储能科学与技术 ›› 2020, Vol. 9 ›› Issue (3): 918-926.doi: 10.19799/j.cnki.2095-4239.2019.0247

• 储能系统与工程 • 上一篇    下一篇

基于天牛须搜索遗传算法的风光柴储互补发电系统容量优化配置研究

李益民1(), 王关平1(), 马建立2, 杨浩1, 朱亮1, 闫红强1, 徐铮2, 朱冬琴1   

  1. 1. 甘肃农业大学机电工程学院,甘肃 兰州 730070
    2. 北京智源新能电气科技有限公司,北京 102628
  • 收稿日期:2019-11-04 修回日期:2020-01-12 出版日期:2020-05-05 发布日期:2020-05-11
  • 通讯作者: 王关平 E-mail:916953199@qq.com;wangguanping@gsau.edu.cn
  • 作者简介:李益民(1995—),男,硕士研究生,主要从事新能源发电系统控制研究,E-mail:916953199@qq.com
  • 基金资助:
    甘肃农业大学科技创新基金(GAU-QDFC-2019-10);国家自然科学基金(11762002);甘肃农业大学盛彤笙创新基金(GAU-STS-2018-25);甘肃农业大学科技创新基金(GAU-XKJS-2018-192)

Study on optimal capacity in the construction of wind-solar-diesel-battery hybrid power system based on bettle antennae search algorithm improved genetic algorithm

LI Yimin1(), WANG Guanping1(), MA Jianli2, YANG Hao1, ZHU Liang1, YAN Hongqiang1, XU Zheng2, ZHU Dongqin1   

  1. 1. Mechanical and Electrical Engineering College, Gansu Agricultural University, Lanzhou 730070, Gansu, China
    2. Beijing Intelligent Power & New Energy Electrical Limited Company, Beijing 102628, China
  • Received:2019-11-04 Revised:2020-01-12 Online:2020-05-05 Published:2020-05-11
  • Contact: Guanping WANG E-mail:916953199@qq.com;wangguanping@gsau.edu.cn

摘要:

依据设计员经验人为确定微电网建设中的容量配比不仅缺乏足够科学依据,易于造成投资浪费,而且还会对后续的经济调度、安全运行等产生负面影响。微电网建设最佳容量配比问题的本质是多目标寻优问题,各电源容量配比与既定指标之间存在复杂的非线性关系。为此,本文以某风光柴储互补发电系统设计为牵引,将容易找到全局最优解且寻优速度较快的天牛须搜索算法(bettle antennae search algorithm,BAS)引入到非线性规划性能较好的遗传算法(genetic algorithm,GA)中,在各电源出力模型和既定调度策略基础上,以兰州某点2018年逐时风速、逐时太阳光辐射强度、逐时环境温度为依据,以年供电可靠性最高作为首要目标、投资经济性最好为次要目标,获取最优容量配比方案。算例结果显示,BAS-GA能够给出最优容量配比且具有更快的收敛速度,每次都能得到相同结果,稳定可靠;其BAS-GA结果在负荷缺电率(loss of power supply probability,LPSP)为0.2%的情况下,比0时基本GA数据投资节约66%,也比2.3%时的基本GA结果效费比更高,其他基本GA的结果亦均无法与BAS-GA相比。因此,BAS-GA是微电网建设容量配比优化方案获取中避免基本GA陷入局部最优的有效措施之一。

关键词: 微电网, 风光柴储互补发电系统, GA, BAS, 最优容量配比, 供电可靠性, 投资经济性

Abstract:

Determination the capacity ratio in microgrids (MGs) not only lacks a sufficient scientific basis, but also incurs large investment waste, with detrimental impact on subsequent economic dispatching and safe operation of the MG. Therefore, MG construction faces a multi-objective optimization problem. Moreover, there is a complex nonlinear relationship between the capacity of each type of power source and the given index. To solve the capacity ratio optimization problem, this work incorporates the Bettle Antennae Search (BAS) algorithm into the genetic algorithm (GA). Based on the design of a wind-solar-diesel battery hybrid power system, the BAS algorithm improves the ease and speed of finding the global optimal solution and thereby improves the nonlinear planning performance. The input data include the hourly wind speed, hourly solar radiation intensity, and hourly environmental temperature at a certain point in Lanzhou City (Gansu Province, Northwest China), where the power output model has an established dispatching strategy. The primary and secondary goals of the optimal capacity ratio are to maximize the annual power-supply reliability and find the best investment economy, respectively. In all tested cases, the BAS-GA optimized the capacity ratio at faster convergence speed than others and consistently obtained the same result, confirming its stability and reliability. In the BAS-GA with an LPSP of 0.2%, the investment saving was 66%, whereas the basic GA with an LPSP of 0% yielded no investment saving. Moreover, the efficiency/cost ratio obtained by the basic GA (2.3%) exceeded the LPSP. The data of other basic GAs were not comparable with those of BAS-GA. The BAS algorithm effectively prevented the basic GA from falling into the local optima of the capacity- ratio optimization scheme in MG construction.

Key words: MG, WSDB-HPS, GA, BAS, optimal capacity ratio, PSR, IE

中图分类号: