The self driving glass melting process
H.P.H., Glass Service, Inc.
Eisenga M. –Glass Service B.V.
The car industry is quickly changing using more sensors such as radar, sonar and visual camera’s into self driving cars. Furthermore most cars are using
today partial or full electric engines to move the car forward. This revolution is also possible for your glass melting furnace. This progress is made thanks to
cheaper sensors and a change into electricity costs. Even in the middle east the sun is powering oil pump jacks (figure 1.) Today renewable resources in the EU
on average generate around 30% of the required electricity. On some windy and sunny days this can be even around 80%, making electricity price going negative see figure 2.
Figure 1. Solar powered Oilfield Pump jack
Also Glass Industry and glass furnaces can make better use of this advantage. But for this they need an intelligent full automatic control system that can use
the dynamic change of availability and pricing of electricity and even plan ahead. The Electricity market even can provide plans of Electricity ahead that can be
fed into a smart control such as Expert System III
Figure 2. Electricity production and pricing including prediction in Germany (snapshot in May 2016 from Fraunhofer ISE website)
This paper will demonstrate how Intelligent Furnace Design & Operating Practices can increase the overall glass furnace efficiency by utilizing advanced furnace modeling
to help select the most-optimal furnace design for a certain type of glass and pull. This can be achieved by installing (more) electric heating in unique ways. The average
residence time of a common glass furnace can be for example, 30 hours, while the minimum residence time is sometimes only 3 hours, so the whole available space volume is
Figure 3. End fired furnace with melter and barrier electric boosting.
Furthermore, flexible top firing energy input is optimized in conjunction with the electric boost in the most optimal combination, for example see figure 3. Altogether this
can save significant energy, increased furnace pull-rate and reduced emissions can also be achieved.
A flexible furnace using natural gas and electric heating can be operated at higher pulls per square meter and cubic meter when needed and at same time reduce energy costs
at a given pull rate.
Such an optimal designed furnace can be controlled in the economical optimum using advanced Model Based Predictive Control (MBPC). MBPC can decide better than Human Operators
when to use which energy input in the most optimal way, keeping the balance between temperature stability, glass quality, furnace lifetime and actual, to-the-minute costs of
the used energy source (especially electricity).
Figure 4. Flexible use of electricity versus gas heating depending on availability and costs while maintaining automatic glass bottom, throat or riser temperature.
Operating the optimal furnace design (with gas firing and electric heating/boosting) with fully automatic Model Based Predictive Control allows the glass producer to operate the furnace in the optimal cost-effective way with minimum use of operators. Figure 4 shows example of full automatic furnace control.
The technology can offer a pull increase and emission reduction, while energy costs reduction can be 2-6%.