doi: 10.52899/24141437_2026_02_227
UDK: 629.5.03-843.6:621.355

Algorithms for load sharing in hybrid propulsion systems combining liquefied natural gas and electrochemical energy storage

Логинова А. А., Логинова Е. А.
Article language:
Citation Link: Loginova E.A., Loginova A.A. Algorithms for Load Sharing in Hybrid Propulsion Systems Combining Liquefied Natural Gas and Electrochemical Energy Storage. Transactions of the Saint Petersburg State Marine Technical University. 2026;5(2):227–234. DOI: 10.52899/24141437_2026_02_227 EDN: DCZTON

Annotation

BACKGROUND: With the tightening of International Maritime Organization requirements aimed at reducing the carbon footprint and achieving net-zero emissions by 2050, liquefied natural gas (LNG) is considered the primary transition fuel for the maritime power sector. However, the environmental advantages of LNG-powered systems are offset by the methane slip effect, which peaks at low and variable load conditions. AIM: This study aimed to develop and validate load distribution algorithms for a hybrid propulsion system consisting of a gasfired piston engine (powered by liquefied natural gas) and a lithium-ion battery. METHODS: The methodology was based on a comprehensive approach combining mathematical modeling, theoretical optimization, adaptive control theory, and numerical simulation. Mathematical modeling involves constructing a statespace representation that integrates submodels of the gas engine, the lithium-ion battery, and the load profile. Theoretical optimization relies on the Equivalent Consumption Minimization Strategy (ECMS) as the core algorithmic framework. Adaptive control is implemented through a dual-loop hierarchical regulator featuring an adaptive equivalence factor. Forecasting techniques are incorporated by introducing a correction method for the equivalence factor based on spectral load prediction, utilizing a moving horizon prediction window and autoregression. Finally, numerical modeling and verification are carried out in the MATLAB/Simulink environment, where the developed algorithms are validated through comparative analysis against alternative strategies, namely the non-hybrid configuration and the conventional peak-shaving approach. RESULTS: The simulations performed confirm the high efficiency of the proposed approach. Firstly, significant fuel savings are achieved: integral LNG consumption is reduced by 7.2% compared to a non-hybrid configuration and by 2.1% relative to the traditional peak-shaving strategy. Even more significant is the environmental benefit: methane emissions are cut by 14.6% compared to the peak-shaving strategy. This is achieved by maintaining the gas engine within the 60%–80% load range of its nominal value, corresponding to the minimum emission zone. Furthermore, the system’s dynamic performance is improved: the frequency of gas engine starts is reduced by 83%, substantially extending its overhaul life. Accurate energy balance maintenance is also ensured: the battery state of charge returns to its initial level by the end of the cycle (69.8% against a 70% target), guaranteeing repeatable operating modes from voyage to voyage. CONCLUSION: The proposed adaptive ECMS algorithm with spectral correction enables not only fuel savings but also a significant reduction in the greenhouse effect by suppressing methane slip. The developed methodology is recommended for implementation in the control systems of next-generation marine power complexes. Keywords: гибридная пропульсивная установка, сжиженный природный газ (СПГ), литий-ионная аккумуляторная батарея (ЛИАБ), алгоритм распределения нагрузки, метановое ускользание, эквивалентная минимизация расхода топлива (ECMS), судовая энергетика, оптимизация энергопотоков

Bibliography

BACKGROUND: With the tightening of International Maritime Organization requirements aimed at reducing the carbon footprint and achieving net-zero emissions by 2050, liquefied natural gas (LNG) is considered the primary transition fuel for the maritime power sector. However, the environmental advantages of LNG-powered systems are offset by the methane slip effect, which peaks at low and variable load conditions. AIM: This study aimed to develop and validate load distribution algorithms for a hybrid propulsion system consisting of a gas-fired piston engine (powered by liquefied natural gas) and a lithium-ion battery. METHODS: The methodology was based on a comprehensive approach combining mathematical modeling, theoretical optimization, adaptive control theory, and numerical simulation. Mathematical modeling involves constructing a state-space representation that integrates submodels of the gas engine, the lithium-ion battery, and the load profile. Theoretical optimization relies on the Equivalent Consumption Minimization Strategy (ECMS) as the core algorithmic framework. Adaptive control is implemented through a dual-loop hierarchical regulator featuring an adaptive equivalence factor. Forecasting techniques are incorporated by introducing a correction method for the equivalence factor based on spectral load prediction, utilizing a moving horizon prediction window and autoregression. Finally, numerical modeling and verification are carried out in the MATLAB/Simulink environment, where the developed algorithms are validated through comparative analysis against alternative strategies, namely the non-hybrid configuration and the conventional peak-shaving approach. RESULTS: The simulations performed confirm the high efficiency of the proposed approach. Firstly, significant fuel savings are achieved: integral LNG consumption is reduced by 7.2% compared to a non-hybrid configuration and by 2.1% relative to the traditional peak-shaving strategy. Even more significant is the environmental benefit: methane emissions are cut by 14.6% compared to the peak-shaving strategy. This is achieved by maintaining the gas engine within the 60%–80% load range of its nominal value, corresponding to the minimum emission zone. Furthermore, the system’s dynamic performance is improved: the frequency of gas engine starts is reduced by 83%, substantially extending its overhaul life. Accurate energy balance maintenance is also ensured: the battery state of charge returns to its initial level by the end of the cycle (69.8% against a 70% target), guaranteeing repeatable operating modes from voyage to voyage. CONCLUSION: The proposed adaptive ECMS algorithm with spectral correction enables not only fuel savings but also a significant reduction in the greenhouse effect by suppressing methane slip. The developed methodology is recommended for implementation in the control systems of next-generation marine power complexes.