How Intelligent Visual Recognition Disrupts Nonferrous Smelting Processes? Decoding the “AI eye” of metal manufacturing

2025-03-10

View: 30

In the non-ferrous metal smelting industry, high-temperature molten 

pool composition fluctuations, electrolyzer plate corrosion, casting 

defects detection and other process challenges, long-term reliance 

on manual experience judgment, resulting in unstable quality control, 

high energy consumption. With the breakthrough of industrial camera, 

multi-spectral imaging and AI algorithm, intelligent visual recognition 

technology is reconstructing the whole process of non-ferrous smelting 

with the “eye of the machine” - from the real-time composition analysis 

in the melting furnace to the micron-level surface defects capture, from

 the prediction of the health of the equipment to the process of 

self-optimization of parameters. From real-time composition analysis 

in the melting furnace to micron-level surface defect capture, from 

equipment health prediction to process parameter self-optimization, 

an industrial change driven by visual intelligence has already started.

A technological breakthrough: Intelligent vision

 cracks the “invisible” problem of 

non-ferrous smelting.

The process environment of non-ferrous smelting is characterized by high 

temperature, high corrosivity and strong interference, which makes 

it difficult for the traditional vision system to operate stably. 

The new generation of intelligent vision technology realizes the

 subversion through three major innovations:

Ultra-high temperature imaging technology:

Adopting short-wave infrared (SWIR) camera and active cooling 

system, it can penetrate smoke and flame inside the smelting 

furnace at 1200℃, and capture the flow state of copper liquid 

and impurity distribution in real time, with an imaging 

accuracy of 0.1mm²/pixel.

Case: A copper smelter deployed a multispectral camera 

at the converter mouth, combined with a CNN algorithm 

to identify Fe3O4 crystallization on the melt surface, 

reducing metal loss due to peroxidation by 3.7%.

Multimodal data fusion:

Synchronized acquisition of visible light, thermal imaging, 

and X-ray images, cross-modal feature extraction through

the Transformer model to accurately identify composite 

defects such as electrolytic anode mud thickness and zinc

 ingot subcutaneous porosity.

Edge Intelligent Computing:

The edge computing box equipped with industrial-grade 

GPU can complete 1280×1024 resolution image processing 

within 0.5 seconds to meet the real-time inspection demand 

of 30 aluminum plates per minute in the casting line.

Second, the scene reconstruction: intelligent vision 

driven by the four core process changes

1. Smelting: from “experience fire control” to “visual alchemy”.

Through the panoramic monitoring system in the furnace, real-time 

analysis of copper matte molten pool color changes and bubble 

morphology, AI model dynamically recommended blowing air 

volume, fuel ratio and other parameters, so that the range of 

fluctuations in crude copper grade from ± 2% narrowed to ± 0.5%.

The infrared thermal camera scans the inner wall of refractory 

material and combines with ResNet to predict the erosion 

location, reducing the maintenance response time from 72 

hours to 4 hours and extending the furnace life by 20%.

2. Electrolysis: the “AI monitor” for the health of the electrodes

High-definition line array camera scans the surface of cathode copper 

plate to detect 13 types of defects such as dendritic crystallization and 

pockmarks, and the detection rate has increased from 82% of manual 

visual inspection to 99.6%, avoiding tens of millions of quality claims 

every year.

Multi-spectral imaging system monitors the corrosion morphology of 

anode plate and predicts the remaining life through LSTM algorithm, 

and the annual power saving of a single tank after optimizing the 

replacement cycle of the plate reaches 42,000 kWh.

3. Pouring process: zero-tolerance defense against micron-level defects

3D structured light camera scans the surface of zinc ingot, reconstructs the 

3D model with 0.01mm precision, identifies cracks, slag and other defects, 

and reduces the defect rate from 1.2% to 0.15%.

High-speed vision system with robotic arm, real-time correction of 

aluminum bar casting offset, dimensional tolerance control within ± 

0.3mm, directly through the high-end aviation aluminum 

certification threshold.

4. Environmental protection control: “full-time eye in the 

sky” of the emission source.

The UV imager is deployed at the flue gas discharge port to dynamically 

monitor the concentration of SO2 and particulate matter, and the data 

real-time linkage DCS system adjusts the dosage of desulfurizer, which 

reduces the incidents of emission exceeding the standard by 90%.

Intelligent inspection drones equipped with hyperspectral cameras in

 the plant identify hidden pipeline leakage points, reducing the loss 

of metal solution leakage by more than 8 million yuan per year.

Value fission: from “quality improvement and cost reduction” 

to “process reengineering”.

The dimension of quality control has been upgraded:

After a lead-zinc smelting enterprise introduced a vision system, 

the surface qualification rate of cathode zinc flakes jumped from 

93% to 99.8%, reducing quality losses by 120 million yuan per year.

Double reduction of energy consumption and material consumption:

Intelligent dosing system based on visual feedback 

of melting pool reduces the amount of copper 

concentrate reducing agent by 18%, and the 

comprehensive energy consumption of tons of copper drops by 14%.

Process knowledge digitization:

Accumulating millions of smelting process image 

data sets, constructing process parameter 

optimization models, and realizing the 

standardized inheritance of the “master's experience”.

Zero safety risk:

Replacement of manual close observation of high-temperature

 melt, toxic gas environment operations, 

to avoid more than 20 major safety accidents.

Fourth, the implementation path: non-ferrous enterprises 

landing visual intelligence triple leap

Hardware selection and weather resistance transformation:

For the melting area (> 800 ℃), electrolysis area (strong acid fog) 

and other scenes, choose explosion-proof, corrosion-resistant

 camera module, with air-cooled / water-cooled protective cover.

Data closed-loop construction:

Open up the data interface between vision system and PLC, MES, 

and establish the millisecond response chain of “image 

acquisition - defect classification - process adjustment”.

Upgrade of human-machine cooperative system:

Develop visualized decision boards, convert AI inspection results into operation

 guidelines, and cultivate new industrial workers who “know how to read data and use AI”.

V. Future picture: the deep integration of visual 

intelligence and industrial meta-universe

Holographic smelting:

Through AR glasses superimposed on the virtual image, 

the operator can observe the reaction process inside 

the electrolysis tank through the lens, realizing “what you see is what you get” process control.

Autonomous optimization system:

Visual data stream is injected into the digital twin in real 

time, and AI autonomously simulates hundreds of 

parameter combinations to continuously approach the process limit.

Visual interconnection of industrial chain:

From mine ore identification to end product traceability, 

visual intelligence will run through the entire non-ferrous

 industry chain, building a quality and credible industrial 

blockchain network.

Conclusion

Intelligent visual identification is pushing non-ferrous metal 

smelting from the era of “black-box operation” to a new era of 

“transparent manufacturing”. This technology not only solves 

the industry's hundred years of pain, but also opens up the

 infinite possibilities of process innovation - when each cluster 

of flames jumping, each piece of metal crystallization by AI 

accurate analysis, the non-ferrous metal industry “smart change” 

tipping point has arrived. In the future, with the cross-fertilization 

of machine vision and quantum sensing, brain-computer interface 

and other technologies, an intelligent smelting ecology of full 

perception, self-decision-making and zero defects is accelerating.