to investigate the effects of the parameters on the transmission loss graph and to determine the number of the extremum points on the transmission loss graph obtained by using the values of the parameters in the desired region for the optimization studies. After the analytical model is validated with the results of the studies in the literature, the effects of some parameters on the DPF acoustic performance were tried to be determined by using this model. The numerical model generated using the program COMSOL were performed with similar work. Numerical analysis was conducted using COMSOL program. By using the model built in COMSOL, the effects of the changes in geometric parameters on transmission loss were investigated . Studies were carried out between 0-2000 Hz frequency range. The results of the numerical analysis are important in order to understand the success of the analytical model. The results of the analytical model were compared with the results obtained with the models that were built in COMSOL. Experimental studies were also carried out in order to verify the analytical and the numerical model and to examine the parameters. The DPF flow and acoustic measurements were carried out. By these experimental studies, sound transmission loss and back pressure were determined to evaluate the acoustic performance of DPF. The obtained results were used to validate the numerical model and the analytical model have previously been created. As a a result of the experimental studies, it can be concluded that the model, that was verified by both models in the literature and the experimental studies, can be used for the solution of the optimization problem. Pressure drop graphs of the DPFs obtained from the experimental data were used in order to get an idea of the order of the pressure drop parameter. After the analytical model is validated with the results of the studies in the literature, the effects of some parameters on the DPF acoustic performance were tried to be determined by using this model. The numerical model generated using the program COMSOL were performed with similar work. The optimization of the DPF was carried out based on 4 parameters, which were defined as design variables. These are channel width, channel wall thickness, DPF length and DPF diameter. Lower and upper boundaries for selected design variables were determined based on the data obtained from various manufacturers’ catalogs, taking into account the area in which the DPF can be placed in a vehicle. Finally in the thesis, the optimization was carried out to determine the acoustic performance of DPF. In literature, optimization was carried out by changing parameter values and examining the effects and proposing a design according to the results. With this study, for the first time in the literature, a new acoustically optimum DPF design was proposed by determining the effect of several parameters of the acoustic optimization of DPF at the same time within targeted limits of the design. Filter design optimization is primarily focused on the acoustic performance of the DPF, which is characterized by the transmission loss parameter. For this reason, it aimed to maximize TL values for the DPF at all frequencies. The optimization problem was also built as a multi-objective optimization by taking into account the back pressure parameter which is also important for DPFs’ acoustics. Pressure drop is the second optimization criteria considered in the present study. When a diesel particulate filter is added to an exhaust system, it causes a pressure drop. A pressure drop means that some type of back pressure occurs on the engine and forces the engine to work harder. Thus, the engine consumes more gas, i.e., it works inefficiently. Therefore, the pressure drop parameter is defined as an another objective function of the optimization problem. Different methods have been used for optimization. First, Sequential Quadratic Programming and Nelder-Mead methods were used, then genetic algorithms, Particle Swarm Optimization and Fast And Elitist Multiobjective Genetic Algorithm: NSGA-II method were used.