The invention provides a dynamic coordination hybrid electric vehicle energy management method based on working condition classification, and the method comprises the steps: employing an improved PSOalgorithm to optimize a K-means clustering analysis algorithm K-means to carry out the division of driving working conditions, and carrying out the adjustment of the dynamic oil consumption of a hybrid electric vehicle for sudden change working conditions. In the aspect of driving condition division, optimizing a clustering center of a K-means algorithm by the improved PSO algorithm, and then performing the condition division according to extracted characteristic parameters of typical conditions; and when a sudden change working condition occurs, entering a throttle valve dynamic coordinationprogram to perform reasonable torque distribution. An intelligent algorithm and a PID control technology are fused to solve the problems of fuel consumption increase, poor riding comfort and the likecaused by improper torque distribution and sudden accelerator change during sudden acceleration or deceleration at present; the power flow between the engine and the motor is reasonably controlled inreal time, the fuel economy is improved, pollutant emission is reduced, and the riding comfort is improved while the dynamic property is guaranteed.