A New Ion Concentration Detection Method Based on Spectral Matching and Fusion for Zinc Leaching Liquid.

Can Zhou,Wentao Yu,Ye Yang,Hongqiu Zhu,Bei Sun,Yonggang Li
DOI: https://doi.org/10.1109/tim.2022.3228251
IF: 5.6
2023-01-01
IEEE Transactions on Instrumentation and Measurement
Abstract:The suspended micro particles (SMP) in the zinc leaching liquid will cause the scattering of detection light, which will make the spectral data of the liquid rise nonlinearly, leading to the difficulty of accurate detection of ion concentration by ultraviolet-visible spectroscopy (UV-Vis). Therefore, based on UV-Vis, a new ion concentration detection method based on spectral matching and fusion is proposed to weaken the scattering interference of SMP from the perspective of the data model. The method includes the improvement of spectrometer structure, spectral matching, spectral fusion, and prediction of the calibration model. By improving the structure of the spectrometer, the spectral curves of different positions of the leaching liquid were obtained. To accurately discriminate and match multiple spectral curves, a similarity measurement method integrating the differences in amplitude, shape, and information between spectra was proposed, named spectral amplitude, shape, and information differences measure method (SASIM). To effectively use the spectral bands' information screened by SASIM, a spectral data level fusion method was proposed. Finally, a partial least square regression model was established to predict the cobalt ion concentration of the zinc-leaching liquid. The experimental results show that SASIM has stronger spectral discrimination power (40.59%) and smaller spectral discrimination entropy (3.426). The proposed detection method obtains better root mean square error (RMSE) (0.563) and R-2 (0.858), which proves this method can improve the detection accuracy of ion concentration in the leaching liquid. This also provides a new idea for the direct detection of water quality parameters.
What problem does this paper attempt to address?