为提高基于量子进化算法(QEA)在宽带极紫外(EUV)多层膜设计中的求解效率和精度，本文利用宽带多层膜的光学性能评价函数的梯度信息改进QEA，建立具有明确进化方向的适用于宽带EUV多层膜设计的改进型量子进化算法(IQEA)。对比分析了基于IQEA和QEA的宽带Mo/Si多层膜的膜系设计过程和结果，结果表明，基于IQEA的多层膜膜系设计理论方法具有更优越的求解效率和精度; 同时，IQEA同样可以小种群规模进行多参数优化。基于IQEA的宽带Mo/Si多层膜的设计理论实现了包括入射角为0°~18°，反射率达50%的宽角度多层膜，以及反射光谱带宽为13~15 nm，反射率达25%的宽光谱多层膜的设计。基于IQEA的宽带高反射率EUV多层膜的理论膜系设计方法为复杂多层膜的理论设计提供了一种可供选择的高效膜系设计方法。

In order to further improve the efficiency and accuracy of quantum evolutionary algorithm (QEA) in the design of broadband Extreme Ultraviolet (EUV) multilayers, the gradient information of evaluation function was applied to improve QEA, and then the improved quantum evolutionary algorithm (IQEA) with defined evolution directions was adopted in the design of broadband multilayers. A comparison between the QEA and IQEA in the design of broadband Mo/Si multilayer was given, and the results show that the IQEA has better efficiencies and higher accuracies. Furthermore, the IQEA can be used to design the multilayer with a population which is smaller than the number of parameters of Mo/Si multilayer. In this research, two kinds of multilayers were designed based on the IQEA, one has a reflectivity of 50% in the incidence angle range from 0° to 18°, and the other allows a reflectivity of 25% in the wavelength range from 13 nm to 15 nm. The theoretical coating designs of broadband EUV multilayers based on IQEA can be an alternative and optimal method for the design of complex non-periodic multilayers.