Technologies

Using AI+big data technology to optimize and control the coupling control of raw material drying, calcination, and cooling equipment in the pellet chain loop production process, combined with infrared temperature measurement and visual flame analysis of pellets, the overall process of the chain loop is optimized. Comprehensive control of pellet production is carried out through laboratory data, visual data, temperature measurement data, etc., and parameters in the production process are optimized to achieve independent optimization and energy saving in the chain loop production process. Taking the 1.2 million tons/year chain loop as an example, the annual economic benefit is over 500000 yuan.

Covering the processes of ore blending, material distribution, ignition, exhaust sintering, and endpoint control in sintering machines, AI+big data is used for multi link prediction to achieve predictive control of the entire process from batching to finished products, stabilizing the sintering production process and reducing solid fuel consumption, in response to the complex, coupled, and lagging characteristics of the sintering process. Taking the 360 ㎡ sintering machine with an annual output of 3.7 million tons as an example, the annual energy savings and economic benefits exceed 6.5 million yuan.

Using the technology route of AI+big data+advanced control, optimize and control the entire process of desulfurization and denitrification. Take over the entire process from the flue gas entering the desulfurization tower and SCR, predict the outlet concentration based on the inlet flue gas pollutant concentration, and control it in a coordinated manner to reduce material and fuel consumption and achieve stable compliance with the outlet pollutant concentration. Taking the desulfurization and denitrification system of a 360 ㎡ sintering machine as an example, the annual reduction in material and fuel consumption benefits is over 300000 yuan.

Using AI+big data technology to optimize and control the flue gas furnace and mill, combined with data such as coal feeding rate to regulate the heating of the flue gas furnace, achieving parameter optimization, stable control, and energy saving and consumption reduction from coal feeding to drying. Taking the 80t/h milling system as an example, the annual economic benefits are over 200000 yuan.

Using AI+big data technology route to analyze and verify the data of lime kiln incoming materials, calcination, and discharge processes, match the composition and particle size of incoming materials, and use heat consumption models to implement different calcination strategies for different incoming materials. Under the premise of optimal discharge quality, the optimal calcination strategy is formed to reduce fuel consumption. Taking a single 600 ton double chamber lime kiln as an example, the annual economic benefit is over 600000 yuan.

Optimize and control the feeding, calcination, and cooling processes of the rotary kiln using the technology route of visual+AI+big data. Establish a predictive model based on the temperature and shape of the kiln flame, operating parameters, and inspection and testing data. Independently optimize the flame control and operation control to reduce energy consumption and improve lime quality. Taking an 800 ton/day rotary kiln as an example, the annual economic benefit is over 900000 yuan.

Using AI+big data technology route to optimize and control the overall production process of LF furnace, intelligent takeover is carried out from the entry, smelting, exit, soft blowing and other links of the steel ladle, optimizing and controlling the refining process in terms of steel temperature, alloy collection, slag sample identification, etc., to achieve the improvement of smelting quality and reduce power consumption and alloy consumption. Taking a single 70 ton refining furnace as an example, the annual economic benefits are over 1.5 million yuan.

The intelligent furnace system for VD vacuum furnace adopts the technology route of AI+big data, with various production technology parameters as the optimization control objectives. It predicts the furnace condition through multiple process model sets, and based on the underlying data, realizes the prediction and feedback of argon blowing flow rate and furnace mechanism model prediction. The reinforcement learning model calculates the optimal control strategy for furnace pressure, argon blowing and other control variables, achieving the effect of stable furnace condition and process optimization, and improving the automation level of VD furnace.

Using AI+big data technology route to optimize the production schedule of all refining furnaces in the steel plant. By collecting production data of refining furnaces and the production situation of converters, continuous casting and other processes, the start and stop time of each refining furnace is reasonably allocated, reducing the number of steel ladles produced at the same time, lowering the electricity demand in the factory, and reducing the operating costs of the enterprise. For example, reducing demand by 10000 kW (10%) would result in an annual economic benefit of over 3.6 million yuan.