JPH02244302A - Fuzzy controller - Google Patents

Abstract

PURPOSE:To facilitate a change of a reasoning rule by connecting a fuzzy controller for rule setting which varies a membership function to a fuzzy controller equipped with a circuit for generating the membership function. CONSTITUTION:This fuzzy controller consists of the fuzzy controller 1 which performs fuzzy control with three inputs X, Y, and Z from a controlled system 6 and the fuzzy controller 25 for rule setting which changes the contents of the reasoning rule of the fuzzy controller 1 according to the state of another signal S obtained from the controlled system 6. Then, when the contents of the reasoning rule for one input is changed corresponding to the state of another signal, the reasoning rule of the fuzzy controller 25 for rule setting is set according to the knowledge of an operator for what rule is set according to the state of the signal. Consequently, the contents of the reasoning rule of the fuzzy controller are changed automatically as the operator intends.

Description

Detailed Description of the Invention <Field of Industrial Application> The present invention relates to a fuzzy control device for performing control using a fuzzy controller, and in particular, the present invention relates to a fuzzy control device for controlling a fuzzy controller. This invention relates to a fuzzy control device that allows changes to be made from the outside.

<Prior Art> FIG. 4 shows a schematic configuration of an analog type fuzzy controller 1, in which a plurality of inference units 2a to 2n,
A fixed value calculation unit 3 and an external interface unit 4 are connected to each other via an analog bus 5.

Input from the controlled object 6 is given to the analog bus 5 via the external interface unit 4. Each of the inference units 2a to 2n receives input from this analog bus 5 and performs fuzzy inference according to the inference rules for each unit.The inference outputs from each of the inference units 28 to 2n are received by the definite value calculation unit 3 and superimposed. Thereafter, a final value is calculated by gravity center calculation and output to the analog bus 5. This fixed value is the external interface unit 4.
The signal is given to the controlled object 6 through the.

FIG. 5 shows an example of the configuration of the inference unit 2a (the same applies to other inference units).

The inference unit 2a is a part for executing fuzzy inference according to inference rules with a fuzzy inference antecedent part 7 and a fuzzy inference consequent part 8, and the fuzzy inference antecedent part 7 is a part that includes a plurality of membership function generation circuits 9. .

The above inference rule is called an if, then (if, then) rule, and in the illustrated example, x, y,
The three-person power of z is expressed as follows.

(Rule 1) if X=X1and Y-Yland Z=Z1
then F = F 1 (Rule 2) if X=X2and Y-Y2and Z=Z2
then F = F 2 (Rule 3) if X-X3and Y-Y3and Z-Z3
then F=F3 Each of the membership function generation circuits 9 to 11 in the fuzzy inference antecedent section 7 has the same configuration, and the inputs x, y, and z are respectively applied to each input terminal.

Each membership function generation circuit 9 to 11 has, for example, 5■
It has a voltage source 15 that outputs a plurality of voltages evenly divided between -5■ and each of these voltages is applied to a switch circuit 16.

The switch circuit 16 is given labels NL to PL and NG for selecting membership functions related to inference rules. Here, NL is a large negative value (negative
NM is a medium-negative value.
dium), and NS is a negative small value (negative
ZR is zero, PS is a positive small value, P
M is a positive medium value.
m), PL is a large positive value (positive la
rge), NG prohibits input or output (negat
ion), respectively.

The voltage source 15 and the switch circuit 16 constitute a selection unit that selects membership functions of NL to PL and prohibition of input of NG, and one of the voltages is selected by the switch circuit 16 and sent to the membership voltage generation unit 17. Given.

This membership voltage generating section 17 has input signals x, y.
.

Z is given, and a voltage corresponding to the input signal is output based on the membership function selected here and the input signal.

Further, if the label given to the switch circuit 16 is NG, a prohibition signal is given from the switch circuit 16 to the prohibition circuit 18, and the output of the membership voltage generation section 17 is fixed to the maximum output voltage, for example, 5■. do.

The voltages selected in each membership function generating circuit 9 to 11 are thus applied to the MIN circuit 12, and the MIN
The circuit 12 selects the minimum value among these inputs and supplies it to the truncation circuit 14 of the fuzzy inference consequent part 8.

The membership function generation circuit 13 in the fuzzy inference consequent section 8 includes a switch circuit 19. Decoder 20. It includes a membership function generator 21 and a prohibition circuit 22, and the switch circuit 19 is an NL-P related to inference rules.
Select the membership function of L and prohibit input of NG,
The decoder 20 decodes the label input and supplies it to the membership function generator 21 and the inhibition circuit 22. The membership function generator 21 generates a predetermined voltage distribution on a plurality of (for example, 25) parallel lines based on the label input from the decoder 20, and its output is transmitted to the prohibition circuit 2.
2 to the truncasing circuit 14.

Note that when the decoder 20 decodes NG, the prohibition circuit 22 sets all the parallel outputs of the membership function generator 21 to zero level.

The truncation circuit 14 compares the minimum value selected by the MIN circuit 12 with the output of each parallel line of the membership function generator 21, and sends the smaller signal of each line to the analog bus 5 as an inference output. give.

The inference outputs from each inference unit) 2a to 2n are taken into the definite value calculation unit 3 and superimposed, and then the center of gravity of the combined output is calculated as the definite value output, and the analog bus 5
and further sent to the controlled object 6 via the external interface unit 4.

FIG. 6 shows the relationship between the controlled object 6 and the fuzzy controller 1. The fuzzy controller 1 is given three human forces X, Y, and Z from the controlled object 6. In the fuzzy controller l, the switch circuit 16.19
Inference is performed according to the inference rule related to the membership function set in , and as a result, a determined value output DO is sent to the controlled object 6.

<Problems to be Solved by the Invention> In such a fuzzy control device, there are cases where it is desired to change the content of the inference rule for any input (for example, X) depending on the state of another signal S obtained from the controlled object 6. Normally, an operator knows what kind of rule setting should be performed depending on the state of the signal S, and conventionally, the operator measures the signal S each time and operates the switch circuit 16 for rule setting of the fuzzy controller 1. are doing.

However, in cases where the inference rules are frequently changed or tests are repeated to determine the inference rules, it is cumbersome to take out the circuit board of the fuzzy controller 1 and operate switches on that board. However, there is a problem that the operability is extremely poor.

The present invention was made in view of the above-mentioned problem, and an object of the present invention is to provide a novel fuzzy control device that allows inference rules for fuzzy control to be changed from outside the fuzzy controller.

Means for Solving the Problems In order to achieve the above object, the present invention provides an inference output based on fuzzy inference to a fuzzy controller equipped with a circuit that generates membership functions related to inference rules for fuzzy control. A fuzzy controller for rule setting that changes the membership function according to the above is connected.

For example, if you want to change the content of an inference rule for one of the inputs depending on the state of a certain signal, you can create a rule based on the operator's knowledge of what kind of rule setting should be done depending on the state of that signal. Set the inference rules of the fuzzy controller for configuration. the result,
The contents of the inference rules of the fuzzy controller are automatically changed as intended by the operator by outputting a fixed value from the fuzzy controller for rule setting.

<Embodiment> FIG. 1 shows a fuzzy control device according to an embodiment of the present invention. 6, and a fuzzy controller 25 for rule setting to change the contents of the inference rules of the fuzzy controller 1 according to the state of another signal S obtained from 6.

The fuzzy controller 1 has the configuration shown in FIG. 4, and FIG. 2 specifically shows an example of the configuration of the inference unit 2a.

The inference unit 2a includes a fuzzy inference antecedent section 7 including a plurality of membership function generation circuits 9 to 11 and a MIN circuit 12, and a fuzzy inference consequent section 8 including a membership function generation circuit 13 and a truncation circuit 14. The fuzzy inference antecedent part 7 and the fuzzy inference consequent part 8 execute fuzzy inference according to the above-mentioned inference rules for the three-person effort.

In the embodiment of FIG. 2, each membership function generation circuit 9 to 11 of the fuzzy inference antecedent section 7 includes a comparator 23 in place of the voltage source 15 and switch circuit 16 in FIG.
In addition, the membership function generation circuit 13 of the fuzzy inference consequent part 8 includes an A/D converter instead of the switch circuit 19 in FIG.
A transducer 24 is provided in each case.

In the case of this embodiment, the input to the membership voltage generation section 17 of the fuzzy inference antecedent section 7 is not selected by a switch, but is given as an analog value by the fuzzy controller 25 for rule setting, and a specific analog input , the comparator 23 operates to activate the prohibition circuit 18,
The generation of membership functions for that input is disabled.

In addition, the input to the fuzzy inference consequent part 8 is also given as an analog value from the fuzzy controller 25, and by converting this analog input into a digital value by the A/D converter 24 and giving it to the decoder 20, the member Even in this fuzzy inference consequent part 8, in which the ship function generation unit 21 generates the desired membership function, it is the fuzzy inference antecedent part that the prohibition circuit 22 operates for analog inputs in a specific range. It is the same as 7.

In FIG. 2, the same components as those in the conventional example shown in FIG. 5 are given the same reference numerals, and the explanation thereof will be omitted.

Returning to FIG. 1, the fuzzy controller 25 for rule setting selects the input of any inference unit of the fuzzy controller 1 (this In the embodiment, the content of the inference rule for the input X) of the inference unit t-2a is changed. Therefore, based on the operator's knowledge of what rules should be set depending on the state of the signal S, the inference rules of the fuzzy controller 25 are set in the following expression.

(Rule 1) if S=S l than G=X 1 (Rule 2) if 5=S2 then G-X2 (Rule 3
) if 5=S3 then G=X3 This fixed value output by the fuzzy controller 25 for rule setting corresponds to the content of the inference rule for the input By giving this definite value output to the fuzzy controller l and inputting it to the shift amount input terminal M1 of the membership function generation circuit 9 in the inference unit 2a, automatic control is performed according to the state of the signal S obtained from the controlled object 6. In other words, the inference rules of the inference unit 2a of the fuzzy controller 1 are rewritten.

FIG. 3 shows a state in which the inference rules are rewritten by shifting the membership function generated by the membership function generation circuit 9 in the direction of the arrow.

In the above embodiment, the content of the inference rule for the input of the inference unit 2a in the fuzzy controller 1 is changed according to the state of one signal S, but the contents are not limited to this, and the content of the inference rule is changed depending on the state of a plurality of signals. The inference rules may also be changed.

<Effects of the Invention> As described above, the present invention provides a fuzzy controller equipped with a circuit that generates a membership function related to an inference rule for fuzzy control, which changes the membership function according to an inference output by fuzzy inference. Since the fuzzy controller for rule setting is connected, for example, the inference rules of the fuzzy controller for rule setting can be set based on the operator's knowledge of what kind of rule setting should be performed depending on the state of a certain signal. By doing so, the contents of the inference rules of the fuzzy controller can be automatically changed as intended by the operator by outputting a fixed value from the fuzzy controller for rule setting. Therefore, when the inference rules of the fuzzy controller are frequently changed or tests are repeated to determine the inference rules, it is easier to change the inference rules compared to the conventional switch operation method. The effect is that it can be done.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a fuzzy control device according to an embodiment of the present invention, FIG. 2 is a block diagram showing a configuration example of an inference unit according to the present invention, and FIG. 3 shows a state in which the membership function is shifted. 4 is a block diagram showing the configuration of a fuzzy controller, FIG. 5 is a block diagram showing the configuration of a conventional inference unit, and FIG. 6 is a block diagram showing a conventional fuzzy control device. ■・・・Fuzzy controller

Claims (1)

[Scope of Claim] A fuzzy control device that executes fuzzy control based on input from a controlled object, the fuzzy controller comprising a circuit that generates a membership function related to an inference rule for the fuzzy control. A fuzzy control device connected to a fuzzy controller for setting rules that changes the membership function according to an inference output from inference.