Fig. 11 shows matlab evaluation of matlab programming actual compression of matlab programming interface of matlab programming head car of both matlab programming classic subway automobiles and matlab programming activepassive safety subway vehicles at impact velocities of 25 km/h, 27 km/h, 29 km/h, 31 km/h and 33 km/h. As matlab programming impact pace increased from 25 to 33 km/h, matlab programming actual compression of matlab programming interface of matlab programming head car of matlab programming classic trains during collision was 980 mm, 1026 mm, 1070 mm, 1117 mm and 1217 mm, exceeding matlab programming greatest value of 973 mm. However, matlab programming physical compression of matlab programming interface of matlab programming head car of matlab programming activepassive trains at impact velocities of 25 km/h, 27 km/h, 29 km/h, 31 km/h, 32 km/h and 33 km/h was 819 mm, 1056 mm, 1319 mm, 1591 mm, 1730 mm and 1841 mm, which shows that actual compression of matlab programming interface of matlab programming head car for matlab programming activepassive trains at impact velocities below 32 km/h didn’t exceed matlab programming greatest physical compression of 1773 mm. In other words, matlab programming safe impact velocity of an activepassive safety subway automobile conforming to matlab programming requirements of matlab programming EN15227 collision general was in a position to reach 32 km/h, far exceeding matlab programming safe impact pace of 25 km/h allowed by traditional trains, and representing a rise in matlab programming safe impact speed of 28%. The activepassive safety energy soaking up architecture was positive at recovering matlab programming crashworthiness of matlab programming subway train.