基于代理模型的电力电缆温度场快速计算方法及其应用

doi:10.11930/j.issn.1004-9649.202306070

中国电力 ›› 2024, Vol. 57 ›› Issue (5): 178-187.DOI: 10.11930/j.issn.1004-9649.202306070

基于代理模型的电力电缆温度场快速计算方法及其应用

黄莉¹^,²(), 梁云¹^,², 黄辉¹^,², 孙晓艳¹^,², 王珊¹^,², 杨玉强³

1. 国网智能电网研究院有限公司，北京　102209
2. 电力智能传感技术国家电网公司实验室，北京　102209
3. 国网浙江省电力有限公司，浙江杭州　310007

收稿日期:2023-06-20 接受日期:2024-01-15 出版日期:2024-05-28 发布日期:2024-05-16
作者简介:黄莉（1986—），女，通信作者，硕士，高级工程师，从事电力信息物理融合、电力通信、传感等研究，E-mail：huangli@geiri.sgcc.com.cn
基金资助:
国家电网有限公司科技项目（基于多物理场耦合分析的电力电缆感知反演模型及状态表征方法研究，5500-202258316A-2-0-QZ）。

Surrogate Model-based Fast Calculation of Power Cable Temperature Field: Method and Application

Li HUANG¹^,²(), Yun LIANG¹^,², Hui HUANG¹^,², Xiaoyan SUN¹^,², Shan WANG¹^,², Yuqiang YANG³

1. State Grid Smart Grid Research Institute Co., Ltd., Beijing 102209, China
2. Electric Power Intelligent Sensing Technology of State Grid Corporation Laboratory, Beijing 102209, China
3. State Grid Zhejiang Electric Power Co., Ltd., Hangzhou 310007, China

Received:2023-06-20 Accepted:2024-01-15 Online:2024-05-28 Published:2024-05-16
Supported by:
This work is supported by Science and Technology Project of SGCC (Research on the Inversion Model and State Representation Method of Power Cable Sensing Based on Multi-physical Field Coupling Analysis, No.5500-202258316A-2-0-QZ).

摘要/Abstract

摘要：

电缆内部温度是电缆管理的重要参数，针对高精度数值计算和实验测量均不能快速准确预测电缆内部温度的问题，提出一种基于多物理场仿真数据构建代理模型，并基于代理模型实现电缆内部温度场快速计算的方法。以110 kV单芯高压电缆为例，首先对比拉丁超立方设计和最大最小思想优化的拉丁超立方方法构造样本空间的效果，选取最优采样方案，再通过径向基神经网络构建代理模型，以环境温度和载流量为变量构建温度数据集进行测试，进一步通过粒子群算法对代理模型进行优化，最后利用网格节点数据使温度场分布可视化。对比110 kV单芯高压电缆内部温度场计算结果，表明所提的基于代理模型的电力电缆温度场快速计算方法具有较高的精度和效率。

关键词: 代理模型, 电力电缆, 有限元仿真, 快速计算

Abstract:

Accurately predicting the temperature inside cables is crucial for effective cable management. However, this process can be time-consuming through numerical calculations and experimental measurements. To address this issue, a new method for constructing surrogate model based on multi-physics field simulation data is proposed to calculate the cable's internal temperature field quickly and accurately. A single-core 110 kV high-voltage cable is taken as a study subject. Firstly, the Latin hypercube design and the Latin hypercube method optimized by the maximum-minimum idea are compared in constructing the sample space. The optimal sampling scheme is selected to construct the surrogate mode through RBF neural network, and a temperature data set is created for testing using ambient temperature and load capacity. Additionally, the surrogate model is optimized by the particle swarm algorithm, and the temperature field distribution is visualized with the grid node data. The calculation result of the internal temperature of the 110 kV single-core high-voltage cable shows that the proposed surrogate model-based fast calculation method for power cable temperature field has high accuracy and efficiency.

Key words: surrogate model, power cable, finite element simulation, fast calculation

黄莉, 梁云, 黄辉, 孙晓艳, 王珊, 杨玉强. 基于代理模型的电力电缆温度场快速计算方法及其应用[J]. 中国电力, 2024, 57(5): 178-187.

Li HUANG, Yun LIANG, Hui HUANG, Xiaoyan SUN, Shan WANG, Yuqiang YANG. Surrogate Model-based Fast Calculation of Power Cable Temperature Field: Method and Application[J]. Electric Power, 2024, 57(5): 178-187.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202306070

https://www.electricpower.com.cn/CN/Y2024/V57/I5/178

图/表 16

图 1 RBF神经网络结构

Fig.1 Topological structure of RBF neural network

图 2 PSO-RBF算法流程

Fig.2 Flow chart of PSO-RBF algorithm

图 3 代理模型的构建流程

Fig.3 Flow chart of the surrogate model

图 4 电缆结构

Fig.4 Power cable structure

表 1 电缆结构参数

Table 1 Parameters of power cable structure

序号	结构名称	标称厚度/mm	标称外径/mm
1	导体	—	29.9
2	导体屏蔽	1.35	32.6
3	交联聚乙烯绝缘	16.50	65.6
4	绝缘屏蔽	1.00	67.6
5	半导电缓冲阻水带	5.50	78.6
6	皱纹铝护套	6.70	92.0
7	PVC护套	4.50	101.0

图 5 电缆模型三维有限元剖分

Fig.5 3 D finite element section of power cable model

表 2 固体传热场仿真材料参数

Table 2 Material parameters for simulating solid heat transfer field

材料	导热系数/ (W·(m·K)^–1)	密度/ (kg·m^–3)	比热容/ (J·(kg·K)^–1)
导体	400.000	8960	385
导体屏蔽	0.160	950	1005
交联聚乙烯绝缘	0.288	1200	2302
绝缘屏蔽	0.160	950	1005
半导电缓冲阻水带	0.190	520	1720
皱纹铝护套	238.000	2700	900
PVC护套	0.167	1380	2100

图 6 电缆损耗密度分布

Fig.6 Distribution of power cable loss density

图 7 不同加载电流下电缆内部温度分布

Fig.7 Temperature distribution inside power cable under different loading currents

图 8 不同载流量下电缆温度变化

Fig.8 Power cable temperature variation under different load capacities

图 9 不同环境温度下电缆温度变化

Fig.9 Power cable temperature variation under different ambient temperatures

图 10 不同的拉丁超立方设计

Fig.10 Different latin hypercube designs

图 11 RBF代理模型计算结果及均方根误差

Fig.11 Calculation results and RMSE of the RBF surrogate model

图 12 PSO-RBF代理模型计算结果及均方根误差

Fig.12 Calculation results and RMSE of the PSO-RBF surrogate model

表 3 RBF、PSO-RBF代理模型和有限元计算时间对比

Table 3 Comparison of the computation time of 3 methods

方法	CPU计算时间/s	效率提升/%
RBF	4.256363	88.18
PSO-RBF	5.292270	85.30
FEM	36.000000	—

图 13 3种方法计算的温度分布

Fig.13 Temperature distribution calculated by three methods

参考文献 25

1	SEDAGHAT A, DE LEÓN F. Thermal analysis of power cables in free air: evaluation and improvement of the IEC standard ampacity calculations[J]. IEEE Transactions on Power Delivery, 2014, 29 (5): 2306- 2314. DOI
2	郑雁翎. 电缆线路的电缆群载流量优化数值计算模型研究[J]. 高电压技术, 2015, 41 (11): 3760- 3765.
	ZHENG Yanling. Research on optimal numerical caculation model of cable groups ampacity in cable route[J]. High Voltage Engineering, 2015, 41 (11): 3760- 3765.
3	梁永春. 高压电力电缆载流量数值计算[M]. 北京: 国防工业出版社, 2012.
4	王伟平, 周恒, 梁国坚, 等. 110 kV单芯电缆缆芯暂态温度径向感知模型研究[J]. 智慧电力, 2023, 51 (10): 103- 110.
	WANG Weiping, ZHOU Heng, LIANG Guojian, et al. Transient temperature radial sensing model for cable cores of 110 kV single core cables[J]. Smart Power, 2023, 51 (10): 103- 110.
5	潘思蓉, 刘友波, 唐志远, 等. 含深度学习代理模型的有源配电网电压无功控制进化算法[J]. 电力系统保护与控制, 2022, 50 (17): 97- 106. DOI
	PAN Sirong, LIU Youbo, TANG Zhiyuan, et al. An evolutionary algorithm for Volt/Var control in an active distribution network with a deep learning surrogate mode[J]. Power System Protection and Control, 2022, 50 (17): 97- 106. DOI
6	徐浩然, 杨华, 刘超. 风力机翼型尾缘加厚修型优化[J]. 太阳能学报, 2015, 36 (3): 743- 748.
	XU Haoran, YANG Hua, LIU Chao. Optimization of enlarging the thickness of airfoil's trailing edge for wind turbines[J]. Acta Energiae Solaris Sinica, 2015, 36 (3): 743- 748.
7	梁煜, 单肖文. 大型民机翼型变弯度气动特性分析与优化设计[J]. 航空学报, 2016, 37 (3): 790- 798.
	LIANG Yu, SHAN Xiaowen. Aerodynamic analysis and optimization design for variable camber airfoil of civil transport jet[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37 (3): 790- 798.
8	LEE S, KANG S, LEE K. Development of a Gaussian process modeling application for the prediction of missile aerodynamic coefficients[J]. Journal of Mechanical Science and Technology, 2021, 35 (3): 987- 997. DOI
9	肖晓, 程绳, 陈思哲, 等. 邻近三相电抗器的电缆线路接地线发热问题分析与治理[J]. 南方电网技术, 2022, 16 (7): 101- 107, 115. DOI
	XIAO Xiao, CHENG Sheng, CHEN Sizhe, et al. Analysis and treatment of heating problem of cable grounding wire adjacent to three-phase reactor[J]. Southern Power System Technology, 2022, 16 (7): 101- 107, 115. DOI
10	赵选民. 试验设计方法[M]. 北京: 科学出版社, 2010: 276.
11	魏文青, 淡淑恒. 基于有限元和神经网络的电缆输送新能源极限能力评估[J]. 电力科学与技术学报, 2023, 38 (1): 191- 200. DOI
	WEI Wenqing, DAN Shuheng. Evaluation of ultimate capacity of cable transmission new energy based on finite element and neural network[J]. Journal of Electric Power Science and Technology, 2023, 38 (1): 191- 200. DOI
12	何元珍. 拉丁超立方设计的均匀性和正交性研究[D]. 北京: 北京大学, 2011.
	HE Yuanzhen. Study of homogeneity and orthogonality of Latin hypercube design[D]. Beijing: Beijing University, 2011.
13	王群. 基于对称拉丁超立方设计的多目标进化算法[D]. 西安: 西安电子科技大学, 2011.
	WANG Qun. Multiobjective evolutionary algorithm based on symmetric Latin hypercube designs[D]. Xi'an: Xidian University, 2011.
14	叶鹏程, 潘光, 高山. 一种快速优化拉丁超立方试验设计方法[J]. 西北工业大学学报, 2019, 37 (4): 714- 723. DOI
	YE Pengcheng, PAN Guang, GAO Shan. Sampling design method of fast optimal Latin hypercube[J]. Journal of Northwestern Polytechnical University, 2019, 37 (4): 714- 723. DOI
15	刘刚, 池骋, 孙立鹏, 等. 基于线性插值函数的二维异构网格数据插值方法研究[J]. 华北电力大学学报(自然科学版), 2017, 44 (6): 8- 13.
	LIU Gang, CHI Cheng, SUN Lipeng, et al. Research on data interpolation for 2D heterogeneous grid based on linear interpolation function[J]. Journal of North China Electric Power University (Natural Science Edition), 2017, 44 (6): 8- 13.
16	杨涛, 白晗, 刘洋, 等. 基于EEMD和改进Elman神经网络的电力电缆故障监测及预警算法研究[J]. 实验室研究与探索, 2018, 37 (10): 59- 62.
	YANG Tao, BAI Han, LIU Yang, et al. Power cable fault monitoring and early-warning algorithm research based on EEMD and improved Elman neural network[J]. Research and Exploration in Laboratory, 2018, 37 (10): 59- 62.
17	梁汇, 张俊峰, 黄坤, 等. 基于改进径向基函数神经网络的10kV电缆接头温度计算[J]. 信息与电脑(理论版), 2021, 33 (22): 209- 211.
	LIANG Hui, ZHANG Junfeng, HUANG Kun, et al. Calculation of 10kV cable joint temperature based on improved radial basis function neural network[J]. China Computer & Communication, 2021, 33 (22): 209- 211.
18	张水平, 仲伟彪. 改进学习因子和约束因子的混合粒子群算法[J]. 计算机应用研究, 2015, 32 (12): 3626- 3628, 3653.
	ZHANG Shuiping, ZHONG Weibiao. Hybrid particle swarm optimization algorithm of new learning factors and constraint factor[J]. Application Research of Computers, 2015, 32 (12): 3626- 3628, 3653.
19	胡明丽. 基于多物理场耦合模型的高压电力电缆温度场与载流量计算[D]. 广州: 华南理工大学, 2015.
	HU Mingli. Calculation of thermal distribution and ampacity for high-voltage power cables by using multi-physics coupled model[D]. Guangzhou: South China University of Technology, 2015.
20	韩雪永. 双循环圆液力缓速器全充液工况流场降阶预测模型研究[D]. 北京: 北京理工大学, 2016.
	HAN Xueyong. Model reduction and prediction of the flow field in dual torus hydrodynamic retarder[D]. Beijing: Beijing Institute of Technology, 2016.
21	唐科, 文武, 阮江军, 等. 基于有限元法的单芯电缆温度场仿真研究[J]. 武汉大学学报(工学版), 2018, 51 (9): 811- 816.
	TANG Ke, WEN Wu, RUAN Jiangjun, et al. Temperature field simulation of single core cable based on FEM[J]. Engineering Journal of Wuhan University, 2018, 51 (9): 811- 816.
22	张磊, 朱立翔, 陈青云. 不同频率谐波下单芯电缆温度场仿真分析[J]. 湖北电力, 2016, 40 (3): 45- 51.
	ZHANG Lei, ZHU Lixiang, CHEN Qingyun. Simulation analysis of single core cable temperature field with different frequency harmonics[J]. Hubei Electric Power, 2016, 40 (3): 45- 51.
23	宋贺. 高压电缆XLPE绝缘的电热特性有限元分析[D]. 保定: 华北电力大学(保定), 2020.
	SONG He. Finite element analysis of electrothermal characteristics of high voltage cable XLPE insulation[D]. Baoding: North China Electric Power University, 2020.
24	唐科, 阮江军, 唐烈峥, 等. 三芯电缆接头温度场计算[J]. 高电压技术, 2019, 45 (11): 3571- 3578.
	TANG Ke, RUAN Jiangjun, TANG Liezheng, et al. Temperature field calculation of three core cable joint[J]. High Voltage Engineering, 2019, 45 (11): 3571- 3578.
25	何晗瑾. 基于代理模型的多目标方法研究及应用[D]. 成都: 电子科技大学, 2022.
	HE Hanjin. Research and application of multi-objective method based on surrogate model[D]. Chengdu: University of Electronic Science and Technology of China, 2022.

[1]	董子凡, 任劼帅, 殷建刚, 陈隽, 文雅钦, 李劲彬. GIS双断口隔离开关不停电试验电场分布及击穿特性[J]. 中国电力, 2023, 56(9): 168-177.
[2]	胡筱曼, 田伟堃, 宋关羽, 于浩. 基于代理模型加速的园区综合能源系统双目标滚动运行优化[J]. 中国电力, 2023, 56(4): 130-137,145.
[3]	陈璨, 白明辉, 张婉明, 席海阔, 尹兆磊, 邵尹池, 王枭枭, 马原. 分布式光伏边界渗透率快速定位及消纳方案择优[J]. 中国电力, 2022, 55(8): 40-50.
[4]	刘浩然, 周凯, 王昱皓, 李蓉, 傅尧, 饶显杰. 基于电磁耦合注入的电力电缆局部缺陷在线定位方法[J]. 中国电力, 2021, 54(12): 177-185,194.
[5]	张成, 王卫东, 杨延滨, 赵洋, 周弋, 张竟成. 新型电缆多参量带电检测装置设计与应用[J]. 中国电力, 2021, 54(10): 144-151.
[6]	柴鹏, 周灏, 张煜, 赵文杰, 杨潇, 周苗, 李明贞. 基于双端行波法的电缆线路短路故障定位改进[J]. 中国电力, 2020, 53(11): 168-174.
[7]	张佳庆, 范明豪, 李伟, 王刘芳, 武海澄, 汪书苹. 电力电缆隧道机械通风防护技术研究[J]. 中国电力, 2017, 50(6): 113-119.
[8]	张磊，郑新龙，高震，彭维龙，林勇，徐政. 电力电缆极限传输长度研究[J]. 中国电力, 2016, 49(8): 45-49.
[9]	黄雄峰，徐彬昭，张宇娇，崔厚坤，王庭华，安增军. 基于参数拟合技术直埋电缆的缆芯暂态温度计算[J]. 中国电力, 2016, 49(10): 60-66.
[10]	贾自杭，李燕青. 智能变电站电缆局部放电在线监测系统的研究[J]. 中国电力, 2015, 48(1): 137-141.

基于代理模型的电力电缆温度场快速计算方法及其应用

Surrogate Model-based Fast Calculation of Power Cable Temperature Field: Method and Application

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 25

相关文章 10

编辑推荐

Metrics

模态框（Modal）标题

基于代理模型的电力电缆温度场快速计算方法及其应用

Surrogate Model-based Fast Calculation of Power Cable Temperature Field: Method and Application

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 25

相关文章 10

编辑推荐

Metrics