A reliability assessment strategy named Structural System-oriented Active learning reliability method combining Kriging and Monte Carlo Simulation (SSo-AK-MCS) was proposed, the number of calls to the limit-state function of SSo-AK-MCS is 0.000121% of Monte Carlo Simulation (MCS), 4.3478% of Subset Simulation (SS), 11.2768% of Importance Simulation (IS). In the meantime, calculation precision of SSo-AK-MCS is 100.2930 times of SS, 11.5431 times of IS. The cost of achieving that outstanding performance is the uplift of the degree of intelligence w.r.t. reliability assessment strategy. At this moment, the reliability assessment strategy is no longer static, but dynamic, which has been equipped with the adaptive active learning strategy. Under the circumstances that uncertainty must be taken into account, the structural reliability assessment w.r.t. complex heterogeneous structure which is extremely sensitive to reliability, such as the duct of two augmented turbofan engines named RD-93 of fighter J-31.