No abstract

In order to extract useful information from XRF (X-ray fluorescence) spectrum, and establish a high-accuracy prediction model of soil heavy metal contents, a hybrid model combined Deep Belief Network (DBN) with tree-based model was proposed. The DBN was firstly introduced into feature extraction of XRF spectral data, which can obtain deep layer features of spectrum. Owing to the strong regression ability of tree-based model, it can offset the deficiency of DBN in prediction ability, so it was used for predicting heavy metal contents based on the extracted features. In order to further improve the performance of the model, the parameters of model can be optimized according to the prediction error, which was completed by sparrow search algorithm (SSA) and the gird search. The hybrid model was applied to predict the contents of As and Pb based on spectral data of overlapping peaks. It can be obtained that R2 of As and Pb reached 0.9884 and 0.9358, the MSE of As and Pb are as low as 0.0011 and 0.0058, which outperform other commonly used models. That proved the combination of DBN and tree-based model can obtain more accurate prediction results.

In this paper, we design and theoretically investigate a silicon notched disk metasurface-waveguide (NDMW) system, realizing a new approach to excite Fano resonance by introducing a waveguide layer to generate quasi-guided mode (QGMs) resonances. By comparing the transmission characteristics of the silicon complete disk metasurface (CDM), the silicon complete disk metasurface-waveguide (CDMW) system, and the NDMW system, the formation mechanism of Fano resonances in multiple different excitation modes is analyzed. Furthermore, the Fano resonance characteristics of the NDMW system for different polarization conditions are studied. It can be found that different polarization states can excite different multiple Fano resonances in the NDMW system. Finally, the effects of various structural parameters on the multiple Fano resonances characteristics are investigated. Utilizing the narrow linewidth and significant near-field confinement of the Fano resonances, an optical refractive index sensor can be obtained with the sensitivity of 113 nm/RIU and the maximum quality (Q) factor of 2*10^5 . The proposed system opens up an avenue to develop high-performance biological sensors.

Based on surface plasmon resonance (SPR) and coupled mode theory (CMT), the formation mechanism of double Fano resonances is studied in a hybrid sensing structure composed of subwavelength dielectric grating/metal-dielectric-metal (MDM) waveguide/periodic photonic crystal. Studies show that the dynamically adjustable double discrete states generated in the subwavelength dielectric grating and MDM waveguide are coupled with the continuous state formed in the periodic photonic crystal respectively to achieve double Fano resonances. And the double Fano resonances can be dynamically modulated under the angular modulation by the structure parameters. Surprisingly, the figure of merit (FOM) value of 4.00 x 104 is also achieved, which demonstrates the proposed structure has potential applications in sensing. In addition, the low and high refractive index sensing regions are set up. Simultaneous detection of refractive index ranges 1.00-1.29 and 3.79-4.14 is achieved. Thus, the proposed structure provides an effective reference for studying the dynamic modulation of double Fano resonances and high throughput sensors.

No abstract