实时获知煤炭发热量对于及时调整电站锅炉风粉配比和提高煤炭燃烧效率具有重要意义，为了实现电力生产中发热量的稳定快速检测，提出了一种近红外光谱（Near Infrared Spectroscopy， NIRS）与X射线荧光光谱（X-ray Fluorescence， XRF）联用的煤炭发热量高稳定检测方法，它结合了NIRS能高稳定检测煤中与发热量正相关的有机基团的优势与XRF能高稳定检测与发热量负相关的成灰元素的特点，大大提高了对煤炭发热量的测量重复性。在光谱预处理中，先将两套光谱融合作为偏最小二乘回归的输入变量进行全谱初步建模，依据回归系数选择NIRS光谱中的有效波段，再将它与XRF光谱中的成灰元素谱线一并融合进行归一化处理。建模时将预处理后的融合光谱数据作为输入变量，利用偏最小二乘回归对煤炭发热量进行建模。实验结果表明，NIRS-XRF联用方法对定标集煤样发热量预测的线性相关度系数（R²）为0.995，对验证集煤样发热量预测的最小均方根误差、平均相对误差和标准偏差分别为0.24 MJ/kg，0.61%和0.05 MJ/kg，测量重复性满足小于0.12 MJ/kg的国家标准。NIRS-XRF联用的煤炭发热量高稳定检测方法有望推广应用于火力发电、煤化工、冶金、水泥和焦化等“高碳”行业，助力我国按期实现碳中和目标。

基于提出的激光诱导击穿光谱（LIBS）和X射线荧光光谱（XRF）的联用多光谱方法，设计了一种基于软件控制的煤质快速分析仪，该分析仪包括LIBS分析模块、XRF分析模块、送样模块、控制模块和操作软件。该仪器不仅发挥了LIBS全元素分析的长处，还继承了XRF高稳定分析的优点，可用于发电厂对压制煤饼进行快速连续的检测。此外，基于偏最小二乘回归方法对数百个煤样进行了光谱分析建模，并完成了工业测试与性能评价。评估结果表明，所建发热量、灰分、挥发分和硫分定标模型的R²分别为0.973、0.986、0.977、0.979，平均绝对误差分别为0.60 MJ/kg、1.24%、0.18%、0.19%，工业分析的平均SD分别为0.11%、0.49%、0.15%、0.09%。模型结果表现出不错的准确度和良好的稳定性，对所有煤炭工业指标的测量重复性均达到甚至优于国标要求。同时，实测结果表明，该仪器对煤炭发热量、灰分、挥发分、硫分的平均绝对误差分别为0.385 MJ/kg、0.830%、0.496%、0.230%，单次样品检测约需5.5 min，能够满足工业现场的实际需求，为煤炭性质的前瞻性预测开辟了道路。

We report the experimental realization of dark state atoms trapping in a nanofiber optical lattice. By applying the magic-wavelength trapping potentials of cesium atoms, the AC Stark shifts are strongly suppressed. The dark magneto-optical trap efficiently transfers the cold atoms from bright (6S_1/2, F = 4) into dark state (6S_1/2, F = 3) for hyperfine energy levels of cesium atoms. The observed transfer efficiency is as high as 98% via saturation measurement. The trapping lifetime of dark state atoms trapped by a nanofiber optical lattice is also investigated, which is the key element for realizing optical storage. This work contributes to the manipulation of atomic electric dipole spin waves and quantum information storage for fiber networks.

Harnessing the dynamics of complex quantum systems is an area of much interest and a quantum simulator has emerged as a promising platform to probe exotic topological phases. Since the flexibility offered by various controllable quantum systems has helped gain insight into the quantum simulation of such complicated problems, an analog quantum simulator has recently shown its feasibility to tackle the problems of exploring topological phases. However, digital quantum simulation and the detection of topological phases still remain elusive. Here, we develop and experimentally realize the digital quantum simulation of topological phases with a solid-state quantum simulator at room temperature. Distinct from previous works dealing with static topological phases, the topological phases emulated here are Floquet topological phases. Furthermore, we also illustrate the procedure of digitally simulating a quantum quench and observing the nonequilibrium dynamics of Floquet topological phases. Using a quantum quench, the 0- and π-energy topological invariants are unambiguously detected through measuring time-averaged spin polarizations. We believe our experiment opens up a new avenue to digitally simulate and detect Floquet topological phases with fast-developed programmable quantum simulators.

We present a method to precisely determine the hyperfine structure constants of the rubidium 5D5/2 and 7S1/2 states in a cascade atomic system. The probe laser is coupled to the 5S1/2→5P3/2 hyperfine transition, while the coupling laser is scanned over the 5P3/2→5D5/2(7S1/2) transition. The high-resolution double-resonance optical pumping spectra are obtained with two counter-propagating laser beams acting on rubidium vapor. The hyperfine splitting structures are accurately measured by an optical frequency ruler based on the acousto-optic modulator, thus, the magnetic dipole hyperfine coupling constant A and quadrupole coupling constant B are determined. It is of great significance for the atomic hyperfine structure and fundamental physics research.