The importance of capacitor bank as a hub device of power system is self-evident. The imbalance of load of the distribution network leads to the distribution problem of capacitor. Simulated annealing

tabu search and genetic algorithm are used to achieve optimization of the distribution

but the above methods treat the distribution network as an invariant system. However

the distribution system is time-varying system in nature and is probabilistic

which leads to the uncertainty of the optimal distribution results. In view of above problem

a kind of single-target probability method based on micro-genetic algorithm considering the randomness of the load is proposed in this paper

which provides more reasonable estimation for the optimal distribution of the capacitor bank. Finally

the method is tested in the IEEE 34-node imbalance distribution network. The experiments prove that the algorithm proposed in this paper has certain effectiveness and the calculation speed is significantly superior to other optimization algorithms.