Method and system for process control with complex inference mechanism
First Claim
1. A process control system including a plurality of control effectors for controlling a process, and a processing device for operating and controlling said control effectors, said processing device comprising:
- (a) a knowledge base storing predetermined membership functions and production rules based on empirical information;
(b) quantitative operation means for determining predicted values of states respectively for a plurality of control objective items by operating on input process information, based on a mathematical model identifying said process;
(c) first inference means for receiving said input process information as its input and for operating on predicted values of states of said plurality of control objective items, based on membership functions predetermined with fuzzy quantities for each input process information;
(d) second inference means, (i) supplied with said input process information, (ii) each predicted value obtained from said quantitative operation means and (iii) each predicted value obtained from said first inference mans as its inputs, for determining the behavior of said process, based on one of said production rules, by diagnosing whether or not the process exhibits a linear behavior, selecting a predicted value from said quantitative operation means when the determination result indicates a linear behavior and selecting a predicted value from said first inference means when the determination result does not indicate a linear behavior; and
(e) third inference means supplied with each predicted value selected by said second inference means, for operating on a satisfaction grade collectively evaluating said plurality of control objective items to thereby determine a combination of control quantities for said control effectors in which the satisfaction grade as a result of said operation is maximum, thereby providing at least one of (i) different types of control effectors and (ii) said plurality of control effectors with an optimal combination of control quantities depending on the behavior of the process.
1 Assignment
0 Petitions
Accused Products
Abstract
A process control system for controlling a process exhibiting both linear behavior and non-linear behavior with a plurality of control quantities has a knowledge base storing therein universal facts, production rules including expert'"'"'s empirical rules, meta rules describing flows of inferences, algorithm methods by mathematical expressions and membership functions referenced in fuzzy inference and a complex inference mechanism. The complex inference mechanism is composed of a complex fuzzy mechanism for inferring directly predicted values on a multi-dimensional space generated by a plurality of elements for evaluation and predicted control objectives evaluated previously with fuzzy quantities, a predicting fuzzy mechanism having the input supplied with the predicted values for determining arithmetically correlation to the membership functions determined previously with the fuzzy quantities and a satisfaction grade obtained by imparting weight to the correlation for each of the control objectives to thereby determine such a combination of the control quantities for control effectors to vary the operation states thereof from the current operation states that the satisfaction grade becomes maximum, and a main inference mechanism having the input supplied with process data to thereby compare selectively the process data with the knowledge stored in the knowledge base for thereby making decision of the process behavior through a forward fuzzy inference with the aid of the production rules based on the empirical knowledge and manage the whole process control system including the complex fuzzy inference mechanism, the predicting fuzzy inference mechanism and the knowledge base. The system can be advantageously employed for intra-tunnel ventilation control and other processes exhibiting both linear and non-linear behaviors.
86 Citations
6 Claims
-
1. A process control system including a plurality of control effectors for controlling a process, and a processing device for operating and controlling said control effectors, said processing device comprising:
-
(a) a knowledge base storing predetermined membership functions and production rules based on empirical information; (b) quantitative operation means for determining predicted values of states respectively for a plurality of control objective items by operating on input process information, based on a mathematical model identifying said process; (c) first inference means for receiving said input process information as its input and for operating on predicted values of states of said plurality of control objective items, based on membership functions predetermined with fuzzy quantities for each input process information; (d) second inference means, (i) supplied with said input process information, (ii) each predicted value obtained from said quantitative operation means and (iii) each predicted value obtained from said first inference mans as its inputs, for determining the behavior of said process, based on one of said production rules, by diagnosing whether or not the process exhibits a linear behavior, selecting a predicted value from said quantitative operation means when the determination result indicates a linear behavior and selecting a predicted value from said first inference means when the determination result does not indicate a linear behavior; and (e) third inference means supplied with each predicted value selected by said second inference means, for operating on a satisfaction grade collectively evaluating said plurality of control objective items to thereby determine a combination of control quantities for said control effectors in which the satisfaction grade as a result of said operation is maximum, thereby providing at least one of (i) different types of control effectors and (ii) said plurality of control effectors with an optimal combination of control quantities depending on the behavior of the process. - View Dependent Claims (2, 3, 4)
-
-
5. A process control system including a plurality of control effectors for controlling a process, and a computer for operating and controlling said control effectors, said computer comprising:
-
(a) a knowledge base storing predetermined membership functions and production rule based on empirical information; (b) quantitative arithmetic operation means for determining predicted values of state quantities for each control objective item by arithmetically operating on input process information, based on a mathematical model identifying the process; (c) complex fuzzy inference means, having an input supplied with said process information, for using complex membership functions predetermined with fuzzy quantities for each input process information and stored in said knowledge base, to qualitatively arithmetically operate on predicted values of said state quantities for each control objective item on a multi-dimensional space of said complex membership functions; (d) main inference means using a group of production rules stored in said knowledge base and including at least one of a rule for evaluating a grade of linearity of a process and a rule for evaluating a grade of non-linearity of the process, and a rule for collectively evaluating both a linearity grade and a non-linearity grade, and having inputs supplied with said process information, each predicted value obtained from said quantitative arithmetic operation means and each predicted value obtained from said complex fuzzy inference means, for diagnosing whether the process exhibits a linear behavior or a non-linear behavior and selecting a predicted value from said quantitative arithmetic operation means when the diagnosis result indicates a linear behavior and selecting a predicted value from said complex fuzzy inference means when the diagnosis result indicates a non-linear behavior; and (e) predicted fuzzy inference means using membership functions stored in said knowledge base ad defining predicted values for said state quantities for each of said control objective items with fuzzy quantities, and having an input supplied with each predicted value selected by said main inference means, for determining evaluation values of each of said control objective items on a plurality of combinations of control quantities for said control effectors which are permitted to be variable from current operation states thereof, arithmetically operating on a grade of satisfaction representing an extent to which all control objective items are collectively satisfied, based on the determined evaluation values and repeating the operation of the satisfaction grade on said plurality of combinations of control quantities, and determining the combination of control quantities for the control effectors in which said grade of satisfaction is maximum from said plurality of combinations based on the result of the operation on the satisfaction grade; thereby arithmetically operating on the control quantities of said control effectors and providing said plurality of control effectors with an optimal combination of control quantities depending on the behavior of the process.
-
-
6. A process control system including a plurality of control effectors for controlling a process, and a processing device for operating and controlling said control effectors, said processing device comprising:
-
(a) a knowledge base storing predetermined membership functions and production rules based on empirical information; (b) quantitative operation means for operating on predicted values of states respectively for a plurality of control objective items from input process information, based on a mathematical model identifying said process; (c) first inference means for receiving said input process information as its input and for operating on predicted values of states of said plurality of control objective items, based membership functions predetermined with fuzzy quantities for each input process information; (d) second inference means, supplied with said input process information, each predicted value obtained from said quantitative operation means and each predicted value obtained from said first inference means as its inputs, for determining the behavior of said process, based on one of said production rules, by diagnosing whether or not he process exhibits a linear behavior, selecting a predicted value from said quantitative operation means when the determination result indicates a linear behavior and selecting a predicted value from said first inference means when the determination result does not indicate a linear behavior; and (e) third inference means supplied with each predicted value selected by said second inference means, for determining a predicted value of the status quantity of each control objective item relative to possible combinations of control quantities which are variable from a current operating state, for evaluating said determined predicted value using a second predetermined membership function, and for determining an optimal one among said possible combinations of control operation quantities based on the evaluation result.
-
Specification