基于Bloch球面坐标的改进量子遗传算法及其应用
An improved bloch spherical quantum genetic algorithm and its application

为解决量子遗传算法(QGA)用于连续多峰函数优化易陷入局部极值的问题, 提出一种基于Bloch球面坐标的改进量子遗传算法(GLBQGA):该算法通过引入新的全局-局部变异算子,在保证全局特性基础上加入局部搜索机制,使算法在搜索到全局最优近似解之后能通过局部邻域搜索收敛到全局最优精确解;算法还进一步优化量子转角取值方案,在保证搜索空间不变的同时提高搜索效率。在机车二系支承载荷均匀性分配优化调整及短时交通流多步预测中的应用表明,GLBQGA有效克服了QGA早熟收敛的问题,在不显著增加搜索时间的前提下提高了求解精度。
An improved Bloch Spherical Quantum Genetic Algorithm (GLBQGA) was proposed to overcome the shortcoming of the quantum genetic algorithm (QGA), i.e., local optimization, when it is used for the optimization of continuous functions with many extreme values. In order to make sure the algorithm can converge to the exact solution of optimal local neighborhood search after searching global optimum approximation, a new variation of global-local operator was introduced, and a local search mechanism was established based on globally attributes. The quantum angular value program was further optimized, while ensuring the search space and improving the efficiency of search. Calculative examples were made in optimization of locomotive secondary uniform load distribution and application of short-time traffic flow prediction, and the results show that GLBQGA can overcome the QGA premature convergence problems, and improve precision without increasing search time significantly