The vehicle was soaked in the lab for 12 h at an ambient temperature of 20.6 C before each individual test. The experiment was conducted according to the NEDC testing standard . Three
fueling strategies were investigated. Test 1 was the original engine test. In Test 2, the engine was fueled with pure hydrogen within the first 7 s from the onset of the cold start. Subsequently, the hydrogen injection duration was gradually decreased while the gasoline injection duration was gradually increased in the following 4 s by the HECU. During the transition time, the global excess air ratio of the hydrogen–gasoline mixtures was roughly kept at
1.00. The engine was finally fueled with the pure gasoline after 11 s from the starting. The engine control strategy in Test 3 was the same as that of Test 2 within the first 11 s from the onset of
the cold start. Differently, the engine in Test 3 was fueled with the hydrogen–gasoline blends at a hydrogen volume fraction in the total intake of about 3% after 11 s from the onset of the cold start when the engine speed was lower than 1100 rpm and the intake manifolds absolute pressure was smaller than 40 kPa. In Tests 2 and 3, the water electrolysis hydrogen generator was controlled to produce hydrogen only when the battery voltage was higher than 13.0 V and the hydrogen cylinder pressure was lower than 400 kPa.