Data processing: generic control systems or specific application – Specific application – apparatus or process – Chemical process control or monitoring system
Reexamination Certificate
2001-02-09
2004-06-15
Patel, Ramesh (Department: 2121)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Chemical process control or monitoring system
C700S267000, C700S268000, C700S270000, C700S271000, C700S273000, C095S001000, C095S016000, C096S025000, C210S141000, C210S143000, C702S022000, C702S031000, C702S032000
Reexamination Certificate
active
06751527
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of chemical process control and in particular the field of ethylene plant control.
BACKGROUND OF THE INVENTION
Chemical processes such as those run in petrochemical plants must be precisely controlled in order for the processes to run efficiently and economically. Various strategies have been used to facilitate process control of individual systems and to coordinate and integrate multiple systems within a plant operation. Multivariable controllers have been commonly used in petrochemical plants for high-level control functions including distillation control, furnace severity capacity enhancement as well as other applications involving rigid product quality specification and constrained operation.
The use of multivariable controllers in the operation of these plants however, has a number of significant disadvantages. Multivariable controllers are based on complex dynamic models, solved by proprietary software packages that run on expensive host computer systems. Continued operation of these platforms requires investment in computer specialists, hardware and software upgrades, and permanent advanced control maintenance staff, who are specially trained in the use of multivariable control software.
Another disadvantage of the multivariable control strategy is that the multivariable controller causes an oscillation effect in the chemical process after a disturbance. The oscillation is a result from the tendency of the multivariable controller to push the chemical processes directly against equipment constraints or performance specifications in the short term. Consequently, the processes are often run with a small gap between the average operating point and the limit when a disturbance occurs. For many variables such as distillation product purity, there is not sufficient time to wait for feedback information before acting. The requirement for fast action combined with even a small model error can result in an oscillating process that is inefficient.
Another disadvantage of multivariable control strategy of moving quickly to respond to disturbances or to maximize the objective function leads to unsteady flows and inventories within large systems, preventing the equipment from operating at maximum efficiency. This may lead to a system appearing constrained when it simply needs to become steady to achieve its performance specification.
Another disadvantage of the multivariable control strategy is that the basic control of product specifications are run under constrained conditions. Basic regulatory control systems are characterized by the one-to-one pairing between controlled variables and manipulated variables. Some of these parings are configured as automatic control loops, while others are under control of the operator. For example, in traditional systems a manipulated variable such as internal reflux may be paired with ethylene purity, a controlled variable.
With this strategy, the operator increases reflux to improve the ethylene purity while the reboiler achieves the heat balance with the tray temperature controller. The strategy keeps both product compositions near target until a constraint is reached. Under constrained operation, the operator can no longer increase reflux to maintain ethylene product purity. The only remaining choices are to lower temperature target or to reduce feed. If the operator decides to lower the temperature, then the control problem changes and becomes more complex.
Ethylene purity becomes the outer loop of a cascade, with tray temperature control being the inner loop. This is not an effective cascade because of the long time constant of temperature control to effect a change in ethylene purity and non-linearity of the gains.
Dynamics are not the only problem with the conventional multivariable controller strategy. For multivariable controller strategy to be successful at both maximizing the capacity of the tower and approaching steady state operation, the precise temperature target must be identified such that the limits of both product purity and equipment capacity are reached at a given reflux rate. Stated another may, the reflux rate and the heat, which are both manipulated variables, must be set so that dependent variables such as temperature, ethylene purity, and constraint are all controlled simultaneously.
Tray temperature had traditionally been considered the most important regulatory control variable and many multivariable control applications continue to keep the temperature controller in service while the computer manipulates its target. Also, it can become quite difficult to identify the correct temperature setting as feed composition and tray efficiency varies over time.
When a multivariable controller is configured with the tray temperature controller in service, there is a constant struggle to find just the right temperature that satisfies the product specification and the objective function simultaneously. Even the slightest model error can cause the multivariable controller to move the temperature target up and down trying to find the precise solution to the over-defined problem with a low controller as the inner loop of the cascade. As the temperature controller wanders, so does the entire plant with it, beginning with the recycle ethane flow.
With the conventional multivariable controller strategy, recycle ethane flow is continually being moved to control the bottoms level. Whenever the reflux is adjusted for ethylene purity control, there is a corresponding response from the tray temperature controller. As this adjustment proceeds, any imbalance between the reflux and the heat appears as a change in the level of the bottoms, which, in turn, requires a response from the relatively small recycle ethane flow. A small percentage error between the reflux and the heat demands a relatively large percentage change in ethane flow via the bottoms level controller.
The problem for the conventional strategy is even more difficult during constrained operation. When the tower becomes constrained and the temperature target is adjusted to control ethylene purity, the likelihood of recycle ethane flow oscillation is increased. As ethylene purity is pushed closer to the limit this effect becomes more pronounced.
If the recycle ethane flow does begin to oscillate, the capacity of the vaporizer can be exceeded at the peaks of the oscillation cycles, causing material to “back up” into the splitter bottom. When this happens, the operator is powerless to remedy the situation. His only recourse is to cut feed so that the peak recycle ethane can be vaporized.
The speed of the computer in performing dynamic calculations cannot eliminate these fundamental problems relating to moving inventory up and down the tower.
Another disadvantage of the multivariable controller strategy is that it does not provide an adequate integrated response to a disturbance such as changes in the fluid levels of vessels used in the plant. Stewart Jr. et al., U.S. Pat. No. 4,956,763 describes a devices that controls surge levels in vessels used in petrochemical plants by neutron backscatter which allows personnel to respond to disturbances in a control variable, such as the flow rate of liquid feed, in order for the plant to operate safely and efficiently. However, the device only detects the liquid level in a vessel and adjusts the inlet or outlet flow to compensate for incoming surges; it does not address disturbances such as changes in feed composition or feed flow rate.
A new method for achieving high-level advanced control of petrochemical processes without complex dynamic models or host computers is needed.
SUMMARY OF THE INVENTION
Accordingly, an improved method for controlling a chemical processing system having a plurality of operating variables is disclosed in the present invention having operating variables, the method comprising:
(a) selecting a loading variable from the system operating variables, the loading variable being directly related to total energy input into the system and equipment constraint
Hedman & Costigan ,P.C.
Patel Ramesh
Stone & Webster Process Technology Inc.
LandOfFree
Heat and material balance method of process control for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Heat and material balance method of process control for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Heat and material balance method of process control for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3302286