Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
2000-04-13
2004-08-03
Picard, Leo (Department: 2125)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C264S040100
Reexamination Certificate
active
06772044
ABSTRACT:
TECHNICAL FIELD
The present invention relates in general to methods and apparatus for the automatic control of manufacturing processes, and particularly to improve sensing and control methodologies for distributed parameter systems.
BACKGROUND OF THE INVENTION
There is an increased demand for methodologies that can improve the efficiency, reduce long-term costs and reduce the occurrence of errors in distributed parameter-manufacturing processes. Of particular importance are the sensing and control methods utilized in a given distributed parameter system. Several control strategies have been developed and utilized. However, most have proved unsatisfactory. For example, industries that use complex composite manufacturing processes, such as resin transfer molding, require powerful sensing and control systems that utilize many variables and can resolve a multitude of disturbances within a process, to effectuate low error rates and to meet product quality goals.
Resin transfer molding (RTM} is a technique of liquid composite molding (LCM) wherein a thermosetting resin is injected into a closed mold into which a fiber preform has been placed. The fiber preform (typically composed of fiberglass, carbon or graphite) imparts tensile strength, stiffness, toughness and mechanical reinforcement to the part. The thermoset resins (typically epoxies, polyamides, polyesters and phenolics) help in binding the fibers together, and impart compressive strength, dimensional and thermal stability and good fatigue properties to the part.
In the RTM process, a fiber is placed in a mold and the mold is closed. A polymer resin is injected into the mold and the fiber preform is impregnated with the resin. Once the filling of the mold is completed, the part is cured by subjecting it to a temperature-pressure recipe over time. During curing, a polymerization reaction occurs where the resin monomers are cross-linked to produce higher molecular weight polymers. Once the part is cured, the mold is opened, and the part is removed.
RTM is gaining popularity because of its net-shape forming capability, easy tailoring of final part properties, ease of use, and lower molding costs and time relative to other techniques like hand-layup and filament winding. RTM has been used to manufacture ship hulls, wheel bases for armored vehicles, and freight car panels, and numerous parts on military aircraft.
The mold filling stage and the resin curing stage of the RTM process are of particular importance because these stages are often particularly sensitive, thus errors often occur. For example, if resin injection and mold filling are not ideal, the fiber preform may not be completely impregnated and hence, voids or “dry spots” can occur in part. Voids can cause local areas of decreased strength in the part.
The control of resin transfer molding (RTM) is essentially a disturbance rejection problem in a batch setting. The main objective of RTM is to successfully fill a given mold without any dry spots. However, permeability changes within the mold and presence of air channels inducing racetracking scenarios cause major disturbances to the above objective.
Furthermore, current techniques to automatically control manufacturing processes have not adequately overcome a major challenge when developing a control strategy, which is to identify the disturbances for a batch and then implement the control action depending upon the type of disturbance. Indeed, many distributed parameter-manufacturing processes, like RTM mold filling, still rely on operator experience and heuristics. Other control strategies developed in the past have tried to generate a good recipe for filling the mold. The process inputs are specified a priori during the off-line design of this recipe. However, the issue of how a disturbance can be detected in the mold during on-line operation, and how to generate a corrective control action once it is detected, has not been addressed thus far.
For the reasons stated above, and for other reasons stated below which will become apparent to those skilled in the art upon reading and understanding the present specification, there is a need in the art for means and methods for detecting and identifying disturbances during on-line manufacturing operations and for automatically generating a corrective control action once a disturbance is detected in a distributed parameter manufacturing system.
SUMMARY OF THE INVENTION
The above mentioned problems with the automatic control of distributed parameter manufacturing processes and other problems are addressed by the present invention, which will be understood by reading and studying the following specification. Apparatus and methods are described which permit the detection of disturbances in a distributed parameter manufacturing system during operation, and the generation of a corrective action control upon detection of the disturbance.
Briefly, the invention provides a computer-implemented method of controlling a manufacturing process, including identifying a set of possible disturbances in the process, determining locations within the process for a set of sensors such that the data for particular disturbances are significantly different from the sensor data for other disturbances, and constructing a set of one or more decision trees. Each of the trees has inputs from multiple ones of the sensors and has an output indicating one of the disturbances. The invention then determines at least one of a plurality of control actions, each control action ameliorating the disturbance indicated by the output of each terminal node of the decision tree.
REFERENCES:
patent: 5528155 (1996-06-01), King et al.
S.Chemishkian Er al: “Intelligent Algorithms for H/Sub Infinity/-Optimal Placement of Actuators and Sensors in Structural Control” Proceedings of the 1999 American Control Conference, vol. 3, Jun. 2, 1999, pp. 1812-1816, XP001025724 USA p. 1812, Left-Hand Column, Line 1 p. 1814, Left-Hnd Column, Line 8.
J.Han et al: “Optimal Placement of Piezoelectric Sensors and Actuators for Vibration Control of a Composite Plate Using Genetic Algirthms” Smart Materials and Structures, vol. 8, No. 2, Apr. 1999 pp. 257-267, XP001025723, UK, p. 259, Right-Hand Column, Line 16—p. 262, Left-Hand Column, Line 33.
P. Defonseca et al: “A Comparative Study of Methods for Optimising Sensor and Actuator Locations in Active Control Applications” Journal of Sound and Vibration, vol. 221, No. 4, Apr. 8, 1999 pp. 651-679, XP001025728 UK, p. 660, Line 18—p. 663, Line 34.
Foslien-Graber Wendy
Mathur Anoop
Parthasarathy Sanjay
Fredrick Kris T.
Honeywell International , Inc.
Picard Leo
Rao Sheela S.
LandOfFree
Sensor placement and control design for distributed... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Sensor placement and control design for distributed..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sensor placement and control design for distributed... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3360353