Digital Twin + Industrial Internet for Smart Manufacturing: A Case Study in the Steel Industry
pipeline
smart
app
seeks
optimization across multiple sub-
processes (blast furnaces, converters,
and continuous rolling and oxygen
generation—not shown in the figure).
The development process of the
smart apps involving analytic
modeling typically include:
o Business
requirement
gathering - identify the
operational/business
problems (pain points) and
understand the requirements
for solving these problems;
o Solution design - determine
the output of the analytics
required to solve the specific
problems, evaluate the input
data needed to support the
analytics, decide the analytic
approach (first-principle, data
modeling or a combination of
both) and explore and
experiment to find the best
algorithms;
o Model development - build
and validate the analytic
models with data collected
from
the
targeted
environments; and
o Model Tuning - after the
model is deployed in the real
environment, fine-tune the
model with real world data
and validate the outputs.
porous mass that can be used in a blast
furnace.
The Problem: Due to variation in the quality
and thickness of mineral materials, as well as
equipment operational conditions, there is a
substantial percentage of sinter with
terminal temperature that does not meet
quality
requirements,
requiring
re-
processing—resulting in additional costs
from energy, time and labor.
The Solution: Gather temperature and
pressure data for various wind boxes, along
with qualitative data about the attributes
(such as moisture of the material and
thickness of the mineral materials on the
trolley where these data are available). Next,
predict the terminal temperature of the
sinter
and
provide
operational
recommendations to the operator to adjust
the speed of the trolley to avoid over-
burning or under-burning the sintered ore to
ensure optimal quality.
Gas Boiler Smart App
Gas boiler is not a primary sub-process in the
iron-and-steel making process but a
necessary supportive sub-process that
consumes the surplus blast furnace gas to
generate electricity to be supplemented in
the other sub-processes.
The Problem: Boilers, especially self-
maintained power plant boilers such as
those deployed in iron-and-steel plants, face
large variations in fuel quality and supply, as
well as large fluctuations in operating load.
After a few years of operation, its thermal
efficiency declines, falling below the design
value and resulting in increased operating
cost. It has been challenging to evaluate
Sintering Smart App
Sintering is an early stage sub-process in the
iron-and-steel making process. It fuses iron
ore fines (dust) with other fine materials at a
high temperature to create sinter, a single
- 22 -
November 2019