中文摘要：

Tikhonov正则化方法是处理一类不适定问题的有效方法，最优正则参数的选取直接影响到最优解的产生．因此，如何选取最优参数极为重要。结合和声算法易收敛到全局最优和量子粒子群算法收敛快的优点．提出了一种和声搜索的量子粒子群算法，首先对基本测试函数进行测试，表明了算法的优越性，然后将算法应用于正则化参数的选取。结果表明，HS-OP-SO算法在选取正则参数时能有效的跳出局部最优解，与其它算法相比具有优更好地全局优化能力。

英文摘要：

The Tikhonov regularization method is an effective method for dealing with the ill posed problem. The selection of the optimal regularization parameter directly affects the optimal solution. In this paper, in view of the fact that the harmony search algorithm is easy to converge to the global optimum and the quantum particle swarm algorithm has fast convergence, we presented a new algorithm--quantum particle swarm optimization algorithm based on harmo- ny search. Firstly, we chose the basic test function, and the experimental results show the superiority of the algo- rithm. And then the new algorithm was applied to the selection of regularization parameters. Experimental results show that the proposed algorithm can effectively jump out of local optimal solution. And the new algorithm has more excellent global optimization performance than other intelligent optimization algorithms.