1. Introduction

Nitrogen oxides (NOx) emitted from ships during operation are a primary cause of photochemical smog and acid rain, making them one of the most critical air-pollution concerns in the shipping sector. The International Maritime Organization (IMO) strictly regulates NOx emissions from marine main engines through MARPOL Annex VI Regulation 13, applying tiered limit values according to the engine construction year — Tier I (post-2000), Tier II (post-2011), and Tier III (post-2016, applicable within Emission Control Areas) (IMO, 2008; IMO, 2021). Most existing commercial vessels currently in operation must comply with Tier I or Tier II, and depending on the operational route, with Tier III as well, while continuously maintaining the validity of the Engine International Air Pollution Prevention (EIAPP) Certificate.

Meanwhile, with the strengthening of ship-efficiency regulations under the IMO greenhouse-gas strategy - including the Energy Efficiency Design Index (EEDI), the Energy Efficiency Existing Ship Index (EEXI), and the Carbon Intensity Indicator (CII) (Lee et al., 2023a; Lim et al., 2023) - Engine Power Limitation (EPL) measures, which substantively restrict the operational output of existing vessels, are becoming increasingly widespread (MAN Energy Solutions, 2022). In this process, as the engine is mainly operated in part-load regions below its design rated output, the efficiency of the turbocharger - which was designed for the rated condition - degrades under actual service conditions.

Among alternative technologies being studied for the reduction of greenhouse-gas emissions during vessel operation are onboard CO₂ capture (Lee et al., 2023b), nuclear propulsion (Jang, 2024), and air-lubrication systems (Park, 2020). As a practical solution that can be applied immediately to existing vessels, however, EPLO (Engine Part Load Optimization) has emerged. EPLO is a technology that lowers the rated output of the engine and replaces the turbocharger components with smaller specifications optimized for the actual service load range (typically 50-65 % MCR), thereby reducing the BSFC in that load range. The concept of EPLO is illustrated in Fig. 1.

Because the turbocharger replacement alters the engine's air-supply characteristics and combustion conditions, however, accurately identifying the resulting impact on NOx emissions and completing the re-certification procedure under the IMO NOx Technical Code 2008 is an essential task from the perspective of environmental regulation.

Studies on the relationship between NOx emissions and engine operating conditions have been actively conducted. Major research topics have included the effect of combustion temperature and residence time on NOx formation (Zeldovich mechanism), the correlation between fuel-injection timing and NOx, and the simultaneous influence of charging conditions on NOx and BSFC (ISO, 2017; Woodyard, 2009). However, empirical analyses of NOx emission characteristics and regulatory compliance through on-board measurement following the IMO E3 test cycle after EPLO application on actual ship main engines remain very limited.

In this study, a case of EPLO application to a 7S60MC Mk6 engine installed on an approximately 100,000-ton class vessel is analyzed in order to identify the effects of turbocharger reconfiguration on combustion characteristics and NOx emissions, and to quantitatively evaluate IMO Tier I compliance through on-board measurement following the E3 test cycle. The results of this study are expected to serve as technical grounds for ensuring NOx regulatory compliance in future expanded application of EPLO to similar engine families.

2. Experimental Equipment and Methods

2.1 Subject Vessel and Engine Overview

The subject vessel of this study is a 100,000-ton class ship, and the installed main engine is a Doosan Engine 7S60MC Mk6 with E3 test-cycle approval. The engine is a two-stroke uniflow-scavenged crosshead-type low-speed diesel engine; each cylinder is equipped with a pump-line-nozzle fuel-injection system and a Variable Injection Timing (VIT) device. The main specifications of the engine are summarized in Table 1.

2.2 EPLO Application

The main engine of the subject vessel has the characteristic that its actual service load is concentrated within the range of 50-65 % MCR (7,140-9,282 kW at 83.3-91.0 rpm, the original MCR being 14,280 kW at 105 rpm). An EPLO strategy was developed to improve BSFC within this range; the core measures are to limit the rated output and to optimize the turbocharger component specifications to the actual service output. The de-rating consists of restricting the original MCR of 14,280 kW (105 rpm) to 79 % of that level, i.e., a de-rated output of 11,229 kW (96.9 rpm) — hereafter referred to as the EPLO rated output — thereby reducing the Mean Effective Pressure (MEP) from 18.0 bar to 15.3 bar. All E3 test-cycle load percentages (25/50/75/100 %) are expressed relative to this EPLO rated output unless otherwise stated. The originally installed turbocharger was an Accelleron TPL73-B12. The component changes accompanying EPLO application are summarized in Table 2. In the table, CT, TF, and TA are the specification codes for the compressor wheel, the turbine frame, and the nozzle ring, respectively, and the trailing number indicates relative size. The compressor wheel was changed from CT75 to the smaller CT65, and the turbine from TF25 to TF15, in order to match the exhaust-gas flow and pressure of the de-rated engine - thereby shifting the turbocharger operating point into the high-efficiency region under the engine's part-load operation. In the turbocharger design, a minimum surging margin of at least 20 % was secured in consideration of overall system durability, and Pmax was controlled through readjustment of the VIT pivot point. Cam-related specifications such as the fuel cam lead angle and pump top lift were not changed.

2.3 NOx Measurement Method and Procedure

NOx measurement was conducted by applying the on-board simplified measurement strategy in accordance with the IMO NOx Technical Code 2008 (IMO, 2008) and ISO 8178-2 (ISO, 2018). The test cycle was the E3 cycle for fixed-pitch propeller (FPP) engines, and measurements were taken under steady-state conditions at four load points: 25 %, 50 %, 75 %, and 100 %. The exhaust-gas analyzer used was a Testo 350 Maritime V2, which simultaneously measures NOx (dry), CO, CO₂, and O₂. The specifications and an image of the Testo 350 Maritime V2 are presented in Table 3 and Fig. 2.

The pre- and post-measurement calibration and drift-check results confirmed that the analyzer was maintained within ±2 % of the permissible relative error, and on-board verification before and after measurement also confirmed compliance within tolerance (NOx span: 1,975 ppm before measurement → 1,974 ppm after measurement, relative drift -0.1 %).

The exhaust-gas sampling point was located in the exhaust gas pipe downstream of the turbocharger turbine outlet, where the measured exhaust gas temperature ranged from 183 to 194 ℃ depending on the load. The on-board measurement was carried out on 18-19 April 2024 during a voyage from Shanghai (China) to Masan (South Korea). Ambient conditions at each load point - barometric pressure 1,013-1,014 mbar, temperature 27.6-28.4 ℃, and relative humidity 43-47 % - were recorded simultaneously. The IMO NOx weighted average was calculated in accordance with the IMO NOx Technical Code by power-weighting the Specific NOx (g/kWh) of each load point using the corresponding engine power and the E3 weighting factor (25 %: 0.15; 50 %: 0.15; 75 %: 0.5; 100 %: 0.2) according to Eq. (1), and the analyzed composition of the actual fuel oil sample (C: 86.38 %, H: 12.39 %, S: 0.093 %, N: 0.056 %, O: 1.08 %) was applied. Wet-dry correction for ambient conditions was performed by applying the ISO humidity correction factor (k_H) given in Eq. (2), and the Specific NOx Emission (g/kWh) at each load point was calculated by Eq. (3).

where, q_NOx,i : ISO-corrected Specific NOx Emission at the i-th load point (g/kWh); P_i : Engine power at the i-th load point; WF_i : Weighting Factor for the i-th load point of the E3 test cycle

where, H_a : Absolute humidity (g/kg dry air)

where, C_NOx : NOx volumetric concentration (ppm, wet basis); Q_exh : Exhaust gas volumetric flow rate (m³/h); ρ_exh : Exhaust gas density (kg/m³); P_engine : Engine output power (kW)

3. Results and Discussion

3.1 Engine Performance and Combustion Characteristics under Turbocharger Reconfiguration

After EPLO application, the load-dependent engine performance parameters are presented in Table 4 and the combustion- and turbocharger-related parameters in Table 5. The scavenge-air pressure increased from 0.3 bar(g) at 25 % load to 1.9 bar(g) at 100 % load, and the turbocharger speed rose from 7,137 rpm (25 %) to 16,164 rpm (100 %). The auxiliary blower was in operation at 25 % load, while at 50 % load and above the operation transitioned to turbocharger-only without auxiliary blower assistance. As shown in Table 5 and Fig. 4, the average maximum combustion pressure (Pmax) was measured to range from 65.4 bar at 25 % load to 125.7 bar at 100 % load, remaining safely below the design limit (140 ± 3 bar) across all load points. The compression pressure (Pcomp) rose from 38.5 bar to 97.1 bar with increasing load, and the VIT index remained in the stable range of 8.1-8.2 across all loads. These results empirically confirm that the engine's combustion conditions are maintained within the normal range even after turbocharger downsizing.

The TC-inlet exhaust temperature increased from 273 ℃ at 25 % load to 354 ℃ at 100 % load, whereas the TC-outlet exhaust temperature decreased from 244 ℃ at 25 % load to 236 ℃ at 100 % load - i.e., it became lower at higher loads (Fig. 5). This suggests that turbocharger efficiency improves with load, allowing exhaust thermal energy to be more effectively converted into mechanical energy (compression work). BSFC (ISO-corrected), as shown in Fig. 3, increased from 181.9 g/kWh at 100 % load to 198.3 g/kWh at 25 % load, exhibiting the typical diesel-engine characteristic of higher BSFC at lower loads. It should be noted that, owing to the operational constraints of the in-service voyage (vessel draft, weather, and propeller/sea margin), the highest attainable steady-state load during the on-board measurement was 10,107 kW, corresponding to approximately 90 % of the EPLO rated output (11,229 kW). Strictly, the E3 100 % point of the IMO NOx Technical Code is to be measured at the rated output, so the on-board simplified measurement did not reach the full rated point. However, given the inverse-load dependency of Specific NOx Emission observed in this study (NOx per unit output decreasing with increasing load), the Specific NOx at the true rated point would be expected to be equal to or lower than the value measured at 90 % load (13.42 g/kWh). The use of this value in the E3 weighted average therefore yields a conservative (i.e., not under-estimated) result, and the conclusion of Tier I compliance is not weakened by this load shortfall. A full-load verification under controlled (e.g., test-bed) conditions remains the basis of the EIAPP certification, while the present on-board simplified measurement serves as in-service confirmation.

3.2 Load-Dependent NOx Emission Characteristics

The exhaust-gas composition and Specific NOx Emission measured at the four load points of the E3 test cycle are presented in Table 6 and Fig. 6. The NOx dry concentrations were measured as 844 ppm at 25 % load, 905 ppm at 50 % load, 893 ppm at 75 % load, and 894 ppm at 100 % load. Because the exhaust-gas volumetric flow itself increases with load, direct comparison of NOx emission by load on a ppm basis alone is limited; on the basis of Specific NOx Emission (g/kWh) per unit output, a clear inverse-load dependency was observed, with a minimum of 13.42 g/kWh at 100 % load and a maximum of 15.92 g/kWh at 50 % load (Fig. 6).

This inverse-load characteristic is attributed to the combined effect of increased BSFC and reduced scavenge-air pressure at low loads (25 % load 0.3 bar.g, 50 % load 0.8 bar.g) - which lowers the air-excess ratio - and consequently increases NOx emission per unit output. As confirmed in Table 5, the TC-inlet exhaust temperature at 50 % load (322 ℃) is significantly higher than at 25 % load (273 ℃), so the effect of increased thermal NOx formation associated with the rise in combustion temperature acts simultaneously.

These measurements are interpreted as reflecting the combustion trade-off induced by EPLO. The measured Pmax across loads ranges from 65.4 to 125.7 bar, and the pressure rise (Pmax-Pcomp) is in the range of approximately 27-36 bar (Table 5), indicating that combustion in the service load region is maintained at an intensity consistent with the EPLO design intent. This is a characteristic of EPLO reconfiguration in which the VIT position and injection timing are readjusted so that the peak cylinder pressure is utilized up to near the design limit (140 ± 3 bar) for BSFC improvement, and is known as a trade-off relationship accompanied by a rise in peak combustion temperature and a corresponding increase in thermal NOx formation (Heywood, 1988). The fact that the measured Specific NOx remains stably within the IMO Tier I limit (17.0 g/kWh) across all loads indicates that this trade-off is appropriately balanced.

The CO concentration generally decreased with increasing load - from 170.0 ppm at 25 % load to 83.3 ppm at 100 % load - indicating that combustion efficiency improves at higher loads. This load-dependent NOx behavior is a noteworthy characteristic from the perspective of NOx emission management for vessels with a high proportion of part-load operation.

3.3 IMO NOx Regulatory Compliance Assessment

The ISO-corrected NOx and the IMO NOx weighted average, calculated by processing the measurement data from each load point of the E3 test cycle through the IMO formulas, are presented in Table 7 and Fig. 7. The ISO-corrected NOx values - i.e., NOx emissions corrected to ISO reference conditions - were determined to be 13.3 g/kWh at 100 % load, 15.0 g/kWh at 75 % load, 15.8 g/kWh at 50 % load, and 15.5 g/kWh at 25 % load.

The final IMO NOx weighted average, obtained by applying the E3 weighting factor, is 14.6 g/kWh, securing a margin of 2.4 g/kWh (approximately 14.1 %) below the IMO Tier I limit of 17.0 g/kWh. A notable point is that, in addition to the weighted average, the ISO-corrected NOx at all four load points individually falls below the Tier I limit. The maximum load-wise emission is 15.8 g/kWh at 50 % load, which still leaves a margin of approximately 1.2 g/kWh (about 7 %) below the limit. The engine therefore satisfies the regulation across the entire operating region, irrespective of the E3 weighting scheme or the actual service load profile. In particular, the EPLO optimal service range (50-65 % of the original MCR, i.e., 7,140-9,282 kW, which corresponds to approximately 64-83 % of the EPLO rated output) lies around the 75 % E3 load point, where the ISO-corrected NOx (15.0 g/kWh) is comfortably below the limit. Moreover, even at the worst-case load point of the cycle (the 50 % E3 point, 15.8 g/kWh), the ISO-corrected NOx remains below the Tier I limit; the fact that compliance is maintained even under this most conservative condition supports regulatory compliance under actual operation. As also confirmed in Table 5 and Fig. 4, Pmax remains stably within the design limit across the entire load range after EPLO application.

4. Conclusions

In this study, a case of EPLO application to a 7S60MC Mk6 main engine installed on an approximately 100,000-ton class vessel was analyzed, and the effects of turbocharger reconfiguration on combustion characteristics and NOx emissions, together with IMO Tier I regulatory compliance, were empirically demonstrated through direct on-board measurement. The main conclusions are as follows.

(1) Verification of combustion characteristics and stability under turbocharger reconfiguration: Even after replacement of the compressor wheel (CT75 → CT65) and the turbine (TF25 → TF15), the maximum combustion pressure (Pmax) at all four load points was stably maintained within the design limit (140 ± 3 bar) - from 65.4 bar at 25 % load to 125.7 bar at 100 % load - and the VIT index also remained in the stable range of 8.1-8.2. This empirically confirms that the mechanical reliability of the engine is sufficiently secured even after turbocharger downsizing.

(2) Identification of inverse-load-dependent NOx emission characteristics after EPLO application: While NOx dry concentrations were distributed in the range of 844-905 ppm, Specific NOx Emission (g/kWh) showed an inverse-load dependency - higher at low loads and lower at high loads (13.42 g/kWh at 100 % load → 15.92 g/kWh at 50 % load). This is the combined result of increased BSFC at low loads, the reduction in air-excess ratio due to lower scavenge pressure, and changes in TC-inlet exhaust temperature; it highlights the importance of managing Specific NOx Emission in modern ship operations in which part-load operation accounts for a substantial proportion.

(3) Empirical verification of IMO Tier I compliance through on-board simplified measurement: The on-board measurement results following the E3 test cycle showed that the IMO NOx weighted average was 14.6 g/kWh, securing a margin of 14.1 % below the Tier I limit (17.0 g/kWh). Even at 75 % load, where the weighting factor is largest (WF = 0.5), the ISO-corrected NOx of 15.0 g/kWh was within the regulatory range, confirming that regulatory compliance is sufficiently maintained under actual operating conditions. These results provide on-vessel empirical evidence that EPLO technology, which involves changes to turbocharger specifications, is a practical technique capable of simultaneously achieving improved energy efficiency and NOx regulatory compliance, and they offer effective technical grounds for establishing NOx certification strategies in future expanded application of EPLO to similar marine engines.