优化多尺度小波变换的里德堡原子EIT-Stark光谱特征提取与高精度电场测量

胡珊珊; 丁超; 宋宏天; 张英; 王保帅; 王浩; 肖冬萍

胡珊珊,丁超,宋宏天,张英,王保帅,王浩,肖冬萍.优化多尺度小波变换的里德堡原子EIT-Stark光谱特征提取与高精度电场测量[J].电测与仪表,2026,63(6):149-158.
Hu Shanshan,Ding Chao,Songhongtian,Zhang Ying,Wang Baoshuai,Wang Hao,Xiao Dongping.Implementation of feature extraction from Rydberg atom EIT-stark spectra of optimized multi-scale wavelet transform and high-precision quantum electric field measurement[J].Electrical Measurement & Instrumentation,2026,63(6):149-158.

优化多尺度小波变换的里德堡原子EIT-Stark光谱特征提取与高精度电场测量

Implementation of feature extraction from Rydberg atom EIT-stark spectra of optimized multi-scale wavelet transform and high-precision quantum electric field measurement

DOI：10.19753/j.issn1001-1390.2026.06.016

中文关键词: 里德堡原子量子电场测量多尺度小波分析类间方差

英文关键词:Rydberg atom, quantum electric field measurement, multi-scale wavelet analysis, between-class variance

基金项目:中国南方电网贵州电网有限责任公司资助项目（GZKJXM20222158、GZKJXM20222147、GZKJXM20222200）；广东省科技计划项目（2021B1212050014）

作者中文名	作者英文名	单位
胡珊珊	Hu Shanshan	南方电网科学研究院有限责任公司
丁超	Ding Chao	贵州电网有限责任公司电力科学研究院
宋宏天	Songhongtian	南方电网科学研究院有限责任公司
张英	Zhang Ying	贵州电网有限责任公司电力科学研究院
王保帅	Wang Baoshuai	南方电网科学研究院有限责任公司
王浩	Wang Hao	输变电装备技术全国重点实验室重庆大学电气工程学院
肖冬萍	Xiao Dongping	输变电装备技术全国重点实验室重庆大学电气工程学院

摘要点击次数: 125

中文摘要:

基于里德堡原子的电场测量是一个新兴发展方向，因其高精度、高稳定性、自校准等优异特性而受到关注，在电力系统中也具有很大的应用空间。非共振频率电场作用于里德堡原子时，所引起的透射光谱峰频移量与电场强度满足特定的数学关系。然而，电场造成里德堡原子简并能级分裂形成包含多个精细能级信息的EIT-Stark光谱，对该光谱峰进行准确和在线特征提取存在诸多难点，并影响对电场的精确反演。为解决上述问题，文中提出了一种基于多尺度小波变换的特征提取方法，解决透射峰重叠和透射峰严重压低等情况下的谱峰特征识别；此外基于类间方差最大化原则对特征提取范围进行限定，降低算法的冗余度，以适应工程应用的需求。与高斯拟合法、基于质谱的多尺度小波变换寻峰方法进行对比，文中所提出特征提取方法在谱峰误检率与真阳性率两个指标上占优。此外，将文中提出的方法融入电场量子测量中，对金属极板不平行以及环境电场干扰进行简单的参数补偿后，实测结果误差小于1%，优于现有传统的电场测量方法以及常规的EIT-Stark光谱处理方法，也对未造成显著能级分裂的小电场的测量误差实现了有效控制。

英文摘要:

Electric field measurement based on Rydberg atom is an emerging development direction, which has attracted attention because of its excellent characteristics such as high precision, high stability, and self-calibration, and also has a large application space in power system. When a non-resonant frequency electric field acts on a Rydberg atom, the frequency shift of the transmission spectrum peak and the electric field strength satisfy a specific mathematical relationship. However, the electric field causes the Rydberg atomic degenerate energy level to split to form an EIT-Stark spectrum containing multiple fine energy level information, which has many difficulties in accurate and online feature extraction of this spectral peak, and affects the accurate inversion of the electric field. In order to solve the above problems, this paper proposes a feature extraction method based on multi-scale wavelet transform to solve the spectral peak feature recognition in the case of transmission peak overlap and transmission peak serious depression. In addition, the scope of feature extraction is limited based on the principle of maximizing variance between classes, and the redundancy of the algorithm is reduced to meet the needs of engineering applications. Compared with the Gaussian fitting method and the multi-scale wavelet transform peak finding method based on mass spectrometry, the feature extraction method proposed in this paper has the advantage in two indicators: the peak false detection rate and the true positive rate. In addition, the method proposed in this paper is integrated into the quantum measurement of electric field, and the error of the measured results is less than 1% after simple parameter compensation for the non-parallel metal plates and the interference of the environmental electric field, which is superior to the existing traditional electric field measurement methods and the conventional EIT-Stark spectral processing method, and also effectively controls the measurement error of the small electric field without significant energy level splitting.

查看全文查看/发表评论下载PDF阅读器

关闭