Model-Free Adaptive Control
of Steam Injection Systems
|• Handles large product inflow and temperature
||• Temperature control is improved by at
least a 50% reduction in temperature variability.
|• Prevents product from over-heating or
||• Product diversion is minimized to achieve
higher production efficiency and less energy consumption.
|• Handles steam system upsets.
||• A smoother operation is achieved.
|• Reduces variation in temperature and product
||• Product quality and production efficiency
|• Improves efficiency & productivity.
||• Full Return-On-Investment is achieved
in less than one season.
CE (left) is mounted on the panel to
replace PID and controls the tomato paste temperature.
The holding tank (right) shows the 3-way valve that
diverts the tomato paste back if its temperature is
too high or too low.
MFA at Small Planet, a General Mills subsidiary, in
During the tomato harvest season, the
plant operates around the clock to produce tomato paste
using hot breaks, evaporators, steam injectors, and
flash coolers. A smooth operation with consistent product
quality is critical to cost efficiency and profitability.
Like other paste plants, one major problem is the steam
injection system that sterilizes the tomato paste product.
Tomato paste is sent from a holding tank through a steam
injection pipe that injects hot steam onto the tomato
paste. Any combination of 8 steam injectors controlled
by 8 manual valves can be in service. Operators may
manually adjust these valves at any time based on the
An automatic controller is used to manipulate the main
steam flow. The objective is to maintain the temperature
of the tomato paste outflow within a narrow specification.
Too high or too low a temperature will trigger a diversion
signal to send the tomato paste back to the holding
tank. The diverted hot tomato paste will go through
the injection pipe again causing sudden temperature
changes and density changes.
PID had a tough time controlling this process. The loop
is typically left in manual. Frequent product diversion
is common resulting in wasted energy, lower product
quality and less efficiency.
The plant installed a CyboCon
CE model-free adaptive (MFA) control
instrument to replace the old Taylor PID controller.
According to Howard Skinner, Evaporation/Aseptic Supervisor,
his team is able to automatically control the temperature
in all operating conditions even during plant upsets.
The number of product diversions is sharply reduced
due to much better temperature control. His operators
like MFA a lot.