基于Markov-ELM的独立混合微网分布式电源多目标容量优化配置

发布时间：2018-09-08 14:00

【摘要】：以微电网经济性、环保性和能源利用效率综合最优为目标,以系统可靠性为约束条件,构建混合微电网分布式电源(distributed generation,DG)优化配置模型。优化过程中需遍历全年数据进行时序分析,计算量大,运行效率低。针对该问题,采用模糊C-均值(fuzzy c-means clustering method,FCM)算法从风速、光照和负荷8760h的数据中提炼出典型的场景,用随机过程方法建立其对应的Markov模型。引入熵理论确定多目标函数各指标的权重系数,避免了人为定权重的主观性。采用种群移动、自适应变步长、基因变异以及按比例系数缩减搜索空间等方法来改进仿电磁学算法(electromagnetism-like mechanism,ELM)的寻优速度及全局搜索能力。最后将前述建立的Markov模型嵌入改进的ELM中进行优化求解。以广西涠洲岛为背景进行算例分析,验证模型和算法的有效性。结果表明,相较于采用年代数据计算的ELM,采用Markov-ELM的求解方法,在快速性和全局搜索能力等方面更具有优势。
[Abstract]:The optimal configuration model of distributed power generation (distributed generation,DG) for hybrid microgrid is constructed with the objective of comprehensive optimization of microgrid economy, environmental protection and energy efficiency, and the system reliability as the constraint condition. In the process of optimization, it is necessary to traverse the whole year data to carry on the time series analysis, the computation quantity is big, the running efficiency is low. To solve this problem, the fuzzy C-means (fuzzy c-means clustering method,FCM) algorithm is used to extract the typical scenes from the wind speed, illumination and load data of 8760h, and the corresponding Markov model is established by the stochastic process method. The entropy theory is introduced to determine the weight coefficients of each index of multi-objective function, which avoids the subjectivity of artificial determination of weights. The optimization speed and global search ability of electromagnetism-like mechanism,ELM are improved by population movement, adaptive variable step size, gene mutation and reducing search space according to proportional coefficient. Finally, the established Markov model is embedded in the improved ELM to optimize the solution. Taking Weizhou Island of Guangxi as the background, the validity of the model and algorithm is verified by an example. The results show that, compared with the ELM, based on chronological data, the Markov-ELM method has more advantages in terms of rapidity and global search ability.
【作者单位】：广西大学电气工程学院;
【分类号】：TM71

【相似文献】