Improved Self-Adaptive Chaotic Genetic Algorithm for Hydrogeneration Scheduling
Improved Self-Adaptive Chaotic Genetic Algorithm for Hydrogeneration Scheduling
复制标题
DOI：
10.1061/(asce)0733-9496(2008)134:4(319
复制
发表时间：
2008-07
影响因子：
3.1
通讯作者：
Xiaohui Yuan;Yongchuan Zhang;Yanbin Yuan
中科院分区：
文献类型：
--
作者：
Xiaohui Yuan;Yongchuan Zhang;Yanbin Yuan
短期水电优化调度问题是一个复杂的带延时约束的非线性优化问题。针对标准遗传算法的不足，提出了一种新的实数编码自适应混沌遗传算法，该算法根据概率分布函数设计了一种新的交叉算子，并结合混沌动力学特性和人工神经网络理论设计了一种自适应混沌变异算子。约束条件可以用一种简单的直接比较罚函数法来处理，而不需要任何罚系数。以两个试验系统的短期发电计划为例，验证了该方法的可行性，并将试验结果与标准遗传算法的解质量和收敛特性进行了比较。仿真结果表明，该方法能够获得更高质量的解决方案。
The short-term optimal hydrogeneration planning is a complicated nonlinear constrained optimization problem with water delay time. To overcome the shortcomings of a standard genetic algorithm, this paper proposes a new real-value encoding self-adaptive chaotic genetic algorithm to solve this problem, which designs a new crossover operator in light of probability distribution function and a self-adaptive chaotic mutation operator combined chaotic dynamic character with artificial neural network theory. Constraints can be dealt with by using a simple direct comparison penalty function method without the need of any penalty coefficient. The feasibility of the proposed method is demonstrated for short-term generation scheduling of two test hydrosystems and the test results are compared with those obtained by the standard genetic algorithm in terms of solution quality and convergence characteristic. The simulation results show that the proposed method is capable of obtaining higher quality solutions.