Multiple swarms immune clonal quantum-behaved particle swarm optimization algorithm and the wavelet in the application of forecasting foundation settlement

To solve the problem of the quantum-behaved particle swarm optimization algorithm (QPSO) easy falling into the local optima, we proposed the multiple swarm immune clonal quantum-behaved particle swarm optimization algorithm in which the swarm was divided into two subgroups dynamically according to the particle´s fitness. In the better fitness subgroup, we carried immune clonal algorithm with Gaussian mutation to do local searching, and in the other subgroup we carried immune clonal algorithm Cauchy mutation to do global searching. And we also made a compare with standard quantum-behaved particle swarm optimization with wavelet de-noise. From these experiment results we can see that this improved method had a better ability of searching global and local optimum and a high forecasting precision.