Modeling Hybrid Electric Vehicles Using Fuzzy Interval Petri Nets

The power distribution control of Hybrid Electric Vehicles (HEVs) must overcome powertrain efficiency degradation and mechanical wear caused by abrupt mode switches. Traditional strategies often disregard real-world sensor noise and physical actuation delays. To address this, this paper proposes a hierarchical supervisory controller based on Fuzzy Interval Petri Nets (FIPNs). The controller integrates bipartite graph state-mapping with fuzzy logic adaptability, replacing rigid binary thresholds to evaluate continuous input parameters such as remaining fuel and battery state of charge (SOC) and calculate fuzzy transition credibility. Finite time intervals are embedded at the regulatory layer to emulate physical, mechanical latency, ensuring local Electronic Control Units (ECUs) receive executable switching commands. Tested within a MATLAB/Simulink R2023b (The MathWorks, Natick, MA, USA)dynamic time-domain environment under stringent transient conditions and evaluated over the 1800-s WLTC Class 3 driving cycle, the framework demonstrates significant performance gains. Spatio-temporal filtering eliminates high-frequency noise-induced chattering (0 spurious oscillations across 100 Monte Carlo runs) and reducing total macroscopic mode transitions over the 1800 s WLTC cycle from 34 down to 6 legitimate operational switches. Furthermore, compared to a baseline Thermostat Control Strategy (TCS), the FIPN supervisor yields an 18.9% reduction in total fuel consumption, offering a structurally transparent and mathematically rigorous solution for HEV energy management.