Neural Network-based CF4 and SF6/CF4 Detection in High Altitude and Extreme Cold Regions

doi:10.11930/j.issn.1004-9649.202310033

Abstract

Abstract:

In extreme cold regions, the need to carry multiple instruments to meet the demands for detecting varying concentration levels of CF₄ gas within SF₆ gas leads to inefficient field operations and high costs for instrument acquisition. To overcome this, an SF₆ gas CF₄ concentration detector utilizing pyroelectric detection technology was initially developed, capable of automatically switching among different ranges by selecting appropriate amplification resistances. Subsequently, two neural network models for temperature-pressure collaborative compensation, BP and PSO-BP, were introduced. Data for model predictions were supported by an effective simulated experimental platform, with results indicating the PSO-BP neural network's superiority over the BP network. The PSO-BP neural network's temperature-pressure collaborative compensation model was then embedded within the multi-range detection instrument for CF₄ gas concentration. Simulation experiments demonstrated that the instrument maintains a detection error and repeatability within ±2% and 1.6% across small and large ranges, and within ±0.5% and 0.2% for mixed ratio ranges, respectively, under varying temperatures and pressures. This technological advancement significantly enhances maintenance operations within the power grids of cold regions.

Key words: CF₄ gas concentration detection, pyroelectric detection technology, high altitude and extreme cold regions, three-range, PSO-BP neural network model, collaborative temperature-pressure compensation

Rukuo MA, Jie DONG, Yatian WANG, Guoxin YI, Xianghao DING, Le MA. Neural Network-based CF₄ and SF₆/CF₄ Detection in High Altitude and Extreme Cold Regions[J]. Electric Power, 2024, 57(3): 103-112.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.electricpower.com.cn/EN/10.11930/j.issn.1004-9649.202310033

https://www.electricpower.com.cn/EN/Y2024/V57/I3/103

Figures/Tables 14

Fig.1 Pyroelectric detector detection principle of single-source differential technology

Fig.2 The operating principle of the three-range CF4 sensor

Fig.3 Three-range CF4 detection instrument workflow

Table 1 Different measuring ranges calibration experiment

量程	标气浓度/ (μL·L^–1)	实测浓度/ (μL·L^–1)	$E_{1}$	$E_{2}$	$Δ E$
小量程	0.0	0.8	41562	41669	107
	50.0	49.1	34502	41581	7079
	100.0	98.5	27338	41734	14396
	150.0	152.1	20153	41673	21520
	200.0	198.7	13168	41696	28528
大量程	0.0	1.2	42975	43821	846
	500.0	489.1	36134	43759	7670
	1000.0	1015.7	29110	43768	14565
	1500.0	1532.3	22384	43785	21467
	2000.0	1968.6	15432	43792	28360
混合比	0.0	0.5	44506	45521	1015
	20.0	20.3	36504	45559	9110
	40.0	41.1	28351	45579	17187
	60.0	58.5	20330	45568	25232
	80.0	80.5	12279	45603	33324

Table 1 Different measuring ranges calibration experiment

量程	标气浓度/ (μL·L^–1)	实测浓度/ (μL·L^–1)	$E_{1}$	$E_{2}$	$Δ E$
小量程	0.0	0.8	41562	41669	107
	50.0	49.1	34502	41581	7079
	100.0	98.5	27338	41734	14396
	150.0	152.1	20153	41673	21520
	200.0	198.7	13168	41696	28528
大量程	0.0	1.2	42975	43821	846
	500.0	489.1	36134	43759	7670
	1000.0	1015.7	29110	43768	14565
	1500.0	1532.3	22384	43785	21467
	2000.0	1968.6	15432	43792	28360
混合比	0.0	0.5	44506	45521	1015
	20.0	20.3	36504	45559	9110
	40.0	41.1	28351	45579	17187
	60.0	58.5	20330	45568	25232
	80.0	80.5	12279	45603	33324

Fig.4 PSO-BP neural network model work flow

Fig.5 Experimental platform

Table 2 Before and after normalization data

参数	归一化前		归一化后
参数	min	max	min	max
$Δ E$	51	38673	–1	1
T/℃	–39.4	39.8	–1	1
P/MPa	0.062	0.119	–1	1
D/%	4×10^–5	79.7	–1	1

Table 2 Before and after normalization data

参数	归一化前		归一化后
参数	min	max	min	max
$Δ E$	51	38673	–1	1
T/℃	–39.4	39.8	–1	1
P/MPa	0.062	0.119	–1	1
D/%	4×10^–5	79.7	–1	1

Fig.6 BP neural network temperature-pressure cooperative compensation model structure

Fig.7 Test set prediction result comparison

Fig.8 Comparison of absolute and relative error of test results

Fig.9 Test result fitting curve comparison

Table 3 The main evaluation indicators of uncompensated , BP and PSO-BP compensation

指标	量程	补偿前	BP补偿后	PSO-BP补偿后
$θ_{R S S}$	小量程	424.5	212.9	160.9
	大量程	149555.0	19396.5	5850.8
	混合比	115.2	13.5	3.4
$θ_{M S E}$	小量程	165.1	26.8	6.5
	大量程	17935.0	1843.6	425.4
	混合比	13.1	0.7	0.1
$θ_{M A P E}$	小量程	12.1	5.0	2.2
	大量程	123.0	1843.6	425.4
	混合比	3.3	0.83	0.24

Table 3 The main evaluation indicators of uncompensated , BP and PSO-BP compensation

指标	量程	补偿前	BP补偿后	PSO-BP补偿后
$θ_{R S S}$	小量程	424.5	212.9	160.9
	大量程	149555.0	19396.5	5850.8
	混合比	115.2	13.5	3.4
$θ_{M S E}$	小量程	165.1	26.8	6.5
	大量程	17935.0	1843.6	425.4
	混合比	13.1	0.7	0.1
$θ_{M A P E}$	小量程	12.1	5.0	2.2
	大量程	123.0	1843.6	425.4
	混合比	3.3	0.83	0.24

Fig.10 Relative error of CF4 gas concentration detection after uncompensated, PSO-BP compensation

Table 4 Repeatable experimental data

实验编号	实验组别			均值c	R_SD/%
实验编号	1	2	3	均值c	R_SD/%
1	97.6 μL/L	97.3 μL/L	98.6 μL/L	97.8 μL/L	0.3
4	51.2 μL/L	52.1 μL/L	50.4 μL/L	51.2 μL/L	1.0
7	1003.4 μL/L	1012.7 μL/L	1006.1 μL/L	1007.4 μL/L	1.6
11	1513.2 μL/L	1508.4 μL/L	1503.7 μL/L	1508.4 μL/L	1.1
14	59.3%	59.2%	59.6%	59.4%	0.1
18	21.2%	21.7%	21.6%	21.5%	0.2

References 25

1	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.
2	林林, 陈庆国, 程嵩, 等. 基于密度泛函理论的SF₆潜在可替代性气体介电性能分析[J]. 电工技术学报, 2018, 33 (18): 4382- 4388.
	LIN Lin, CHEN Qingguo, CHENG Song, et al. The analysis of SF₆ potential alternative gas dielectric strength based on density functional theory[J]. Transactions of China Electrotechnical Society, 2018, 33 (18): 4382- 4388.
3	王越. SF₆/N₂混合气体间隙击穿与沿面闪络特性研究[D]. 沈阳: 沈阳工业大学, 2021.
	WANG Yue. Research on gap breakdown and flashover characteristics of SF₆/N₂[D]. Shenyang: Shenyang University of Technology, 2021.
4	和彦淼, 黄印, 颜湘莲, 等. SF₆混合气体和环保替代气体设备标准化研究[J/OL]. 中国电力, 1–8 [2023-06-08] (2023-11-07).http://kns.cnki.net/kcms/detail/11.3265.tm.20230607.1158.008.html.
	HE Yanmiao, HUANG Yin, YAN Xianglian, et al. Standardization research and construction on electrical equipment using SF₆ gas mixture and eco-friendly alternative gas[J]. Electric Power, 1–8 [2023-06-08] (2023-11-07).http://kns.cnki.net/kcms/detail/11.3265.tm.20230607.1158.008.html.
5	赵晓民, 韩国辉, 刘文魁, 等. 高压断路器用SF₆/CF₄混合气体状态参数计算及液化分析[J]. 高压电器, 2016, 52 (12): 204- 208.
	ZHAO Xiaomin, HAN Guohui, LIU Wenkui, et al. Liquefaction and state parameters of SF₆/CF₄ mixed gas in high voltage circuit breaker[J]. High Voltage Apparatus, 2016, 52 (12): 204- 208.
6	周倩, 柯锟, 张晓星, 等. 基于SF₆混合气体绝缘性能的设备补气策略研究[J]. 电力工程技术, 2021, 40 (4): 175- 181. DOI
	ZHOU Qian, KE Kun, ZHANG Xiaoxing, et al. Air supply strategy of equipment based on SF₆ mixed gas insulation performance[J]. Electric Power Engineering Technology, 2021, 40 (4): 175- 181. DOI
7	国家质量监督检验检疫总局, 中国国家标准化管理委员会. 工业六氟化硫: GB/T 12022—2014[S]. 北京: 中国标准出版社, 2014.
8	李予全, 吴司颖, 董曼玲, 等. SF₆气体绝缘设备局部放电分解特征组分三角形诊断法[J]. 绝缘材料, 2022, 55 (11): 86- 92.
	LI Yuquan, WU Siying, DONG Manling, et al. Research on insulation fault diagnosis method of SF₆ equipment based on decomposed components analysis[J]. Insulating Materials, 2022, 55 (11): 86- 92.
9	WU Z T, PANG X B, XING B, et al. Development of a portable and sensitive CO₂ measurement device with NDIR sensor clusters and minimizing water vapor impact[J]. Sustainability, 2023, 15 (2): 1533. DOI
10	熊涛, 高明, Des Gibson, 等. 新型光室结构的主流式NDIR呼吸CO₂监测系统[J]. 红外与激光工程, 2020, 49 (6): 248- 256.
	XIONG Tao, GAO Ming, GIBSON D, et al. Mainstream NDIR breathing CO₂ monitoring system based on new light chamber structure[J]. Infrared and Laser Engineering, 2020, 49 (6): 248- 256.
11	陈胜源. 基于NDIR技术的SF₆气体浓度检测系统设计[D]. 武汉: 华中科技大学, 2018.
	CHEN Shengyuan. Design of SF₆ gas detector-based on NDIR technology[D]. Wuhan: Huazhong University of Science and Technology, 2018.
12	张加宏, 朱涵, 顾芳, 等. 非色散红外CO₂气体传感器的抗温湿度干扰设计[J]. 电子测量与仪器学报, 2022, 36 (7): 160- 169.
	ZHANG Jiahong, ZHU Han, GU Fang, et al. Anti-interference design of temperature and humidity fornon-dispersive infrared CO₂ gas sensor[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36 (7): 160- 169.
13	王富忠. NDIR二氧化碳传感器信号读出电路研究[D]. 成都: 电子科技大学, 2023.
	WANG Fuzhong. Research on signal readout circuit of NDIR CO₂ sensor[D]. Chengdu: University of Electronic Science and Technology of China, 2023.
14	PROKOPIUK A, BIELECKI Z, WOJTAS J. Improving the accuracy of the NDIR-based CO₂ sensor for breath analysis[J]. Metrology and Measurement Systems, 2021: 803–812.
15	孙亚飞, 顾芳, 黄亚磊, 等. 基于GA-WNN温度补偿的红外CO₂气体传感器系统研究[J]. 传感技术学报, 2018, 31 (10): 1613- 1620. DOI
	SUN Yafei, GU Fang, HUANG Yalei, et al. Research on infrared CO₂ gas sensor system with temperature compensation based on GA-WNN[J]. Chinese Journal of Sensors and Actuators, 2018, 31 (10): 1613- 1620. DOI
16	潘甫钱, 胡斌, 梁晓瑜, 等. 非色散红外CO₂传感器温度补偿方法研究[J]. 激光与红外, 2023, 53 (6): 887- 894. DOI
	PAN Fuqian, HU Bin, LIANG Xiaoyu, et al. Research on temperature compensation method of non-dispersive infrared CO₂ sensor[J]. Laser & Infrared, 2023, 53 (6): 887- 894. DOI
17	裴昱, 陈远鸣, 卞晓阳, 等. 基于RBF神经网络气压补偿的非色散红外SF₆气体传感器[J]. 应用光学, 2018, 39 (3): 366- 372.
	PEI Yu, CHEN Yuanming, BIAN Xiaoyang, et al. Non-dispersion infrared SF₆ gas sensor with air pressure compensation based on RBF neural network[J]. Journal of Applied Optics, 2018, 39 (3): 366- 372.
18	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.
19	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. DOI
	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. DOI
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	宋伟业, 刘灵玥, 阎洁, 等. 基于深度强化学习的海上风电集群自进化功率平滑控制方法[J]. 中国电力, 2023, 56 (3): 36- 46.
	SONG Weiye, LIU Lingyue, YAN Jie, et al. Self-evolving power smooth control method for offshore wind power cluster based on deep reinforcement learning[J]. Electric Power, 2023, 56 (3): 36- 46.
22	梁恩豪, 孙军伟, 王延峰. 基于自适应樽海鞘算法优化BP的风光互补并网发电功率预测[J]. 电力系统保护与控制, 2021, 49 (24): 114- 120.
	LIANG Enhao, SUN Junwei, WANG Yanfeng. Wind and solar complementary grid-connected power generation prediction based on BP optimized by a swarm intelligence algorithm[J]. Power System Protection and Control, 2021, 49 (24): 114- 120.
23	陈鹏, 郎需军, 国震, 等. 基于改进BP神经网络和多目标粒子群算法的四回路导线布置优化[J]. 电力科学与技术学报, 2023, 38 (4): 151- 161.
	CHEN Peng, LANG Xujun, GUO Zhen, et al. Optimization of four-circuit wire arrangement based on improved BP neural network and multi-objective particle swarm optimization algorithm[J]. Journal of Electric Power Science and Technology, 2023, 38 (4): 151- 161.
24	张沛, 朱驻军, 刘曌, 等. 基于RReliefF-BP网络的新型电力系统静态电压稳定在线评估方法[J]. 南方电网技术, 2023, 17 (3): 65- 74.
	ZHANG Pei, ZHU Zhujun, LIU Zhao, et al. Online assessment method for static voltage stability of new power system based on RReliefF-BP network[J]. Southern Power System Technology, 2023, 17 (3): 65- 74.
25	王海燕, 刘佳康, 邓亚平. 基于预估-校正综合BP神经网络的短期光伏功率预测[J]. 智慧电力, 2023, 51 (3): 46- 52. DOI
	WANG Haiyan, LIU Jiakang, DENG Yaping. Short-term photovoltaic power forecasting based on predict-correct combination BP neural network[J]. Smart Power, 2023, 51 (3): 46- 52. DOI

Neural Network-based CF₄ and SF₆/CF₄ Detection in High Altitude and Extreme Cold Regions

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 14

References 25

Related Articles 0

Recommended Articles 0

Metrics