Effect of Mild-calcined Coal Gangue Additionon Hot Strength and Refractoriness Under Loadof Ultra Low Cement Castablesin Al_2O_3-SiO_2 System

This work explored the way to improve hot modulus q/＇ rupture （HMOR） and refractoriness under load （RUL） by adding mild-calcined coal gangue （MCG） in Al2O3 -SiO2 ultra low cement （ULC） castables, making use of the in-situ effect of the MCG during heating-up. The influence of respective additions of 5%, 10% and 1.5% of the MCG powders calcined at 700℃ was investigated on HMOR at 1400 ℃ and RUL of the castables. With increased addition of the MCG, HMOR and RUL become significantly enhanced. At 10% of the MCG addition, HMOR reaches 3 MPa, as compared to 0. 3 MPa in the case of no MCG addition. RUL of the specimens dried at 110 ℃for 24 h can be increased by some 270 ℃ with 10% of the MCG addition. RUL 0.11 the specimens preheated at 1 500℃ for 3 h maintains the growth trend with the MCG addition increasing. The microstructure of the heated castable samples was investigated by means of SEM. The in-situ formed needle-like and interlaced mullite in the matrix is contributive to the tmprovement.

放顶煤工作面煤矸混合度自动识别研究进展

自动化放煤是实现自动化综放开采的关键技术,放煤过程中对于煤矸混合度的自动识别是研究的难点。文章依次论述了基于自然射线、声波信号、图像的放顶煤工作面煤矸混合度识别技术的研究进展,提出了基于图像的煤矸混合度识别技术是未来的发展方向,以及综采工作面的煤岩界面的识别技术及分选中的煤矸识别技术的研究进展,总结了为放顶煤工作面煤矸混合度识别提供的宝贵经验。提出了将深度学习理论引入到煤矸混合度识别研究中,为人工智能技术在采矿行业中的应用提供了新思路,对提高综放开采顶煤回收率、提高煤质、实现工作面自动化有重要意义。