基于不确定参数的变电站碳储量预估方法

doi:10.11930/j.issn.1004-9649.202311078

摘要/Abstract

摘要：

设备资产运维精益管理系统（power production management system，PMS）SF₆气体量数据不全且误差较大，无法为电网企业核算碳储量以及实现待建变电站碳规划提供基础数据。针对上述情况，研究了计及母线和断路器的变电站碳储量核算方法，并结合宁夏电网现场实测数据，通过MIC法筛选神经网络输入参数，构建了6输入参数的GA-BP、PSO-BP、HPO-BP神经网络模型，结果表明HPO-BP神经网络模型的评估指标及预估结果相对误差（6.28%）均优于其余2种神经网络模型，可以准确核算断路器SF₆气体量。针对参数不确定情况，根据PCCs法分析不同参数之间的线性关系，构建了3输入参数的HPO-BP神经网络模型，预估结果相对误差为9.72%。通过遍历输出方式，在参数不确定情况下输出多组断路器SF₆气体量预估数据，利用求和累积方法获取变电站总SF₆气体量，并量化为变电站碳储量，从而为电网企业实现“双碳”目标提供数据支撑。

关键词: SF₆气体量, 碳储量, 神经网络模型, PSO, 不确定参数

Abstract:

The SF₆ gas volume data from the PMS system is incomplete and contains significant errors, rendering it inadequate for power grid enterprises to calculate carbon reserves and formulate carbon planning for future substations. To address this problem, this paper studies the carbon storage accounting method for substations, taking into account buses and circuit breakers. The input parameters of the neural network are selected using the MIC method based on the field measured data in Ningxia grid, and three neural network models (GA-BP, PSO-BP, and HPO-BP) with six input parameters are developed. The verification results indicate that the HPO-BP neural network model outperforms the other two models in evaluation index and prediction results with a relative error of 6.28%, and it can accurately calculate the SF₆ gas volume of circuit breakers. Additionally, considering the parameter uncertainties, the linear relationship between different parameters is analyzed using the PCCs method, and a three-input-parameter HPO-BP neural network model is constructed, yielding a relative error of 7.69% for the prediction results. Concurrently, the ergodic output mode is examined to generate estimated SF₆ gas volume data for multiple groups of circuit breakers under uncertain parameter conditions. The total SF₆ gas volume of the substation is determined using the cumulative sum method, consequently quantifying the total carbon storage of the substation, thereby offering data support for power grid enterprises to accomplish the objective of "double carbon".

Key words: SF₆ gas volume, carbon storage, neural network model, PSO, uncertain parameters

陈巳阳, 韩利, 方济中, 丁五行, 成诚, 李文, 张源, 钱勇. 基于不确定参数的变电站碳储量预估方法[J]. 中国电力, 2024, 57(4): 200-210.

Siyang CHEN, Li HAN, Jizhong FANG, Wuhang DING, Cheng CHENG, Wen LI, Yuan ZHANG, Yong QIAN. Prediction Method for Carbon Storage in Substation Based on Uncertain Parameters[J]. Electric Power, 2024, 57(4): 200-210.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202311078

https://www.electricpower.com.cn/CN/Y2024/V57/I4/200

图/表 12

图 1 实测数据与PMS系统设备台账信息差值

Fig.1 Difference between measured data and PMS system equipment ledger information

表 1 不同参数MIC值

Table 1 MIC values of different parameters

参数		最大信息数
设备名称		0.95
设备型号		0.91
电压等级		0.79
额定电压		0.77
额定电流		0.76
SF₆气体额定压力		0.64
操作机构型式		0.57
结构型式		0.53
生产厂家		0.41
SF₆气体闭锁压力		0.27
SF₆气体报警压力		0.21
出厂日期		0.14

图 2 BP神经网络拓扑结构

Fig.2 BP neural network topology

表 2 不同隐藏层的误差值

Table 2 Error values of different hidden layers

H值		3		4		6		7		8		9		10		11		12
平均相对误差		30.2		29.1		27.2		25.7		24.1		23.6		26.5		27.2		31.1

图 3 GA-BP、PSO-BP、HPO-BP神经网络训练样本测试集结果对比

Fig.3 Comparison of test set results of GA-BP, PSO-BP and HPO-BP neural network training samples

表 3 BP、GA-BP、PSO-BP和HPO-BP神经网络评价指标结果

Table 3 Evaluation index results of BP, GA-BP, PSO-BP and HPO-BP neural networks

模型类型	δ_RMSE/kg	${K}_{{\text{R}}^{2}} $	δ_MAPE
GA-BP	171.460	0.954	0.163
PSO-BP	187.801	0.979	0.112
HPO-BP	80.091	0.991	0.061

图 4 GA-BP、PSO-BP、HPO-BP神经网络模型对同类型SF6气体预测结果对比

Fig.4 Comparison of prediction results of the same type of SF6 gas by GA-BP, PSO-BP and HPO-BP neural network models.

图 5 参数线性分析热值

Fig.5 Parametric linear analysis of calorific value

图 6 不同输入参数PHO-BP神经网络测试集预估结果对比

Fig.6 Comparison of the estimated results of PHO-BP neural network test sets with different input parameters

表 4 参数遍历取值范围

Table 4 The parameter value range

序号	操作机构型式	结构型式
1	弹簧	GIS
2	电磁	PASS
3	液簧	瓷柱式
4	液压	罐式
5	/	其它

图 7 2种模型预估结果与现场实测数据对比

Fig.7 The prediction results of the two models are compared with the field measurement data

表 5 变电站碳储量预估结果对比

Table 5 Comparison of estimated carbon storage results of substations

${Z_{\rm C}}$/t		${Z_{{Y_3}}}$/t		${Z_{{Y_6}}}$/t		${\delta _{{Y_3}}}$/%		${\delta _{{Y_6}}}$/%
12015.725		12949.02		12920.01		7.78		7.53

参考文献 25

1	CRESSAULT Y, CONNORD V, HINGANA H, et al. Transport properties of CF₃I thermal plasmas mixed with CO₂, air or N₂ as an alternative to SF₆ plasmas in high-voltage circuit breakers[J]. Journal of Physics D:Applied Physics, 2011, 44 (49): 495202. DOI
2	张咪, 高克利, 侯华, 等. SF₆替代绝缘气体的虚拟筛选与分子设计综述[J]. 高电压技术, 2023, 49 (7): 2816- 2830.
	ZHANG Mi, GAO Keli, HOU Hua, et al. Review on computational screening and molecular design of replacement gases for SF₆[J]. High Voltage Engineering, 2023, 49 (7): 2816- 2830.
3	张智刚, 康重庆. 碳中和目标下构建新型电力系统的挑战与展望[J]. 中国电机工程学报, 2022, 42 (8): 2806- 2818, S1.
	ZHANG Zhigang, KANG Chongqing. Challenges and prospects for constructing the new-type power system towards a carbon neutrality future[J]. Proceedings of the CSEE, 2022, 42 (8): 2806- 2818, S1.
4	彭静, 王军, 亓富军, 等. “双碳” 目标下配电网多阶段扩展规划[J]. 电力系统保护与控制, 2022, 50 (7): 153- 161.
	PENG Jing, WANG Jun, QI Fujun, et al. Multi-stage expansion planning of a distribution network with double-carbon policy[J]. Power System Protection and Control, 2022, 50 (7): 153- 161.
5	王延峰, 申永鹏, 唐耀华, 等. 双碳目标下NB-IoT能源物联网安全构架及关键技术[J]. 电力系统保护与控制, 2022, 50 (8): 179- 187.
	WANG Yanfeng, SHEN Yongpeng, TANG Yaohua, et al. Framework and key technologies for NB-IoT energy Internet of Things to achieve carbon peak and neutrality goals[J]. Power System Protection and Control, 2022, 50 (8): 179- 187.
6	童光毅. 基于双碳目标的智慧能源体系构建[J]. 智慧电力, 2021, 49 (5): 1- 6.
	TONG Guangyi. Construction of smart energy system based on dual carbon goal[J]. Smart Power, 2021, 49 (5): 1- 6.
7	任大伟, 侯金鸣, 肖晋宇, 等. 支撑双碳目标的新型储能发展潜力及路径研究[J]. 中国电力, 2023, 56 (8): 17- 25.
	REN Dawei, HOU Jinming, XIAO Jinyu, et al. Research on development potential and path of new energy storage supporting carbon peak and carbon neutrality[J]. Electric Power, 2023, 56 (8): 17- 25.
8	张逸豪, 王振雷. 基于最大信息系数的分组支持向量数据描述故障检测[J/OL]. 化工学报, 2023: 1–21. (2023-08-25).https://kns.cnki.net/kcms/detail/11.1946.tq.20230824.1808.006.html.
	ZHANG Yihao, WANG Zhenlei. Fault detection using grouped support vector data description based on maximum information coefficient[J/OL]. CIESC Journal, 2023: 1–21. (2023-08-25).https://kns.cnki.net/kcms/detail/11.1946.tq.20230824.1808.006.html.
9	任佳, 孙思宇, 鲍克. 基于最大信息系数和深度残差图卷积的工业过程故障诊断方法[J]. 高校化学工程学报, 2023, 37 (1): 111- 119.
	REN Jia, SUN Siyu, BAO Ke. Fault diagnosis method in industrial processes based on maximal information coefficient and depth residual graph convolution[J]. Journal of Chemical Engineering of Chinese Universities, 2023, 37 (1): 111- 119.
10	赵帅, 宋冰, 侍洪波. 基于加权互信息主元分析算法的质量相关故障检测[J]. 化工学报, 2018, 69 (3): 962- 973.
	ZHAO Shuai, SONG Bing, SHI Hongbo. Quality-related fault detection based on weighted mutual information principal component analysis[J]. CIESC Journal, 2018, 69 (3): 962- 973.
11	周威振, 马越, 邓集瀚, 等. 基于图像分析的气体继电器积气体积辨识及自动取气方法[J]. 南方电网技术, 2023, 17 (9): 112- 119, 150.
	ZHOU Weizhen, MA Yue, DENG Jihan, et al. Image analysis-based gas volume identification and gas automatic collection of gas relay[J]. Southern Power System Technology, 2023, 17 (9): 112- 119, 150.
12	郭志瑞. 基于BP神经网络的启动子序列的预测分析[D]. 重庆: 重庆大学, 2020.
	GUO Zhirui. Prediction and analysis of promoter sequences based on BP neural network[D]. Chongqing: Chongqing University, 2020.
13	吕品, 左金宝, 倪小军. 基于BP神经网络的矿井淋水井筒风温预测[J]. 煤矿安全, 2008, 39 (12): 11- 13.
	LV Pin, ZUO Jinbao, NI Xiaojun. BP neural network for predicting air temperature of watering well in mine[J]. Safety in Coal Mines, 2008, 39 (12): 11- 13.
14	朱历新, 周竞涛, 高俊杰, 等. 基于神经网络的工时定额技术研究[J]. 机械科学与技术, 2004, 23 (6): 702- 704,747.
	ZHU Lixin, ZHOU Jingtao, GAO Junjie, et al. A study on the development of man-hour ration based on artificial neural networks[J]. Mechanical Science and Technology for Aerospace Engineering, 2004, 23 (6): 702- 704,747.
15	周中, 邓卓湘, 陈云, 等. 基于GA-BP神经网络的泡沫轻质土强度预测[J]. 华南理工大学学报(自然科学版), 2022, 50 (11): 125- 132.
	ZHOU Zhong, DENG Zhuoxiang, CHEN Yun, et al. Strength prediction of foam light soil based on GA-BP neural network[J]. Journal of South China University of Technology (Natural Science Edition), 2022, 50 (11): 125- 132.
16	陈明. MATLAB神经网络原理与实例精解[M]. 北京: 清华大学出版社, 2013.
17	杨桢, 马钰超, 李丽, 等. 基于HHT和GA-BP的电压暂降源定位方法[J]. 中国电力, 2022, 55 (3): 97- 104.
	YANG Zhen, MA Yuchao, LI Li, et al. A novel method for voltage sag source location based on HHT and GA-BP[J]. Electric Power, 2022, 55 (3): 97- 104.
18	刘伟吉, 冯嘉豪, 祝效华, 等. 基于动量自适应学习率PSO-BP神经网络的钻速预测模型研究[J]. 科学技术与工程, 2023, 23 (24): 10264- 10272.
	LIU Weiji, FENG Jiahao, ZHU Xiaohua, et al. Prediction model of penetration rate based on PSO-BP neural network with momentum adaptive learning rate[J]. Science Technology and Engineering, 2023, 23 (24): 10264- 10272.
19	王彤彤, 张剑, 涂川, 等. IPSO-BP神经网络在渭河天水段水质评价中的应用[J]. 环境科学与技术, 2013, 36 (8): 175- 181.
	WANG Tongtong, ZHANG Jian, TU Chuan, et al. Application of IPSO-BP neural network in water quality evaluation for Tianshui section of Wei River[J]. Environmental Science & Technology, 2013, 36 (8): 175- 181.
20	甘智超, 郭硕昌, 陶盈盈, 等. 基于PCA-BP神经网络的管道内壁几何形状识别[J]. 固体力学学报, 2023, 44 (5): 622- 636.
	GAN Zhichao, GUO Shuochang, TAO Yingying, et al. Identification of pipeline inner wall geometry based on the PCA-BP neural network[J]. Chinese Journal of Solid Mechanics, 2023, 44 (5): 622- 636.
21	王栋, 张江石, 张琪, 等. 基于HPO-BP神经网络的隧道粉尘处理预测模型[J]. 工业建筑, 2023, 53 (S1): 561- 564,574.
22	NARUEI I, KEYNIA F, SABBAGH MOLAHOSSEINI A. Hunter–prey optimization: algorithm and applications[J]. Soft Computing, 2022, 26 (3): 1279- 1314. DOI
23	朱文广, 李映雪, 杨为群, 等. 基于K-折交叉验证和Stacking融合的短期负荷预测[J]. 电力科学与技术学报, 2021, 36 (1): 87- 95.
	ZHU Wenguang, LI Yingxue, YANG Weiqun, et al. Short-term load forecasting based on the K-fold cross-validation and Stacking ensemble[J]. Journal of Electric Power Science and Technology, 2021, 36 (1): 87- 95.
24	郑焕坤, 曾凡斐, 傅钰, 等. 基于E-C-K-均值聚类和SOP优化的分布式电源双层规划[J]. 太阳能学报, 2022, 43 (2): 127- 135.
	ZHENG Huankun, ZENG Fanfei, FU Yu, et al. Bi-level distributed power planning based on E-C-K-means clustering and SOP optimization[J]. Acta Energiae Solaris Sinica, 2022, 43 (2): 127- 135.
25	赵源上, 林伟芳. 基于皮尔逊相关系数融合密度峰值和熵权法典型场景研究[J]. 中国电力, 2023, 56 (5): 193- 202.
	ZHAO Yuanshang, LIN Weifang. Research on typical scenarios based on fusion density peak value and entropy weight method of Pearson's correlation coefficient[J]. Electric Power, 2023, 56 (5): 193- 202.

[1]	马汝括, 董杰, 王雅湉, 伊国鑫, 丁祥浩, 马乐. 基于神经网络的高寒地区CF₄和SF₆/CF₄检测[J]. 中国电力, 2024, 57(3): 103-112.
[2]	李强, 史元浩, 曾建潮, 陈晓龙. 基于PSO-Elman神经网络的燃煤机组受热面清洁状态预测[J]. 中国电力, 2019, 52(5): 48-53.
[3]	王仕俊, 平常, 薛国斌. 考虑共享储能的社区综合能源系统协同优化研究[J]. 中国电力, 2018, 51(8): 77-84.
[4]	祝晓燕, 张金会, 付士鹏, 朱霄珣. 基于改进PSO的SVM参数优化及其在风速预测中的应用[J]. 中国电力, 2013, 46(11): 105-108.
[5]	高亚静, 赵开轩, 张占龙, 梁海峰. 含风电场与电动汽车充换电站的电力系统经济调度[J]. 中国电力, 2013, 46(11): 84-88.
[6]	王振华, 朱桂萍. 并网状态下微电网中微电源出力优化算法[J]. 中国电力, 2012, 45(1): 55-58.