The invention provides a long oil pipeline leakage detection method and system based on AFPSO-K-means. The method comprises the following steps: initializing a particle swarm, and defining the initialspeed and position of each particle in the particle swarm; determining a fitness value of each particle; traversing the fitness value of each particle to obtain an optimal position corresponding to the fitness value of each individual particle and a global optimal position in the optimal fitness value of the particle swarm; iteratively updating the speed and the position of each particle according to the optimal position corresponding to the fitness value of the individual particle; and when the distance between the positions of all the particles and the global optimal position in the optimalfitness value of the particle swarm is smaller than a preset threshold value or the number of iterations reaches a preset number of times, outputting a current iteration updating result to determinethe actual position of pipeline leakage. The scheme is suitable for long oil pipeline fault detection, and can accurately obtain a global optimal value on the premise of avoiding a local optimal result.