On-line detecting the tuyere coke size and temperature distribution of raceway zone in a working blast furnace

Developing the new injection energy including full oxygen, hydrogen and natural gas injection in raceway zone of blast furnace is one of the effective and feasible measures to reducing the carbon emission and satisfying the carbon neutral plan. While the tuyere combustion condition under all-coke smelting model of blast furnace could supplying a direct manner to observing how the new energy affecting the coke combustion and temperature variety process during raceway process. In this paper, the tuyere coke particle sizes and raceway temperature distribution, as the representative parameters of raceway combustion, are online detected at the same time in a working 2500 m(3) blast furnace under the all-coke smelting model by machine vision, Fully Convolutional Networks (FCN), and colorimetric thermometry method (CTM). The results showed the same ranges compared to research of coke sampling and average temperature in tuyere zone. And the calculation of coke belt length (CBL) in raceway zone by empirical model were modified to evaluate the coke combustion and size decaying under all-coke smelting model. Which could not only benefit to study the coke cracking and formation mechanism in the raceway zone in a blast furnace after the new energy injection, but also provide a new method to online detect particle size and temperature distribution at the same time in complex environment by deep learning and machine vision methods.

Automated summary

Developing new injection energy in blast furnace is an effective measure to reduce carbon emission and achieve carbon neutrality. This study used machine vision and deep learning methods to simultaneously detect coke particle size and temperature distribution in the tuyere zone, providing insights into the combustion process and temperature variation in the all-coke smelting model.