Model-free adaptive control for flexible production systems

Details Model-free adaptive control for flexible production systems Model-free adaptive control for flexible production systems

1998-10-16

2002-03-19

Grant, William (Department: 2121)

Data processing: generic control systems or specific application

Generic control system, apparatus or process

Optimization or adaptive control

C700S028000, C700S048000, C700S052000, C700S055000

Reexamination Certificate

active

06360131

ABSTRACT:

FIELD OF THE INVENTION
The invention relates to industrial process control, and more particularly to a method and apparatus for adaptively controlling various continuous and semi-continuous processes so that a flexible production system can be built to achieve Just-in-Time (JIT) manufacturing to improve quality, increase efficiency, reduce waste, and sharpen competitive edge.
BACKGROUND OF THE INVENTION
A new industrial revolution is evolving. Continuous and semi-continuous process industries are under ever increasing pressure to improve efficiency, profitability, and competitive position. The old concept that a continuous process is the most efficient may not be correct any more.
In the continuous process industry (petrochemical, chemical, power and utility, water treatment, etc.), the processes are usually running 24 hours a day year round. A refinery today may be forced to make only certain products at certain times in specific quantities to achieve the highest economic benefits based on market demands and prices. Over producing can cause big burdens in storage, cash flow, and price, etc. A combined cycle power plant used for cogeneration of electric and thermal energy may need to provide a large amount of steam on demand while at the same time maintaining balance in power generation.
In the semi-continuous process industry (iron and steel, pharmaceutical, semiconductor, cement, pulp and paper, tire and rubber, glass and plastic, food and beverage, etc.), producing different products as needed is a more common practice. However, switching batches or recipes usually causes waste in material, time, and resources. The new challenge is how to cut down on scrap, reduce setup time, and facilitate smooth transitions while keeping products within specifications.
In this evolution, Just-in-Time (JIT) manufacturing methodology can be applied to the process industry to meet the new demands. Just-in-Time covers a wide array of business and production objectives. (1) A schedule that satisfies product demand; anything more or less is waste. (2) Quality is measured not in percentage points but as defects in parts per million; ‘zero defects’ or ‘six sigma’ is the ultimate goal. (3) Inventory does not stand or sit; it flows.
JIT needs 3 major elements to implement: (1) the people to implement and carry out the objectives of JIT; (2) a physical process with the capability to manufacture “zero defects” parts; and (3) a computer system with the intelligence to plan, schedule, optimize, and control the process and operations.
JIT has been adapted and implemented in the discrete manufacturing industries. For instance, most of the cars produced in Japan and United States today are based on JIT manufacturing. Similar to the discrete industry, JIT in the process industry also relies on a flexible production system that combines the effort of people, process, and system. However, the challenge to implement such a system in the process industry is quite different than in the discrete industry.
In the process industry, it is difficult to change production configurations because of the resulting major disturbances in energy balance and material balance. When a process loses its energy and material balance, the consequences can be significant. It can change the process dynamic behavior so much that the process goes out of automatic control. This can cause severe system safety, productivity, and product quality problems. A process going wild has to be put in manual control by experienced engineers or operators. A large process usually takes a long time to settle down and recover the balance. Therefore, although highly desirable, flexible production systems are not popular in process industries because of these associated problems.
The key to implementing a flexible production system in the process industry is a control technology that can deal with large production configuration changes. Processes with changes in batch, load, configuration, and specification are extremely difficult to control. The processes will typically be nonlinear, multivariable, time-variant, structure-variant, and specification-variant. They usually have small open-loop stable ranges and some of them may have bad transient behavior or even open-loop unstable behavior.
The Model-Free Adaptive (MFA) control methodology described in U.S. Pat. No. 6,055,524 and patent application Ser. No. 09/143,165 filed on Aug. 28, 1998 is able to deal with various complex processes in practice. However, it may have difficulty in effectively controlling the following processes:
a) a process whose dynamic behavior changes so much that the MFA control system is out of its operating range; and
b) a process that has bad transient behavior when the process input has a sudden change.
In addition, the Model-Free Adaptive control methodology described in the above-identified patent applications did not address the question of how to deal with certain controller constraints which are important for controlling various continuous or semi-continuous complex processes.
SUMMARY OF THE INVENTION
The present invention overcomes the above-identified limitations of the prior art by using a multifunction MFA controller, in which those controller parameters that govern the operations of the controller at various times are stored in a multifunction advisor, and are selectively applied to the controller as needed from time to time. A measurement filter may be used to cope with special conditions arising during parameter changes. The flexibility and adaptive capability of the inventive MFA controller allows it to control a very broad range of processes. A model-free adaptive transient control system is disclosed to deal with processes that have bad transient behavior due to sudden input changes. This system operates by modifying the process output signal component of the controller's error input upon the occurrence of a transient-generating condition to prevent the transient from affecting the controller. An MFA controller constraint handling method is introduced which allows the user to configure MFA controller constraints easily by maintaining the controller output within dynamically varying ranges. When a control system is bounded by certain constraints, the operator can run the system much more freely because the system will not go beyond the constraints. Higher efficiency and throughput call thereby be achieved while the plant safety is also protected.


REFERENCES:
patent: 5587896 (1996-12-01), Hansen et al.
patent: 5748467 (1998-05-01), Qin et al.

Affiliated with

Also associated with

Rating

Model-free adaptive control for flexible production systems does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Model-free adaptive control for flexible production systems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Model-free adaptive control for flexible production systems will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2831592