In order to meet the smart distribution grid's demand for communicating, a heterogeneous wireless network selection model is built. This model has 3 goals, real-time services' blocking rate, non-real-time services' mean transmission time, and network load. And the model is simulated by using the quantum clonal immune multi-objective optimization algorithm. Compared to the MLB algorithm and RBF-FNN algorithm, the simulating results show that the proposed algorithm has better performance in real-time service blocking rate, mean transmission time of non-real-time services and network load balancing.