Piston engines are a crucial component of hybrid power systems used in transportation and energy generation. As the demand for improved fuel economy and reduced emissions grows, it is essential to design models that optimize the power balance among the components of these hybrid systems. This optimization considers operational conditions such as temperature and load levels The research focuses on modern approaches to integrating piston engines in hybrid power plants for vehicles and energy generation. The primary focus is building a neuro-mathematical model that optimises energy flow distribution between an internal combustion engine (ICE) and an electric motor. To optimize fuel consumption, the model considers the total power requirement (Ptotal = 100 kW), the ICE efficiency (ηICE = 0.3), as well as temperature conditions (−10 ℃ to 30 ℃) and load (from 20% to 100%). Numerical modelling results show that fuel loss in the traditional approach can reach 50,000 conventional units at full power of the ICE. The proposed model presented a 15–20% reduction in costs depending on operating conditions, which indicates its effectiveness in real use. The model demonstrates its efficacy in lowering fuel consumption compared to traditional methods. Additionally, the study highlights the crucial role of artificial intelligence in optimizing control systems for hybrid power plants in real-time. The proposed model can potentially enhance the energy efficiency of vehicles and energy systems and outline directions for future research.