JPH02273802A - Fuzzy controller - Google Patents

Abstract

PURPOSE:To easily decide whether or not the set and inputted membership function is proper and then to easily correct the membership function if it is not proper by displaying the input value and the membership function related to the input value stored in a storage means with the correlation secured between them. CONSTITUTION:A fuzzy processing means 6 calculates and outputs the fixed value corresponding to the given input value. Then a display means 4 for the membership function is added together with an input means 5 for the membership function, a storage means 3 for the input membership function, and a display control means 2 which displays the given input value and the antecedent part membership function stored in the means 3 with the correlation secured between them. As a result, if the input value is given, the input value and its corresponding membership function are displayed with the correlation secured between them. Thus it is possible to decide whether the set and inputted membership function is proper or not and also to easily correct the improper membership function.

Description

[Detailed description of the invention] (Industrial application field) The present invention relates to fuzzy controllers.

(Prior art) In a fuzzy controller, the degree of fitness for each antecedent membership function of a given input value is calculated, and the antecedent degree of fitness, which is the minimum value of these degrees of fitness, is calculated for each rule. The consequent membership function is truncated in each rule based on the antecedent fitness, then the truncated consequent member membership functions are superimposed, and the determined value corresponding to the center of gravity is input as the input value. I'm going out on a limb.

(Issues to be Solved by the Invention) As mentioned above, in the fuzzy controller, the determined value corresponding to the input value is calculated based on each membership function input and set, and therefore, the determined value corresponding to the input value is calculated based on each membership function input and set. If the determined membership function is inappropriate, an appropriate determined value will not be obtained, and as a result, the controlled device will not be controlled correctly.

For example, when the equipment to be controlled is a heating furnace, and the variable related to its input value is the environmental temperature of the heating furnace, the membership functions related to the environmental temperature in the fuzzy controller are: quite high (PL), somewhat high (PM), and somewhat high. Settings are input in seven levels of labels corresponding to linguistic expressions, such as high (PS), normal (ZR), slightly low (NS), somewhat high (NM), and quite low (NL).

When the fuzzy controller actually performs heating control of the heating furnace, it is assumed that the current environmental temperature is 20° C. and the degree of conformity is 1.0 for the label ZR. On the other hand, if a worker in the environment where the heating furnace is installed feels that the environmental temperature is normal, the setting of the above temporship function is considered to be correct, but if the worker feels that the temperature is a little high, the setting is incorrect. Therefore, it becomes impossible to obtain a correct heating signal as a fixed value.

Conventionally, if the target device was not controlled correctly, it was determined whether the set membership function was appropriate, and if it was determined that it was inappropriate, it was repaired. Therefore, there is a problem in that it takes a lot of effort to repair when the membership function is not appropriate.

The present invention has been made in view of the above points, and includes determining whether or not a set input membership function is appropriate.
An object of the present invention is to provide a fuzzy controller that can easily be repaired when it is not appropriate.

(Means for Solving the Problem) In order to achieve the above object, the present invention provides a fuzzy controller with a boozy processing means for calculating and outputting a definite value corresponding to a given input value, and a display of membership functions. means for inputting a membership function; storage means for the membership function inputted by the input means; The present invention is configured to include display control means for displaying antecedent part membership functions regarding input values on the display means so as to maintain a correlation between them.

(Operation) According to the above configuration, when an input value is given, the display means displays the input value and the membership function corresponding to the input value stored in the storage means so that they each maintain a correlation. be done.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing the overall configuration of the fuzzy controller of the present invention.

In the figure, reference numeral 1 denotes a control section, and the control section 1 is composed of a CPU 2 as a display control means and a memory 3 as a storage means. 4 is a CRT as a means for displaying membership functions, and 5 is a keyboard as means for inputting membership functions. The membership function is input using the keyboard 5 while checking the display on the CRT 4, and this input value is given to each membership function generator of the fuzzy processing device 6 as a fuzzy processing means via the control unit 1. . The fuzzy processing device 6 is
There are a number of inference boards corresponding to the number of rules that obtain input signals from the sensor side and execute inference from ■r to then, and one unit that comprehensively judges the inference results obtained from each inference board to obtain and output a definitive value. Consists of a confirmed board. When sensor signals from various sensors a installed on the controlled device A are input, the inference board performs inference and the final value is outputted on the determination board, and the output signal is transferred to the controlled device A.
The control target unit is configured to be provided with the control target.

Reference numeral 7 denotes an A/D converter, which is provided to A/D convert a sensor signal given from the control target machine mA side and give the result to the control section 1.

Next, the operation of the fuzzy controller will be explained.

FIG. 2 is a flowchart showing the operation when setting the input of the antecedent membership function.

When the antecedent membership function is input on the keyboard 4 (step Sl), the input signal is given to the fuzzy processing device 6 (step S2), and the antecedent membership function related to the input signal is stored in the memory 3.
(Step S3). In the fuzzy processing device 6, the input value is given to the membership function generator of each inference board, and in this state, each membership function generator is in a state of outputting a corresponding fitness signal when given the input value. .

FIG. 3 is a flowchart showing operations during display operation.

When there is an input from the sensor a of the controlled device A, the fuzzy processing device 6 outputs the corresponding determined value, and the input value is given to the control unit l via the A/D converter 7 ( Step S1), whereby a membership function corresponding to the input value is searched among the antecedent membership functions stored in the memory 3 (step S2);
On the CRT 4, the input value and the retrieved membership function are displayed in a state where they have a mutual correlation (step S3).

FIG. 4 is a display example when the sensor a that provides the input value is an environmental temperature sensor, and the input value name lO is shown at the top and the input temperature scale 11 is shown along the bottom. Four correspondingly different labels (NL, NS, P
The membership functions of S and PL) are displayed, and furthermore, the input value is displayed 11a on the scale 11, and the degree of suitability x and y are displayed on the labels NS and PS. In the display example in the figure, the input values are 20°C, N
The goodness of fit in S is 0.7, and the goodness of fit in PS is 0.7.
It is 3.

As described above, if each of the suitability indicators x and y is set to a different shape for each label, such as a square or a circle, the suitability display for each label can be easily confirmed.

Further, each suitability display may be displayed in a different color, or a specific one may be displayed blinking.

By viewing the above display, the field worker can recognize in real time the degree of adaptation of each membership function to the actual environmental temperature, and can determine whether the membership function settings are appropriate. If it is determined that the membership function setting of label NS is inappropriate, for example, by operating the keyboard 5 while looking at the display, the membership function setting of label NS can be set as shown in FIG. change.

FIG. 6 shows another display example. In this display example, input value name 10 is displayed, and input value display 11 is displayed.
The fitness degrees Z 1 , X J , and Z for each label corresponding to a and its input value are all displayed numerically. The display contents are the same as those shown in FIG.

(Effect of the Invention) Therefore, according to the present invention, when an input value is given, the input value and the membership function related to the input value stored in the storage means are maintained in a correlation relationship in the display means. This makes it easier to determine whether the entered membership function is appropriate and to repair it if it is not.

[Brief explanation of drawings]

Figures 1 to 6 relate to embodiments of the present invention, with Figure 1 being a block diagram showing the configuration of the fuzzy controller, Figure 2 being a flowchart showing the operation when membership functions are input, and Figure 3 being the display operation. Flowchart showing the fourth
5 is a diagram showing an example of display, FIG. 5 is a diagram showing an example of repairing the display example of FIG. 4, and FIG. 6 is a diagram showing another example of display. 2... CPU (display control means), 3... memory (storage means), 4... CRT (display means), 5... keyboard (input means), 6... fuzzy processing device (fuzzy processing means).

Claims (1)

[Claims] (1) A fuzzy processing means for calculating and outputting a definite value corresponding to a given input value, a membership function display means, a membership function input means, and a memory for the membership function input by the input means. means, displaying the given input value and the antecedent part membership function related to the input value in the antecedent part membership function stored in the storage means on the display means so that they each maintain a correlation. A fuzzy controller comprising a display control means for displaying a display.