A Nonlinear Integrated Modeling Method of Extended Kalman Filter Based on Adaboost Algorithm

In the zinc hydrometallurgical purification process, the concentration ratio of zinc ion to trace nickel ion is as high as 105, so that the nickel spectral signal is completely covered by high concentration zinc signal, resulting in low sensitivity and nonlinear characteristics of nickel spectral signal. Aiming at the problem that it is difficult to detect nickel in zinc sulfate solution, this paper proposes a nonlinear integrated modeling method of extended Kalman filter based on Adaboost algorithm. First, a non-linear nickel model is established based on nickel standard solution. Second, an extended Kalman filter wavelength optimization method based on correlation coefficient is proposed to select wavelength variables with high signal sensitivity, large amount of information and strong nonlinear correlation. Finally, a nonlinear integrated modeling method based on Adaboost algorithm is proposed, which uses extended Kalman filter as a basic submodel, and realizes the stable detection of trace nickel through the weighted combination of multiple basic models. The results show that the average relative error of this method for detecting nickel is 4.56%, which achieves accurate detection of trace nickel in zinc sulfate solution.

Mesoporous adsorbent for competitive adsorption of fluoride ions in zinc sulfate solution

Al–La hybrid gel was prepared using an innovative acid-catalyzed and calcination free sol–gel process, and showed excellent adsorption performance for fluoride ions in zinc sulfate solution.

Choosing sorbent for fluoride ion removal from zinc sulfate solutions

The process of zinc sulfate solution purification determines process, economic and environmental production results. Since recently there has been a constant increase in the content of halides in pregnant solutions of zinc production due to the processing of technogenic zinc-containing raw materials, it is relevant to search for methods for removing halides, in particular fluorine, from zinc solutions using a variety of materials. The purpose of this paper was to investigate the effectiveness of akaganeite as an sorbent for fluoride ion removal from zinc sulfate solutions. When using akaganeite, it is especially important to choose a carrier for the sorbent since the nanosized particles of akaganeite make it difficult to clean the solution from the sorbent. Most suitable carriers for this purpose in terms of surface characteristics and physicochemical properties are gypsum and red mud of alumina production. Experiments used a zinc sulfate solution (100 g/dm3 Zn2+, pH = 4.5) containing 26.8–111.4 mg/dm3 F–. The maximum fluoride ion capacity was shown by red mud due to the formation of F–Al complexes. The highest fluorine recovery was demonstrated by red mud with impregnated akaganeite at elevated temperature that facilitates showing akaganeite properties and accelerates the surface ОН– ↔ F– exchange process. The gypsum-based adsorbent successfully removed fluorine due to calcium ions released and calcium fluoride formed. The amount of fluorine removed depends on the sorbent material, its consumption, sorption duration and temperature. The optimal processing conditions were (at pH = 5.5): temperature – 60 °C, process duration – 120 min, composite sorbent consumption – 20÷30 g/dm3. It was shown that composite sorbents based on red mud or gypsum with impregnated akaganeite (β-FeOOH) are most suitable for cleaning zinc solutions from halides. These adsorbents make it possible to achieve the greatest capacity and degree of fluoride ion removal (up to 98–99 %) in the actual pH range of process solutions. The abovementioned sorbents can be regenerated in an alkali solution, and then reused (up to 3–4 cycles).