Objective For the fuel-cell + supercapacitor hybrid power battery system adopted in virtual track tram, existing parameter matching methods have shortcomings such as low estimation accuracy and difficulty in achieving simultaneous multi-objective optimization. To accurately calculate vehicle operating conditions and thereby avoid capacity redundancy caused by extreme conditions, it is necessary to study a parameter matching method based on road-condition operating data.

Method Volume and mass models are established for the fuel-cell system, supercapacitor system, and hydrogen storage system, and key indicators such as vehicle driving range are introduced. Based on actual line speed data, a working condition calculation method using dual moving-average filtering is proposed to estimate line operating conditions of the vehicle. According to the estimated data, the NSGA-III (non-dominated sorting genetic algorithm III) in the multi-objective evolutionary algorithm framework is used to obtain the Pareto front of power system configuration schemes, thereby completing the parameter matching of the hybrid power system. Verification is conducted using the actual operation data of a virtual track tram.

Result & Conclusion  Calculation results based on actual operating data indicate that the dual moving-average filter structure not only avoids errors caused by direct calculation, but also maintains a high degree of conformity with the actual power curve, fully demonstrating the effectiveness of the filtering structure. The Pareto front results of multi-objective optimization show that the volume and mass of the power system may directly affect the vehicle operating mileage. The proposed parameter matching method can effectively improve the driving range of the vehicle while maintaining normal operation, achieving coordinated optimization of the power system mass, volume, and driving range.