基于改进TOPSIS-PSO-SVM的配电网适应性评价研究

doi:10.16180/j.cnki.issn1007-7820.2022.06.009

摘要/Abstract

摘要：

我国可再生能源发展迅速,新能源发电已成为发电的主力之一。文中针对分布式能源并网后配电网的适应性问题,对配电网适应性内涵进行了新的定义,建立了可靠性、负载率、电流、电能质量、运行年限、新能源利用率6个一级指标的配电网适应性评价指标体系。通过主、客观赋权法得到组合权重,并将其与TOPSIS进行结合,确定了评价模型的期望输出值。文中提出了一种基于改进TOPSIS-PSO-SVM的配电网适应性评价模型,并以宁夏5个地区配电网进行实例分析。结果表明TOPSIS-PSO-SVM评价模型的评价相对误差区间为[0.94%,1.03%],相对误差绝对值平均数为0.885 4%,说明该评价模型在配电网适应性评价中的评价误差更小,评价精度更高。

关键词: 配电网, 分布式能源, 适应性评价, 层次分析法, 熵权法, 组合权重, 理想点法, 改进的粒子群优化支持向量机

Abstract:

With the rapid development of renewable energy, new energy generation has become one of the main forces of power generation in China. In view of the adaptability of distribution network after distributed energy is connected to the grid, this study proposed a new definition of the adaptability of distribution network. A distribution network adaptability evaluation index system with six first-level indexes, including reliability, load rate, current, power quality, service life and new energy utilization rate is established. Through the subjective and objective weighting method, the combined weight is obtained. Combined with TOPSIS, the expected output value of the evaluation model is determined. This study proposes a distribution network adaptability evaluation model based on improved TOPSIS-PSO-SVM. A distribution network adaptability evaluation model based on improved TOPSIS-PSO-SVM is proposed, and the distribution network in 5 regions of Ningxia is used for example analysis. The results show that the evaluation relative error interval of the TOPSIS-PSO-SVM evaluation model is [0.94%, 1.03%], and the average absolute value of the relative error is 0.885 4%, which indicates that the evaluation model has smaller evaluation error and higher evaluation precision in the adaptability evaluation of distribution network.

Key words: distribution network, distributed energy, adaptability evaluation, analytic hierarchy process, entropy method, combination weight, ideal point method, improved particle swarm optimization support vector machine

中图分类号:

TP13

黄元生,姜雨晴,王静. 基于改进TOPSIS-PSO-SVM的配电网适应性评价研究[J]. 电子科技, 2022, 35(6): 54-63.

HUANG Yuansheng,JIANG Yuqing,WANG Jing. Research on Adaptability Evaluation of Distribution Network Based on Improved TOPSIS-PSO-SVM[J]. Electronic Science and Technology, 2022, 35(6): 54-63.

图/表 12

表1

图1

图2

表2

表3

表4

表5

图3

图4

图5

图6

图7

参考文献 31

[1]	田鑫, 张健, 李雪亮, 等. 基于时序运行模拟的山东电网光伏发电消纳能力评估[J]. 电力系统保护与控制, 2020, 48(8):34-43.
	Tian Xin, Zhang Jian, Li Xueliang, et al. Evaluation photovoltaic accommodation ability for the Shandong power system based on sequential operation simulation[J]. Power System Protection and Control, 2020, 48(8):34-43.
[2]	甘磊. 考虑大型新能源发电基地接入的大电网规划方法研究[D]. 北京: 华北电力大学, 2017.
	Gan Lei. Research on planning method of bulk power network integrated with large-scale new energy power bases[D]. Beijing: North China Electric Power University, 2017.
[3]	国宗. 含新能源发电的配电网供电能力研究[D]. 上海: 上海电力学院, 2015.
	Guo Zong. Study on power supply capacity of distribution network containing distributed generation[D]. Shanghai: Shanghai University of Electric Power, 2015.
[4]	黄湛华. 基于供电可靠性的配电网规划实践[J]. 供用电, 2017, 34(3):42-46.
	Huang Zhanhua. Distribution network planning practice based on power supply reliability[J]. Distribution and Utilization, 2017, 34(3):42-46.
[5]	钱科军, 袁越, Zhou Chengke. 分布式发电对配电网可靠性的影响研究[J]. 电网技术, 2008, 32(11):74-78.
	Qian Kejun, Yuan Yue, Zhou Chengke. Study on impact of distributed generation on distribution system reliability[J]. Power System Technology, 2008, 32(11):74-78.
[6]	张嘉堃. 含分布式电源的配电网供电能力研究[D]. 上海: 上海电力学院, 2016.
	Zhang Jiakun. Study on power supply capacity of distribution network including distributed generation[D]. Shanghai: Shanghai University of Electric Power, 2016.
[7]	张伟. 含分布式电源的配电网供电能力评估[D]. 呼和浩特: 内蒙古工业大学, 2018.
	Zhang Wei. Evaluation of power supply capability for distribution network with distributed generation[D]. Hohhot: Inner Mongolia University of Technology, 2018.
[8]	李刚, 张巍, 奚培锋. 考虑经济安全性的有源配电网供电能力研究[J]. 电力科学与工程, 2020, 36(2):7-12.
	Li Gang, Zhang Wei, Xi Peifeng. Study on power supply capability of active distribution network considering economic security[J]. Electric Power Science and Engineering, 2020, 36(2):7-12.
[9]	王丽娜. 智能配电网规划经济性综合评价研究[J]. 电力学报, 2017, 32(5):414-420.
	Wang Lina. Study on comprehensive evaluation of smart distribution network economy[J]. Journal of Electric Power, 2017, 32(5):414-420.
[10]	姜楠, 王琦, 王恩荣, 等. 分布式光伏电源接入对配电网可靠性的影响研究[J]. 南京师范大学学报(工程技术版), 2016, 16(2):1-9.
	Jiang Nan, Wang Qi, Wang Enrong, et al. Research on influence of the distributed photovoltaic supply on distributed network reliability[J]. Journal of Nanjing Normal University(Engineering and Technology Edition), 2016, 16(2):1-9.
[11]	袁修广. 配电网可靠性与风险性评估方法的研究[D]. 长沙: 湖南大学, 2015.
	Yuan Xiuguang. Research on reliability and risk evaluation methods of distribution network[D]. Changsha: Hunan University, 2015.
[12]	赵学会, 梁骏, 王克峰. 供电安全标准在配电网网格化规划设计中的应用研究[J]. 能源研究与利用, 2019(1):36-39.
	Zhao Xuehui, Liang Jun, Wang Kefeng. Application research of power supply safety standard in grid planning and design of distribution network[J]. Energy Research and Utilization, 2019(1):36-39.
[13]	陶鹏, 张焰, 邱生. 基于GIS的配电网供电可靠性定量评估系统研究[J]. 电力系统保护与控制, 2006, 34(13):67-71.
	Tao Peng, Zhang Yan, Qiu Sheng. Research on quantitative assessment system for distribution network reliability based on GIS[J]. Power System Protection and Control, 2006, 34(13):67-71.
[14]	方欢欢. 智能配电网经济性评估方法研究[D]. 上海: 上海交通大学, 2012.
	Fang Huanhuan. Research on economic evaluation method of smart distribution network[D]. Shanghai: Shanghai Jiao Tong University, 2012.
[15]	陈强, 刘瑾, 杨海马, 等. 基于改进遗传算法的配电网无功优化研究[J]. 电子科技, 2019, 32(5):11-15.
	Chen Qiang, Liu Jin, Yang Haima, et al. Research on reactive power optimization of distribution network based on improved genetic algorithm[J]. Electronic Science and Technology, 2019, 32(5):11-15.
[16]	南钰, 宋瑞卿, 陈鹏, 等. 基于改进熵权-灰色关联法的配电网可靠性影响因素分析[J]. 电力系统保护与控制, 2019, 47(24):101-107.
	Nan Yu, Song Ruiqing, Chen Peng, et al. Study on the factors the reliability analysis in distribution network based on improved entropy weight gray correlation analysis algorithm[J]. Power System Protection and Control, 2019, 47(24):101-107.
[17]	何英静, 李帆, 沈舒仪, 等. 新能源及分布式电源接入浙江配电网适应性研究[J]. 浙江电力, 2018, 37(1):32-36.
	He Yingjing, Li Fan, Shen Shuyi, et al. Adaptability research on integration of new energy and distributed generation into Zhejiang distribution network[J]. Zhejiang Electric Power, 2018, 37(1):32-36.
[18]	王伟华. 新能源分布式电源大量接入吐鲁番配电网适应性研究[D]. 北京: 华北电力大学, 2018.
	Wang Weihua. Study on the adaptability of new energy and distributed power supply to Turpan distribution network[D]. Beijing: North China Electric Power University, 2018.
[19]	谭家茂, 黄少先. 基于模糊理论的电能质量综合评价方法研究[J]. 电力系统保护与控制, 2006, 34(3):55-59.
	Tan Jiamao, Huang Shaoxian. Research on synthetic evaluation method of power quality based on fuzzy theory[J]. Power System Protection and Control, 2006, 34(3):55-59.
[20]	邓雪, 李家铭, 曾浩健, 等. 层次分析法权重计算方法分析及其应用研究[J]. 数学的实践与认识, 2012, 42(7):93-100.
	Deng Xue, Li Jiaming, Zeng Haojian, et al. Research on computation methods of AHP wight vector and its applications[J]. Mathematics in Practice and Theory, 2012, 42(7):93-100.
[21]	郭金玉, 张忠彬, 孙庆云. 层次分析法的研究与应用[J]. 中国安全科学学报, 2008, 18(5):148-153.
	Guo Jinyu, Zhang Zhongbin, Sun Qingyun. Research and application of analytic hierarchy process[J]. China Safety Science Journal, 2008, 18(5):148-153
[22]	肖枝洪, 王一超. 关于“评测指标权重确定的结构熵权法”的注记[J]. 运筹与管理, 2020, 29(6):145-149.
	Xiao Zhihong, Wang Yichao. Notes on structure entropy weight method to confirm the weight of evaluating index[J]. Operations Research and Management Science, 2020, 29(6):145-149.
[23]	周志恒, 陈玲玲, 吕雅琪, 等. 基于熵权法对幸福感的模糊综合评价[J]. 电子科技, 2012, 25(11):18-20.
	Zhou Zhiheng, Chen Lingling, Lü Yaqi, et al. Fuzzy synthetic evaluation of the sense of happiness based on the entropy weight method[J]. Electronic Science and Technology, 2012, 25(11):18-20.
[24]	王艳, 周航, 丁松滨. 基于TOPSIS的客舱布局选项评估要素分析方法[J]. 电子科技, 2015, 28(9):23-26.
	Wang Yan, Zhou Hang, Ding Songbin. Assessment of cabin layout options based on TOPSIS[J]. Electronic Science and Technology, 2015, 28(9):23-26.
[25]	周亚. 多属性决策中的TOPSIS法研究[D]. 武汉: 武汉理工大学, 2009.
	Zhou Ya. The Study of TOPSIS method in multiple attribute decision making[D]. Wuhan: Wuhan University of Technology, 2009.
[26]	刘树林. 多属性决策理论方法与应用研究[D]. 北京: 北京航空航天大学, 1997.
	Liu Shulin. Multi-attribute decision making:theories methods, and their applications[D]. Beijing: Beijing University of Aeronautics and Astronautics, 1997.
[27]	房乐楠, 何腾鹏, 刘宇红. 一种改进型PSO算法在SVM参数寻优中的应用[J]. 电子科技, 2018, 31(6):17-19.
	Fang Lenan, He Tengpeng, Liu Yuhong. Application of an improved PSO algorithm in SVM parameter optimization[J]. Electronic Science and Technology, 2018, 31(6):17-19.
[28]	陈潇雅, 刘志坚, 刘晓欣, 等. 基于长期负荷预测和联络分析的配电网规划[J]. 广东电力, 2019, 32(3):59-65.
	Chen Xiaoya, Liu Zhijian, Liu Xiaoxin, et al. Distribution network planning based on long-term load forecasting and tie-line analysis[J]. Guangdong Electric Power, 2019, 32(3):59-65.
[29]	Wang J G, Zhang Z J, Zhang W X. Support vector machine based on double-population particle swarm optimization[J]. Journal of Convergence Information Technology, 2013, 8(8):898-905. doi: 10.4156/jcit.vol8.issue8.106
[30]	赵倩, 陈芳芳, 甘露. 基于改进粒子群算法优化支持向量机的风电功率预测[J]. 电气技术, 2020, 21(12):12-16.
	Zhao Qian, Chen Fangfang, Gan Lu. Wind power prediction based on support vector machine trained by improved particle swarm optimization[J]. Electrical Technology, 2020, 21(12):12-16.
[31]	黄洋, 鲁海燕, 许凯波, 等. 一种动态调整惯性权重的简化均值粒子群优化算法[J]. 小型微型计算机系统, 2018, 39(12):2590-2595.
	Huang Yang, Lu Haiyan, Xu Kaibo, et al. Simplified mean particle swarm optimization algorithm for dynamic adjustment of inertia weight[J]. Journal of Chinese Computer Systems, 2018, 39(12):2590-2595.

一级指标	二级指标	符号表示
可靠性指标	变压器可靠性	A₁
可靠性指标	线路可靠性	A₂
负载率指标	变压器满载率	B₁
	变压器容载比	B₂
	线路满载率	B₃
电流指标	线路最大电流	C₁
电流指标	线路额定电流	C₂
电能质量指标	电压偏差超标率	D₁
	谐波畸变超标率	D₂
	谐波电流超标率	D₃
	电压波动超标率	D₄
	电压不平衡度超标率	D₅
运行年限指标	变压器运行年限	E₁
运行年限指标	线路运行年限	E₂
新能源利用率	新能源渗透率	F₁
新能源利用率	新能源消纳率	F₂

序号	AHP权重	EM权重	组合权重
A₁	0.160 700	0.000 029	0.000 096
A₂	0.160 700	0.000 110	0.000 366
B₁	0.078 400	0.077 346	0.125 354
B₂	0.039 200	0.111 579	0.090 418
B₃	0.039 200	0.077 155	0.062 522
C₁	0.079 200	0.000 036	0.000 060
C₂	0.039 600	0.107 523	0.088 020
D₁	0.103 600	0.065 611	0.140 515
D₂	0.037 400	0.093 454	0.072 253
D₃	0.058 000	0.100 939	0.121 025
D₄	0.063 000	0.138 452	0.180 313
D₅	0.024 100	0.138 452	0.068 977
E₁	0.047 300	0.000 883	0.000 863
E₂	0.015 800	0.000 581	0.000 190
F₁	0.036 100	0.043 669	0.032 589
F₂	0.018 000	0.044 180	0.016 439

样本序号	期望输出值	模型训练输出	相对误差
1	0.435 6	0.429 9	-1.307%
2	0.515 4	0.521 1	1.114%
3	0.713 5	0.719 2	0.802%
4	0.342 9	0.348 5	1.608%
5	0.565 9	0.560 3	-0.985%
6	0.391 0	0.392 0	-0.255%
7	0.583 1	0.588 9	0.996%
8	0.745 4	0.741 4	-0.534%
9	0.199 0	0.193 5	-2.780%
10	0.610 8	0.599 1	-1.917%
11	0.457 4	0.462 7	1.154%
12	0.561 9	0.563 2	0.221%
13	0.519 3	0.539 9	3.975%
14	0.568 9	0.563 4	-0.971%
15	0.518 7	0.524 2	1.047%
16	0.397 4	0.402 7	1.336%
17	0.429 8	0.424 2	-1.293%
18	0.499 9	0.503 5	0.710%
19	0.649 7	0.655 2	0.848%
20	0.640 8	0.637 7	-0.479%
相对误差绝对值平均数			1.217%

样本序号	改进PSO-SVM
样本序号	期望输出值	评价测试值	相对误差
1	0.391 1	0.394 80	0.948 1%
2	0.430 9	0.428 60	-0.533 0%
3	0.506 7	0.503 02	-0.726 0%
4	0.705 1	0.696 80	-1.182 0%
5	0.642 9	0.636 20	-1.038 0%
相对误差绝对值平均数			0.885 4%

样本序号	期望值	PSO-SVM 评价测试值	PSO-SVM 相对误差	SVM 评价测试值	SVM相对误差
1	0.391 1	0.403 7	3.22%	0.413 2	5.66%
2	0.430 9	0.446 0	3.51%	0.455 5	5.71%
3	0.506 7	0.519 9	2.61%	0.486 2	-4.04%
4	0.705 1	0.680 2	-3.54%	0.689 7	-2.19%
5	0.642 9	0.620 3	-3.51%	0.601 0	-6.51%
相对误差绝对值平均数			3.28%		4.82%