Controlling Nox in Modified High Pressure Split Injection Single Cylinder Diesel Engine with EGR-A Mathematical Approach

Increase in number of vehicles globally possessing a serious threat to environmental pollution all over the world. In addition to these various reasons such as fuel specifications, fuel control, injection pressure, spray pattern, quantity and fuel timing, engine operating parameters like load, speed, turbocharging & super charger controlling techniques, amount of exhaust gas recirculation (EGR) etc. directly effects the emissions released from the engine. Light duty low capacity diesel generator sets and agricultural diesel engines are one of the major contributors for deteriorating the atmospheric air quality. Controlled fuel injection is one of the promising techniques in reducing the NO_x and smoke. EGR circulation has proven history in controlling the NO_x but however there is an increase in soot. Hence to control both NO_x and smoke a combined strategy of fuel injection control along with EGR circulation will be effective in reducing the engine emissions. Hence in the latest emission control strategy adopted in diesel engines split injection with EGR is used to regulate the emissions. This method is proven to be the most prominent technique in IC engines to regulate the NO_x and Smoke. Advanced combustion technologies use EGR technique for Gas and biodiesel fuels [1]. Research results show that the engine performance will be affected by the circulation of exhaust gas. It was reported that cooled EGR is more effective in reducing the NO_x emissions as compared to the hot EGR. However, trade off results reported with EGR that engine brake thermal efficiency is better with hot EGR as compared to the cooled EGR [2]. In take charge temperature is higher in hot EGR which leads to better brake thermal efficiency [3]. The effect of cooled EGR for different flow rates in turbocharged heavy duty diesel engines reduced the amount of bsfc and Soot with limited effect on reducing NO_x emissions [4]. Yu et al. [5] investigated the effect of EGR flow rates, equivalence ratio and flame temperature on engine brake thermal efficiency. They found that when the equivalence ratio is increased to near about 1.0 then amount of Carbon monoxide (CO) concentration in the exhaust was decreased. Studies carried by Feng et al. [6] on low combustion engine reported that there is no increase in peak pressure when the flow rates of EGR further increased. Simultaneously they noticed an increase in ignition delay, due to increase in the boiling point at the initial stage. Verschaeren et al. [7] conducted experiments on low, medium speed engines having two camshaft configurations with combined EGR and variable valve timings. They achieved higher amount of NO_x reduction and also a decrease in trade-off results between unburnt hydrocarbons (UHC) and NO_x emissions. In addition to this carbon monoxide and particulate matter (PM) also reduced in the entire engine loading conditions. They concluded this is mainly due to the higher volumetric efficiency. Homogeneous charge compression ignition (HCCI) combustion with emphasis on EGR is proven technique to reduce the NO_x emissions. Fang et al. [8] investigated the effect of pilot fuel injection which is injected much early compared to the main or secondary injection in HCCI heavy duty engine with EGR. Experiment results revealed that an increase in quantity of pilot injection decreased NO_x emissions with low flow rates of EGR. Engelmayer et al. [9] tested a 300kW power single cylinder diesel engine with high pressure fuel injection system. In his experiments he studied spray characteristics, combustion phenomenon and soot formation. He reported that there is an effective reduction in soot that can be lowered both at part and full load engine operation under high pressure fuel injection while circulating EGR through the inlet manifold. There is a unique demand for more torque, higher power output, decrease in fuel consumption and vehicle emissions in passenger cars. To meet this requirement common rail direct injection (CRDI) was modified to run on pulse or multiple injections either with separation or without it. Delphi Lucas has developed efficient fuel injection system (FIP) to meet the above requirements which suits mostly for diesel fuel. The fuel injection equipment developed by Delphi Lucas was capable of injecting the fuel at very high pressure at the same time it reduces the separation time between two pulse injection duration even close to zero micro seconds. Because of this the engine is capable of avoiding the longer premixed combustion and diffusion combustion was improved with reduction in NO_x emissions when EGR is also circulated. But in this technique higher amount of HC and CO is noticed when the fuel injection is advanced with EGR circulation. Lean combustion and decrease in combustion temperature is reported to be the reason for this higher amount of HC and CO emissions [10]. Lee et al. [11] achieved significant improvement of in power out of the engine when the pilot fuel is injected with multi hole injector consisting of 14 holes when the engine is running under idling conditions along with EGR. Simultaneous reduction in PM and NO_x was also reported. When the temperature of the exhaust gas is kept very low then greater mass of exhaust gas can be easily handled in less available space due to the decrease in the density of the exhaust gas. Hence the engine can take more amount of fresh air allowing higher amount of EGR percentage in the fresh air. Low temperature exhaust gas (EGR) was tested in Sports utility vehicle (SUV) engine by Mehrotra et al. [12]. Their research findings shown that it is possible to supply more heat by EGR technique due to the increase in capacity of intake air by the engine. In test results they found that around 16.4% of NO_x and 23.3% of PM was reduced for the same amount of fuel consumption when low temperature EGR is supplied to the engine. When EGR is cooled to very low temperature designated as ultra-cooled EGR it was observed that there is an increase in thermal efficiency with reduction in NO_x and PM but however there is an increase in HC and CO emission as per the experimental results of Brijesh et al. [13]. The above results reported when the fuel injection pressure is maintained at lower values. Cheng et al. [14] studied the effect of EGR on PCCI combustion under the advanced fuel injection timing. They reported that when fuel injection is advanced in PCCI combustion there is an increase in cylinder peak pressure, shorter ignition delay with simultaneous reduction in NO_x, Particulate matter (PM) and soot emissions. However, in this technique there is an increase in HC and CO emissions. High EGR flow rates and 50% mass share of pilot and main injection respectively in single cylinder direct injection diesel engine were investigated by the Sarangi et al. [15]. The results reveal that split injection along with EGR circulation has direct influence on the amount of EGR that needed to be recirculated. They concluded that in split injection along with EGR supply the engine requires less amount of EGR as compared to the conventional direct injection technique. NO_x can be effectively reduced without any compromise on the brake thermal efficiency. It has been observed that almost 17% amount of smoke was reduced with this split injection and EGR supply for diesel engines. Yang and Chung [16] in his experiments on diesel engine with high pressure split injection equipment which injects pressure up to 2000bar studied the effect of number of pulse injection in a cycle on engine emissions. The split injections are delivered by the electronically controlled fuel injection equipment which injects the fuel for 6 pulse injections in a given cycle. Their research findings shown that up to 1000bar fuel injection pressure there is high combustion gas temperatures are noted due to the intense and rapid combustion that is existing inside the combustion chamber. Yin et al. [17] noticed that higher amount of EGR is needed for advanced fuel injection without any split injection when the fuel is injected under very high pressure to reduce NO_x and soot emissions. There is a drastic reduction in soot released however NO_x decreased marginally with increase in bsfc. When fuel injection is advanced the quantity of pilot injection can be reduced to minimum quantity in the case of split injection at the same time the brake specific fuel consumption is also improved. The effect of number of pulse injections along with EGR was investigated using numerical simulations by Wang et al. [18]. They reported that a greater number of pulse injections without any EGR is proven to be more effective than a smaller number of pulse injections. The simulation results show that 5 pulse injections together with EGR is more effective that 3 pulse injections. In pulse injections the role of EGR is highly limited and much reduction in NO_x and Soot is not observed in pulse injections with EGR. Grey-Fuzzy Taguchi method is applied to the study the experimental results conducted by Roy et al. [19] on light duty modified single cylinder engine to run on CRDI and EGR is also supplied. They concluded that EGR is effective parameter and plays a significant role in reducing the engine emissions. Limited information regarding the effect of dual combination of ultra-cooled EGR and pulse injections in light duty low powered diesel engines under low and high engine operation conditions are not available. Thus, present research work mainly focusses to investigate the research gap mentioned above on a single cylinder low powered diesel engine of 3.5kW which is modified to work on CRDI for injecting the fuel under very high pressure with electronically controlled fuel injection system capable of delivering 3 pulse injections at 1000bar injection pressure at the engine speed of 1500rpm. The FIS is capable of controlling precisely both the injection quantity and timing by programmable electronic controller. In this present research work 2 pulse injections at different pressures ranging from 210bar to 1000bar in which 1^st pulse is pilot injection contains 10% of mass share injected at -54CAD and 2^nd pulse is main injection contains remaining 90% mass share of the fuel injected at -11CAD. Two ultra-cooled EGR flow rates were chosen consisting of 5% and 10% on volume basis of the intake air.

In this regression orthogonal numerical calculation xj (j=1, 2, 3, 4) is the level of each factor and X_1j and X_2j are upper and lower limits of the factors, then the zero level of the factor can be represented by X_0j=(X_1j+X_2j)/2. The Eq. (1) is used for determining the required number of numerical calculations that are necessary for the experimental programme. Then these results are compared with the 5 levels and 4 factors (i.e., 5⁴=625 times), only needs 17 times. The calculation period is reduced to 2.72% of the original one.

$n=m_0+m_c+m_r$               (1)

where, m₀, the center tests’ number, is 1; m_c, the times of two level test, is 2p-1; m_r, the asterisk test, is 2p; p, the number of factors, is 4; r, the asterisk arm length, is: r=(√n m_c-m_c)^1/2/√2 (i.e., r=1.353).

The test level is changed for number of steps that can be find by the lower and upper limits of the factors and asterisk arm length.

$\Delta_{\mathrm{j}}=\mathrm{X}_{1 \mathrm{j}}-\mathrm{X}_{2 \mathrm{j}} / 2 \mathrm{r}$               (2)

These factors should be kept in consistent dimension and they should not be deviate to a large degree in the orthogonal regression test. Thus the natural variables of each factor should be treated in the centralization. Suppose zj is code value obtained after the centralization of each factor level value, then Z_j (X_j+X_0j)/∆j. Here both the test results and simultaneously the analysis results are represented here. A regression equation which is unique and unified form was developed after processing the data from the orthogonal table given above:

$\mathrm{s}=\sum_{\mathrm{j}=1}^{\mathrm{n}} Y \mathrm{j}_{-1 / \mathrm{n}}\left(\sum_{\mathrm{j}=1}^{\mathrm{n}} \mathrm{Yj}\right)^2$               (3)

3.1 Single variable test

A single variable test is performed to identify the most influencing working engine operating parameter on NO_x emissions. Separate single variable test is conducted for each injection variable of 5%SI, 10%SI, 5%MPFI and 10%MPFI. In this analysis the most influence variable for reducing the engine emissions is NO_x and hence this single variable was chosen and using and an orthogonal regression analysis is conducted using 4 different factors of fuel injection strategies of SI, MPFI and EGR flow rates. Quadratic orthogonal regression array was computed using 26 arrays corresponding to 4 variables as seen in Table 3.

3.2 Quadratic orthogonal regression test

Quadratic regressive orthogonal test involving 4 factors is conducted to bring a relationship among adjustable working parameters for NO_x emissions in the engine [20, 21]. Engine tests are conducted to identify the most significant factor effecting the NO_x emissions in the modified single cylinder low-capacity engine which runs on CRDI system and these results are performed statistical analysis applying the quadratic orthogonal regressive technique. The quadratic relation coefficients are calculated using the equation in section 5.1. The weighting value of each quadratic coefficient was chosen from 0 to 1. In this present study the weighting value for the constraints assigned is+/-1.483. The results are analyzed for the engine operating parameters of ¼ Full load (0.875kW) and b) ¾ Full load (2.625kW) at different injection pressures of 200bar, 250bar, 300bar and 350bar respectively. Table 3 shows the arrangement of the test parameters and levels of the selected variables.

The results of quadratic regressive orthogonal test are shown in Table 4. After significance test on regression coefficients and regression formulas, the equations that relate the four working parameters to NO_x for the two operating conditions were formulated as follows [23, 24].

Table 4. Variables and computed test results obtained from orthogonal quadratic regression analysis

5.1 Optimization of working parameters

$\begin{aligned} & Y_2=251.972+39.298 x_1-76.258 x_2+52.989 x_3 -20.864 x_4+146.75 x_1 x_2+131 x_1 x_3-42.125 x_1 x_4 -79.875 x_2 x_3-2 x_2 x_4-68.75 x_3 x_4+425.082 x_1^2 -19.38 x_2^2+18.587 x_3^2-28.246 x_4^2 .\end{aligned}$

5.2 Constraint condition

$X_i-1.483 \leq 0$

where, i=1, 2, 3, 4.

With the help of statistical analytical software, the optimal combinations of the above 4 working parameters will be determined. The following results were obtained:

$\begin{aligned} & X_1=514.6, X_2=194.873, X_3=553.73, X_4=199 \\ & Y_{\min }=134.2790 \mathrm{ppm}\end{aligned}$

The present research work is conducted on the test facilities sanctioned by Department of science and technology, Govt of India under the project, Technology systems development scheme (TSD) vide order no. (DST/TSG/AF/2011/125). The corresponding author Dr.Veera Venkata Satyanarayana Murthy Yeditha, Associate professor & Placement officer, Associate Dean Research & Consultancy, Marine Engg Dept, Indian Maritime University, Central University, Govt of India, Chennai, 600119 Tamilnadu, India, wishes to  thank authorities of Indian Maritime University, for encouragement to publish the research findings.