Study On Diesel Particulate Filter And Its Regeneration Technique

Diesel particulate matter （PM） can cause serious environmental pollution and beharmful to human health, its emission control technology has been a focus. DieselParticulate Filters （DPFs） are the most effective aftertreantment device for removingPM from diesel engine exhaust. To comply with tightening emission regulations,DPFs are increasingly applied. In order to promote the practical and industrializationprocess of diesel particulate filter, the filtration, preesure drop and regenerationcharacteristics of a DPF were studied with theoretical analysis, numerical calculationand experimental research in this study.In order to achieve DPF structure design and optimization as well as DPFperformance prediction using computer simulation， DPF Filtration and pressure dropmodels were developed in Matlab computing platform, which could provide technicalsupport for the development of the DPF with high filtration efficiency and low flowresistance. Based on packed bed filtration theory, Brownian diffusion and interceptioncapture mechanisms, filtration model of the wall-flow particulate filter was built.Capture process of DPF was analyzed theoretical and pressure drop models of cleanand loaded filters were built. Experimental data were used to calibrate and validatingthe models. The models can predicte the pressure drop of a clean andparticulate-loaded DPF, and be used to analyse the effect of parameters of filterstructure and engine exhaust on the filtration and pressure drop performance. Thetrends of pressure drop and its components vs PM loading during cake filtrationregime were determined.For effective control of DPF regeneration process to ensure safe and completeregeneration, on the basis of summarizing the advantages and disadvantages ofvarious regeneration methods, a DPF active regeneration method with In-cylinderLate Post Injection （LPI） was pointed out, and DPF drop to idle （DTI） uncontrolledregeneration, and the methods of setting regeneration timing, and control strategy ofDPF active regeneration with LPI were studied. The study results show that:（1） LPI takes place far after the Top Dead Center （TDC）, large amont of the injected fuel does not combust completely in cylinder and produces Hydrocarbon （HC）emission, HC undergoes exothermic oxidation on the DOC to increase DPF inlettemperature for DPF active regeneratiom. Test results show that the effective inlettemperature of DOC is above250℃. The influence of LPI injection timing on engineperformance and DOC temperature rise characteristics is little. At the same injectintiming, with the increase of LPI amount, engine torque keep constant while fuelconsumption increases, DOC inlet temperature does not change dramatically, HCemissions at DOC inlet exhaust increase significantly, DPF inlet temperature increase.DOC temperature rise characteristics are obvious under low speed conditions. LPIcould achive safe and complete DPF regeneration with effective regeneration time ofabout200s through the experimental validation. During the regeneration process, DPFpressure drop increased firstly and then decline sharply until it reaches relativelystable. The temperature of each point on the same axis line with same radial positionsuccessively higher along the flow direction of exhaust gas, DPF internal peaktemperature is near DPF outlet at the central axis. The radial temperature gradient islarger than the axial temperature gradient, its peak value appears early in theregeneration, and is near the filter skin.（2） For DTI uncontrolled regeneration, the engine goes to idle during the normalDPF regeneration process. The study results show that: during DTI uncontrolledregeneration, DPF internal temperature distribution regulation is the same as DPFnormal regeneration. At idle condtions, low exhaust flowrate ang high oxygenconcentration cause dramatical increase in peak temperature and peak temperaturegradient. Through EGR control, intake throttling and rising idle speed could reducethe maximum peak temperature to avoid DPF damage.（3） The methods of setting regeneration timing involved according back pressure,according driving time, according soot emissions, according mathematical models, aswell as according soot loading were analyzed respectively. The method of setingregeneration timing according soot loading obtained from pressure drop washighlighted, the SML determination of the filter through DTI was discussed. Based onjudging regeneration timing according soot loading and DPF active regeneration withLPI, DPF active regeneration control strategy was proposed, the method fordetermining the optimal value of LPI amount was analyzed. For better particle emsission control, particle number emission and particle sizedistribution of different fuels with different physical and chemical characteristics.weremearsured, and DPF number efficiency （NE） and fractional efficiency （FE） wereanalyzed. The study results show that:（1） For diesel, BTL, GTL and FBC, exhaust particle number size distributionsare unimodal, the sizes of the particles are typically less than200nm during all thetest modes. Compared with engine speed, the engine load greatly influences particlesize distributions. As the engine load increases, the particle size distributions fordiesel, BTL and FBC turn to nucleating. The peak diameters become smaller, whereasfor GTL, this value becomes larger and the emissions of the nucleation mode particledecrease.（2） BTL has lower calorific value and higher oxygen content, it has differentcombustion and emission characteristics with diesel. When fueled with BTL, theignition timing advances, and theignition delay period are shortened, the engine has ahigher BSFC. The CO₂, HC and CO emissions decrease as NOx increases,particularly at higher load conditions. BTL has the highest total particle andnanoparticle number concentrations. How to reduce its nanoparticle emissions is achallenge.（3） GTL has higher cetane number and less aromatic hydrocarbons than the otherfuels. GTL results in the advancement of the beginning of ignition, a shorter ignitiondelay period, a shorter pre-mixed combustion period and complete combustion. Whenfueled with GTL, the CO₂, HC, CO and NOx emissions all decrease, particle numberemissions of nucleation mode and accumulation mode decrease.（4） For FBC, the change of particle size distribution、 the percentage ofnanoparticle and particle number concentration with engine load or engine speed aresimilar with diesel. The use of Fe-based additive could premote combustion andreduces the emissions of larger size particle. Due to the generation of metal oxide incombustion chamber, FBC has highest total particle number and nanoparticleemissions, the percentage of nanoparticle is above90%.（5） Diesel engine fueled with fuels of different physical and chemicalcharacteristics will produce different exhaust gas parameters and particle sizedistributions, thus influencing the filtration performance of the DPF. The change of NE and FE of the DPF with engine load and speed is not regular. For BTL, GTL andFBC, the NE of the filter reduces. Under most conditions, DPF number efficiency fordiesel is above90%, which is far higher than that for BTL, GTL and FBC. Over themain size range of exhaust, DPF fractional efficiency for diesel is sufficiently high,and behaves as predicted by filtration theory. The Greenfield Gap locates in theaccumulation size range and near150nm. For BTL, GTL and FBC, the FE decreasesduring the main size range compared with that for diesel, the filter has an extraGreenfield Gap located in the nanometer size range, which is disagree with filtrationtheory. Overall, when the diesel engine is fueled with alternative fuels, to avoidjeopardizing the potential emission benefits of alternative fuels and to maximize theDPF system efficiency, it is necessary to improve the filtration performance of thefilter or to choose the appropriate filters for the application according to the particleemission characteristics of alternative fuels.