JPH0424740A - Learning method for prewired neuro of rule part - Google Patents

Abstract

PURPOSE:To efficiently learn to adjust the weight of a rule part and that of the antecedent implementing part of a membership function by learning the weight of the rule part and subsequently the weight of the antecedent implementing part of the membership function after initializing a weight value for the rule part, the antecedent implementing part of the membership function and the consequent implementing part of a membership function. CONSTITUTION:An initial weight value to which preliminarily stored knowledge is introduced is set for each of an antecedent implementing part 10 of the membership function, a rule part 12 and a consequent implementing part 14 of the membership function. Then, the weight of the rule part 12 is learned by using a learning data and after that, the weight of the antecedent part 10 is learned. The weight value can be initialized for the rule part 12 by using a random number or by using both the preliminarily stored knowledge and the random number. Thus, it is possible to efficiently learn to adjust the weight of the rule part and that of the antecedent implementing part of the membership function.

Description

[Detailed Description of the Invention] [Summary] A control output corresponding to an input is generated according to a hierarchical network structure that realizes a fuzzy control rule, which is composed of an antecedent membership function implementation unit, a rule unit, and a consequent membership function implementation unit. Regarding the learning method of the rule part pre-wired neuro, we have developed an antecedent membership function realization part, a rule part and a consequent part membership function realization part, with the aim of efficiently learning to perform control suitable for the target system. After setting initial weight values that introduce pre-existing knowledge for each, the weights of the rule part are first learned using the training data, and after the weight learning of the rule part is completed, the antecedent part membership function is The implementation unit is configured to learn the weights. Note that the initial setting of the weight values of the rule section may be performed using random numbers, or may be performed using a combination of knowledge held in advance and random numbers.

[Industrial Application Field] The present invention relates to a learning method for a rule section prewired neuron that combines fuzzy control and a neural network.

In information processing using neurons, learning data consisting of a set of input patterns of the target system and desired output patterns for that input pattern is used in a hierarchical format.
It is presented to the physical network to learn and perform adaptive processing. In particular, a processing method called pack propagation method is attracting attention due to its high practicality in this learning process.

On the other hand, fuzzy theory was proposed by Zadeh in 1960, and is a theory that models the process of thinking and judgment based on ambiguity that humans make, such as whether the temperature is "low" or "high." This fuzzy theory is mathematically based on fuzzy set theory, which introduces membership functions to represent ambiguity.

As an application of fuzzy theory, in 1974 Maqmda
Various types of fuzzy control have been used since ni was first used to control steam engines.

Since the system that performs this fuzzy control handles uncertain things, it is necessary to have a configuration that can adaptively respond to adjustments and changes in the control algorithm.

[Prior Art] Generally, the following procedure is taken to construct a fuzzy control system.

■Obtain ambiguous control rules such as ``If the temperature is high, reduce the fire'' held by experienced operators.

■The meaning of the words (propositions) in the control rule is quantified in the form of a membership function, and the control rule is described in a rule expression of the type ``rlF-THEN part''.

■Conduct inspections through simulations and on-site tests.

■Evaluate inspection results and improve membership functions and rules.

To explain in more detail, the fuzzy control rules are, for example, rif x, is big and l2is sm
It is written as all thenllis bigJ. Here, the IF section is called an antecedent section, and is a section that describes conditions regarding control state quantities (control inputs) such as temperature. Further, the THEN section is called a consequent section, and is a section that describes the conditions regarding the control operation amount (control output) of the operating terminal, etc.

Fuzzy control manages multiple fuzzy control rules prepared as the control logic of the target system, and also manages the meaning of ambiguous linguistic expressions such as "large" or "small" written in each fuzzy control rule. Quantify and manage it as a ship function.

The control procedure of the completed fuzzy control system is as follows.

■When a controlled state quantity such as temperature is input from the controlled object, the truth value is calculated according to the membership function of the antecedent part that has been set.

(2) Next, in accordance with the antecedent computation such as selecting the minimum value, a process is executed to determine the value to be applied to the consequent in each fuzzy control rule. For example, r I, is b
Truth value of igJ = 0.8 "x2is smallJ
When the truth value of = 0.5, the truth value = 0.5 is output as the applied value to the consequent part of the fuzzy control rule according to the antecedent part operation that selects the minimum value. Process.

(2) Next, a process is executed to determine an output application value from the input value (applied value) from the antecedent part to the membership function of the consequent part, which has the same amount of control operation.

For example, with two fuzzy control rules, r y+
Applicable value of is bigJ = 0.5ry, is bi
When the applied value of gJ=0.6 is obtained, the applied value=0.6 is selected and determined according to the consequent operation that selects the maximum value.

■Subsequently, the membership function of the control operation amount is reduced (or expanded) according to the determined applied value, and the center of gravity of the figure of the sum of the membership functions for the same control operation amount is calculated. The control operation amount, which is a fuzzy inference value, is calculated and output.

By the way, fuzzy control rules are planned to be generated according to the knowledge possessed by a skilled operator, and it is difficult to establish fuzzy control rules that can realize optimal control from the beginning. Therefore, it is necessary to improve the generated fuzzy control rules through trial and error while evaluating them through simulations and field tests.

Therefore, the inventors of the present application developed a hierarchical network structure that realizes fuzzy control rules by applying neurology.
That is, we have proposed an adaptive fuzzy control system using prewired neuro in the rule section (Patent Application No. 2-602).
(See No. 56, “Hierarchical Network Configuration Data Processing Device and Data Processing System”). That is, the rule part prewire neuro has a hierarchical network configuration that realizes a fuzzy control rule consisting of an antecedent part membership function realization part, a rule part, and a consequent part membership function.

Considering the automatic extraction of fuzzy control rules by the pre-wired neuron of the rule section, the following procedure may be taken.

■Set the membership function of the rule section prewire neuro.

■The rule section prewire neuro is presented with the input pattern of the target system and a set of desirable output patterns for that input pattern, and is made to learn.

■Extract the optimal fuzzy control rule for the target system by analyzing the weights of connections in the rule section.

[Question 14 to be solved by the invention] However, the learning performed for extracting fuzzy control rules for the rule section pre-wired neuron is
The learning method for learning, including the weights of the antecedent membership function part, rule part, and consequent part membership function realization part, has not been determined, and the performance of the roll part prewire neuron has not yet been determined. There is a need to develop efficient learning methods that can bring out the best in students.

The present invention has been made in view of the above situation, and an object of the present invention is to provide a learning method for a rule section pre-wired neuron that can efficiently perform learning for controlling a target system.

[Means to solve the problem] FIG. 1 is a diagram explaining the principle of the present invention.

First, the present invention provides antecedent membership function realization unit 10.1
Or it has a hierarchical network structure consisting of a rule part 12 and a consequent part membership function realization part 14 which are made up of multiple hierarchies,
The rule section 12 is an antecedent membership function realization section 10 in the previous stage.
and/or with the consequent membership function realization unit 14 in the subsequent stage, without internally connecting all the units, some of them are internally combined according to the control rule, and the input control state quantity (X,
.

X2. .

In the present invention, as a learning method for such a rule part prewire neuron 100, [each of the first step antecedent membership function realization unit 10, rule part 12, and consequent membership function realization unit 14] [Second process] Only the weights of the rule section 12 are learned using pre-prepared learning data; [Third process: A priori calculation is performed using pre-prepared learning data. A learning method is configured that includes the following processing steps: learning only the weights of the section membership function realizing section 10.

Here, as a method of setting the initial weight values in the first process, the knowledge (in the figure Set the initial weight values created based on the information (shown as "a priori knowledge").

Another method for setting the initial weight values in the first process is to set initial weight values created using random numbers in the rule unit 12, and then use the antecedent membership function realization unit 10 and the consequent membership function The implementation unit 14 is set with initial weight values created based on knowledge held in advance. Further, in setting the initial weight values of the rule section 12 in the first step, weights that can be set based on knowledge held in advance are initialized using this knowledge;
Others may be initialized using random numbers.

On the other hand, at the start of learning in the second process, the learning flag provided for each connection of the rule section 12 is turned on, and the learning flag provided for each connection of the antecedent membership function realization section 10 and the consequent membership function realization section 14 is turned on. The provided learning flag is turned off, and only the weight values of the connections having the learning flag in the on state are optimized through learning processing.

Furthermore, at the start of learning in the third process, a learning flag provided for each connection of the antecedent membership function realization section 10 is turned on, and a learning flag provided for each connection of the rule section 12 and the consequent membership function realization section 14 is turned on. The learning flag that has been set is turned off, and only the weight values of the connections that have the learning flag that is on are optimized through learning processing.

[Operations] According to the learning method of the rule section pre-wired neuron of the present invention having such a configuration, the following operations can be obtained.

First, for the rule part of the prewired neuro, initialize the weight value based on the knowledge that you have in advance, initialize the weight value using random numbers, or set the weight value in combination with the antecedent part membership function. Initial weight values are set for the realization part and the consequent membership function realization part based on knowledge held in advance.

Next, the learning process is started, the input cover turn for learning is presented to the rule unit pre-wire neuron, and the output cover turn from the rule unit pre-wire neuron is a learning ratio cover turn (the teacher For example, learning of weights using a pack propagation method is performed only for the rule section, in which the initially set weight values are corrected so as to substantially match the signal).

Once the weight learning of the rule part is completed, weight learning is similarly performed on the antecedent membership function realization part using the pack propagation method.

In this way, by learning the rule part of the rule part pre-wired neuron and then learning the antecedent part membership function realization, the antecedent part member can be initialized by introducing pre-existing knowledge. While effectively utilizing the weights of the ship function realization section and the consequent membership function realization section, it is possible to adjust the weight of the rule section and to adjust the weight of the antecedent membership function realization section.

In addition, by initializing the weights of the rule section that incorporates pre-existing knowledge, learning can be facilitated and learning efficiency can be increased.

Furthermore, by initializing the weights of the rule section using random numbers, it is possible to maximize the neural learning function.

[Embodiment] FIG. 2 is a learning operation flowchart showing a processing procedure as an embodiment of the learning method of the rule section pre-wire neuro according to the present invention. The rule section pre-wire neuron that is the object of the learning method according to the procedure of FIG. 2 is, for example, the antecedent membership function realization section 10 and the rule section 12 shown in FIG.
, a consequent membership function implementation section 14, and a center of gravity calculation implementation section 18, which have a hierarchical network structure for realizing fuzzy control rules.

Here, the rule section pre-wire neuro means the rule section 12.
has a network structure in which all units are not internally connected with the antecedent membership function realizing unit 10 in the previous stage and/or the consequent membership function realizing unit 14 in the subsequent stage, but some are internally connected according to the control rule. , is compared to a rule-part fully connected neuron that internally connects everything between hierarchical units.

In the learning process shown in FIG. 2, the following steps 81 to S6 are performed.

[Step SL] Weight Cab of the antecedent membership function realization unit 10.

Cbc is initialized based on the introduction of prior knowledge to realize a given membership function.

[Step S2: Initialize the weights Ccd and Cde of the co-rule unit 12 based on the introduction of pre-existing knowledge so as to realize the given fuzzy rule.

[Step S3] The weight Cel of the consequent membership function realization unit 14 is initialized based on the introduction of knowledge held in advance so as to realize the given membership function.

[Step S4: Initialize the weight Cfg of the center-of-gravity calculation implementation unit 18 so as to realize center-of-gravity calculation.

[Step S5] A phase/group correspondence table indicating neuron groups for each phase in which learning processing is performed and a group/connection correspondence table indicating connection groups belonging to each neuron group are set. In the present invention, the phase/group correspondence table is set so that the weight of the rule section 12 is learned in need 1, and the weight of the antecedent membership function realization section 10 is learned in phase 2.

[Step S6] Start the learning scheduler and perform learning processing using the pack propagation method.

Hereinafter, steps 81 to 81 shown in the learning operation flow of FIG.
Each of S6 will be explained in detail.

[Configuration of Rule Section Prewire Neuron] The rule section prewire port, which is the object of the present invention, has a hierarchical network structure capable of realizing the fuzzy control rules shown in FIG. 3, for example.

In FIG. 3, each node shown in each part represents one basic unit, and two types of basic units are used: a sigmoid function unit shown by a black circle, and a linear function unit shown by a white circle.

FIG. 4 shows the configuration of the basic unit shown in FIG. 3.

In FIG. 4, the basic unit 20 constitutes a multi-power single output system, and includes a multiplication section 22 that multiplies a plurality of inputs by the weight value of each internal connection, and a multiplication section 22 that adds all the multiplication results of the multiplication section 22. It is composed of an adding section 24 and a threshold processing section 26 that performs threshold processing on the output of the adding section 24.

Here, if the first layer is the h layer and the second layer is the i layer, the multiplier 22
The addition unit 24 executes the calculation of equation (1) below, and the threshold processing unit 26 executes the calculation of equation (2) below.

X p I=Σy, hW+h
In (1), y,, = 1 / (1+ e−”p′−”)
(2) However, h Unit number i of the 76th layer: Unit number p of the i layer: Pattern number θl of the input signal: Threshold value W+h of the first unit of the i layer: 11 Weight value xp of the internal connection between the i layers: h
Sum of products yph of inputs from each unit of the layer to the 1st unit of the i layer: Output Y-1 from the h-th unit of the h-layer for the input signal of the p-th pattern: 1 of the i-layer for the input signal of the p-th pattern Output from the number unit Here, the above equation (2) is known as a sigmoid function, and is the threshold value processing of the basic unit indicated by the black circle shown in FIG.

In contrast, the threshold processing unit 26 ypl=x.

The linear function unit set as follows becomes the white circle linear function unit in FIG. However, the input section, which is the first unit of the antecedent membership function realization section 10, does not necessarily have to be a linear function unit. The hardware configuration of the basic unit shown in FIG. 4 is shown in Japanese Patent Application No. 63-216865 titled ``Network Configuration Data Processing Apparatus.''

Referring again to FIG. 3, the configuration of each part of the rule section pre-wire neuro will be explained.

First, the antecedent membership function realization unit 10 processes each input variable X, , X2 . ...Xn is input, input variable X1
Output the grade value of the membership function of ~Xn. In the figure, each membership function is a one-variable function for simplicity, but it may be a multi-variable function. In other words, variable X,
The membership function may be XI &X2 (SA) with inputs XI and X2 as inputs.

The rule section 12 includes an antecedent section rule section 12-1 and a consequent section rule section 12-2. The antecedent part rule part 12-1 is
Membership function X1(S A
) to X n (LA) as input, and outputs the grade value of each rule.

Further, the consequent part rule parts 12 to 2 have the grade value L of the rule.
H3-1 to LH8-n are input, and grade values of membership functions y (SA) to y (LA) of output variables are output.

Furthermore, the consequent membership function realization unit 14 (including the centroid calculation realization unit 18) calculates the membership function y(S
Input the grade values of A) to y (LA) and output the value of the output variable Y.

Here, the antecedent part rule part 12-1 and the consequent part rule part 12
Since both the input and output of the rule unit 12 including -2 are the grade values of the membership function, these can be cascaded and multi-stage inference is possible.

Further, the neurons connecting each module 10.12-1.12-2.14 can be shared. For example, the output neuron of the antecedent membership function implementation unit 10 and the input neuron of the antecedent rule unit 12-1 may actually be the same.

Further, in FIG. 3, each unit is described as one neuron, but if the function of the unit can be specified, it may be used as a gate circuit, an arithmetic unit, etc. that realizes the unit function instead of a neuron.

FIG. 5A shows a specific embodiment of the rule section pre-wired neuron of the present invention, taking as an example a case where two inputs, X, , and X2 are given to obtain a control output Y.
The configuration of each part will be explained in detail with reference to the drawings as follows.

The input unit 16 is composed of separation units 28-1, 28-2, and divides the inputs X, . . . Antecedent part.

The membership function implementation unit 10 is composed of four sigmoid function units h30-1 to 30-4 and four linear function units 32-1 to 32-3.

The rule section 12 includes, for example, sigmoid function units 34-1 to 34-8 that realize eight fuzzy rules, and the antecedent membership function realization section 10 connects each fuzzy rule according to the descriptions 1 to L8. It is being done. The outputs of the sigmoid function units 34-1 to 34-8 are connected according to fuzzy rules to two linear function units 36-1 and 36-2 which perform consequent operations according to fuzzy rules.

The consequent membership function realization unit 14 is composed of, for example, nine linear function units 38-1 to 38-9, and performs the same operation based on the two unit output values (grade values) output from the rule unit 12. The membership function of the quantity position is reduced (or enlarged) and the function sum of the two reduced (or enlarged) membership functions is calculated.

Furthermore, the center of gravity calculation implementation unit 18 is composed of a center of gravity determining element output device 40 and a center of gravity calculation device 42. The linear function units 40-1 and 40-2 of the center-of-gravity determining element output device 40 calculate the weighting of the unit output from the consequent membership function realization unit 14 by calculating the sum and calculating the two center-of-gravity calculation elements Y,... Y2 is calculated, and finally the center of gravity calculation device 42 calculates the two center of gravity calculation elements Y,
, Y2, and generates a control output Y as a fuzzy inference value.

FIG. 5B shows neuron groups and connection groups for realizing the learning method according to the present invention for the rule section pre-wire unit shown in FIG. 5A.

In FIG. 5B, there are seven neuron groups, Ga-Gg, among which neuron group G in the input section
a and 2 of neuron group Gg of center of gravity calculation realization unit 18
As for the first one, it is excluded from the weight learning target by the pack propagation method because it merely performs separation and addition/synthesis. Therefore, there are five neuron groups that are subject to weight learning by the pack propagation method: Gb, Gc, Gd, Ge, and Gf.

On the other hand, there are six connection groups, Cab-ClH, representing the input connections of each two-neelon group, and the weight values of the input connections located in the previous stage of the neuron group to be learned are learned by the pack propagation method.

Regarding the learning operation flow shown in Fig. 2, which targets the rule section pre-wired neuron having the structure shown in Figs. 5A and 5B,
First, the initial setting of weight values for each part will be explained as follows.

[Initial setting of the antecedent membership function realizing unit] The initial setting of the weight value for the antecedent membership function realizing unit 10 is to set the weight of each connection included in the connection groups Cab and Cbc to the given membership function. Initial settings are made to achieve this.

FIG. 6 shows the antecedent membership function realization unit 10 of FIG. 5A.
The control human power X, side is taken out and shown. In FIG. 6, the output value y output from the sigmoid function unit 30-1.30-2 is V=1/(1+e-1"-")
It is expressed as (3).

In the initial setting of the weight value for the antecedent membership function realizing unit 10 of the present invention, based on the knowledge that is held in advance, for example, as the weight value W of the input connection of the sigmoid function unit 30-1, W=- 20 is set, and θ=-10 is set as the threshold value θ in the sigmoid function unit 30-1. Another sigmoid function unit 3
The weight value W of the input connection 0-2 is set to W=20, and at the same time, the threshold value θ of the sigmoid function unit 30-2 is set to θ=10. Note that the linear function unit 32-1. 32-
The weight value of the input connection No. 2 is set to the initial weight value:=1.

In this way, in the sigmoid function unit 30-1, W〈0
When the weight value and threshold value are initially set such that θ<0, the unit output value x1 (SS) for the control output X1 shown in the characteristic curve 44 in FIG. 7, that is, as the control human power X increases, the unit output value It is possible to set a truth value calculation function for a membership function having a function shape such that (SS) becomes small.

Furthermore, in the sigmoid function unit 30-2, W〉0,
By initially setting the weight value and threshold value such that θ〉0, the unit output value X 1 (L A) for the control human force x1 shown in the characteristic curve 46 in FIG. It is possible to realize a function of calculating the truth value of a membership function having a function shape in which LA) becomes large.

Sigmoid function unit 30-1.3 shown in FIG.
The initial setting of the weight value and threshold for 0-2 is such that the desired antecedent membership function is also realized for the remaining sigmoid function units 30-3 and 30-4 in the antecedent membership function realization unit 10. Initial settings are made based on prior knowledge.

[Initial Setting of Rule Section] The initial setting of the weight value for the rule section 12 in FIG. 5A is as follows.
The weight values of the connections belonging to the connection groups Ccd and Cde shown in FIG. 5B are initialized by introducing knowledge held in advance so as to realize a given fuzzy rule.

FIG. 8 is an explanatory diagram showing a specific example of initial setting of weight values for the rule section 12 together with fuzzy rules.

In FIG. 8, eight sigmoid function units 34-
1 to 34-8 correspond to fuzzy rules 1 to L8 shown on the right side, respectively. For each of the sigmoid function units 34-1 to 34-8, fuzzy rules are calculated from the linear function units 32-1 to 32-4 of the antecedent membership function realization section 10 located at the previous stage.
Input connections are made in accordance with ~L8. Furthermore, fuzzy rules are applied to the two linear function units 36-1 and 36-2 provided following the eight sigmoid function units 34-1 to 34-8, and the consequent parts of 1 to L8 (T HE
Connections are made in accordance with N). Here, the sigmoid function unit 34-1 in the rule section 12 of FIG.
The function shape of the fuzzy logic operation based on the weight value initially set to the input connection of ~34-8 and the threshold value set to the unit itself is explained as follows.

FIG. 9 shows one sigmoid function unit 34 of the rule section 12, in which truth values produces an output value y.

The output value y output from the sigmoid function unit 34 in FIG. 9 is as follows: 5=Wx,・X1+Wx1・X2−θ (4)y
It is calculated as =1/(1+e-s) (5). The unit output value y given by the above equation (5) is the weight value Wx of the input connection.

By arbitrarily setting Wx□ and the threshold value θ, an appropriate function shape can be realized.

Specifically, the sigmoid function units 34-1 to 34-1 in FIG.
The numerical value shown in the input connection of 34-8 is set as the initial weight value, and the sigmoid function unit) 34-1 to 34-8
The numerical value shown below indicates the threshold value θ. For example, in the sigmoid function unit 34-1 that realizes the fuzzy rule Ll, is set, and by setting these initial values, the sigmoid function unit 34-1 realizes the fuzzy rule L1)
34-1 performs threshold processing of the function shape shown in the three-dimensional coordinates of FIG.

Hereinafter, similarly, sigmoid function units 34-2 to 34-
By initializing the weight values and threshold values for each of the 8 fuzzy rules 2 to L8, the function shapes shown in FIGS. 11 to 17 can be obtained.

Note that the linear function unit 36 provided at the output stage in FIG.
The weight value of the connection for -1.36-2 is initially set to 1, and the threshold processing is a linear function, and θ
=0 is set.

[Initial setting of the consequent membership function realization unit] The initial setting of the weight value for the consequent membership function realization unit 14 in FIG. 5A is based on the weight value of the connection belonging to the connection group Cel in FIG. 5B. Initial settings are made to realize the given membership function based on the knowledge of the given membership function.

FIG. 18 is an explanatory diagram specifically showing the initial setting of weight values for the consequent membership function realization unit 14.

In FIG. 18, the consequent membership function realization unit 14
Nine linear function units 38-1 to 38- provided in
9, a connection input is performed from the linear function unit 36-1, 36-2 of the rule section 12 at the previous stage as shown in the figure. These nine units 38-1 to 38-9 are
They correspond to points in the range of 0 to 1 on the training line of the control output y at equal intervals. These linear function units 38-1 to 38-9
The initial setting of the weight value of the input connection for
The weight value of the connection of the output y (SS) from the unit 36-2 is calculated from the unit 38-1 side according to the characteristic curve 46 of the graph shown on the right side with the horizontal axis as the control output Y1 and the vertical axis as the unit output value. Weight value on unit 38-9 side: 1, 1. 1.0.75.0.5.0.25. 0.
0. Initialize 0.

Further, regarding the weight values of the input connections from the unit output y (LA) of the rule section 12, the weight values change from 0 to 0 from the unit 38-1 side to the unit 389 side according to the characteristic curve 48. 0.0.25. Q, 5, 0.75.
1. 1. Initialize 1.

Furthermore, the threshold value θ of linear function unit) 38-1 to 38-9 is θ
= 0.

Due to the initial setting of the weight values in the consequent membership function realization unit 14, the unit output values y (SS) and y (LA) output from the linear function unit 36-1.36-2 of the rule unit 12 are are reduced by multiplication by initial weight values, added, and output to the next center-of-gravity calculation implementation unit 18.

In FIG. 18, a weight value of 1 or less is initially set so as to reduce the unit output, but it is also possible to set a weight value of 1 or more to enlarge it.

[Initial Setting of Center of Gravity Calculation Implementation Unit] The weight values for the center of gravity calculation implementation unit 18 in FIG. 5A are initialized so as to realize center of gravity calculation for the connection group C1g in FIG. 5B.

FIG. 19 shows a specific example of initial setting of weight values for realizing the center of gravity calculation performed on the center of gravity determining element output device 40 of the center of gravity calculation implementation unit 18.

For the two linear function units 40-1 and 40-2 provided in the centroid determining element output device 40, the nine linear function units 38-1 provided in the consequent membership function realization unit 14 of the preceding stage ~389 are connected.

Then, as shown in the figure, weight values for calculating two gravity center derived values Y, , Y2 used in the gravity center calculation are set for this input connection.

In fuzzy control, a method is generally adopted in which the control output Y, which is a fuzzy inference value, is calculated by finding the center of gravity of a figure consisting of the sum of functions of reduced membership functions for the same control output position according to the following formula: do.

Y-f gtade(y)ydy / f grade
(y) dy (6) In order to realize this equation (6), as the weight value of the connection from the linear function units 38-1 to 38-9 of the consequent membership function realization unit 14 to the linear function unit 40-1. , minimum value Y=0 of control output Y
Starting from 0, weight values are assigned that increase from 0 at equal intervals up to the maximum value Y=1. On the other hand, linear function unit 40-
For input connection No. 2, a weight value that decreases from 0 to -1, for example, from 0 to -1 at the same equal intervals as above, is set from the maximum value Y = 1 of the control output Y to the minimum value Y = 0. Assign.

By weighting the input connections to the linear function unit 40-1 and 40-2, if the unit output values of the linear function units 38-1 to 38-9 in the previous stage are 01 to C, the linear function unit 40-1 is, Y 1 = I C, +0.875 C2+0.75C
300, 625 C4+0.5 C5+0.375 C
6+0.25C7+0.125 CB+OC9(
7), and the linear function unit 40-2 outputs Yl
"OCI O, 125C20, 25C3-QJ75
C4-θ, 5 c, -0,625C6-0,75C,
-0,875C8+IC1(8) is output.

The centroid derived values Y, Y2 output from the linear function unit 40-1, 40-2 are given to the centroid calculation device 42, and Y=Y2/(Y2-Y+) (9)
using the control output Y, which is the center of gravity, that is, the fuzzy inference value
is calculated.

By substituting the above equations (7) and (8) into the above equation (9), it can be seen that Y in the above equation (9) matches Y in the above equation (6).

With the above processing, the initial setting of the weight value for the rule section prewire neuron shown in steps S1 to S4 in FIG. 2 is completed, and the process proceeds to step S5.

[Setting of learning schedule] In step S5 of FIG. 2, following the completion of the initial settings, a learning plan, ie, a learning schedule, for causing the rule section pre-wired neuro to perform learning processing is set. This learning schedule is set by setting the phase/group correspondence table shown in FIG. 20(a) and setting the group/connection correspondence table shown in FIG. 20(b).

First, the phase-group correspondence table sets neuron groups to be learned in each phase as the learning phase progresses. In the present invention, after setting the initial weight values, the rule unit 12 and the antecedent membership function realizing unit 10 perform weight learning in this order, so that the phase 1 In phase 2, the neuron groups Gd and Ge belonging to the rule unit 12 are set as learning target groups, and in phase 2, the antecedent membership function realization unit 10 is set.
The neuron groups Gb and Gc belonging to are set as learning target groups.

On the other hand, the group/connection correspondence table shows the correspondence between each of the two knee long groups Ga to Gg in the rule section prewire neuron and the input connection group Cab-C1g in Figure 20 (b
).

When the setting of the phase/group correspondence table and the group/connection correspondence table is completed, the process proceeds to step S6 in FIG. 2, starts the learning scheduler, and actually gives learning data to the rule section prewire neuron. Weight learning is performed using the pack propagation method.

[Processing and Configuration of Learning Scheduler] FIG. 21 shows a flowchart of the learning process of the rule section pre-wire neuro of the present invention. This processing flow diagram is, for example, the 22nd
First, the device configuration for learning processing that can be realized by the device configuration shown in the figure will be explained.
In the figure, a learning instruction unit 50 and a learning unit 70 are provided for the rule section pre-wired neuron 100. The learning instruction unit 50 has a learning scheduler 52, and a learning signal storage section 54, shown in FIG. 20, stores a learning signal consisting of a control input and a desired control output corresponding to the control input. A phase/group correspondence tape 56 storing a phase/group correspondence table, a group/connection correspondence table 58 storing a group/connection correspondence table, and a learning convergence determination section 60 are provided.

On the other hand, the learning unit 70 is provided with a learning instruction reading section 72, a learning flag setting section 74, and a weight changing section 76 that changes weight values using the pack propagation method.

Furthermore, a feature of the device configuration in FIG. 22 is that a learning adjuster 80 is provided in the connection connecting each unit in the rule section prewire neuron 100. The learning adjuster 80 has the configuration shown in FIG. Note that the weight change unit 76 is shown together with the learning adjuster 80. Learning regulator 8
0 is provided with a flag storage section 82, a weight change information reading section 84, and a weight change amount adjustment section 86.

On the other hand, the weight change unit 76 includes a weight calculation unit 88 that performs weight calculation based on the pack propagation method, and a weight storage unit 9.
0 and a weight change amount storage section 92 are provided.

In this way, by providing the learning adjuster 80 for each connection of the rule section pre-wire neuron 100, it is possible to determine whether the learning scheduler 52 of the learning instruction unit 50 changes the weight value of the connection for the re-rule section pre-wire neuro 100. It is possible to set a learning flag that determines whether the learning flag is valid, that is, it is possible to perform weight learning only for connections for which the flag is turned on.

In addition, the flag set for the learning coordinator 80 sets the antecedent membership that can execute fuzzy rules for a general hierarchical network that does not have a network structure that can implement fuzzy rules as shown in FIG. It is also used when specifying a hierarchical network structure consisting of the function realization section 10, the rule section 12, and the consequent membership function realization section 14.

When a hierarchical network conforming to fuzzy rules is configured, the weight values are adjusted as follows when actually learning is performed by the pack propagation method.

In FIG. 23, the weight change information reading unit 84 of the learning adjuster 80 monitors whether the weight change amount has been written by the weight calculation unit 88 to the weight change amount storage unit 92 of the weight change unit 76. There is. When the weight change information reading unit 84 detects that the weight has changed, it notifies the weight change amount adjustment unit 86 of the fact. The weight change amount adjustment unit 86 checks the learning flag in the flag storage unit 82, and does nothing if the learning flag is on. That is, writing to the weight change amount storage section 92 is made valid. On the other hand, if the learning flag is off, the weight change amount adjustment section 86 changes the weight change amount storage section 9 of the weight change section 76.
2 is set to zero to disable the weight change. The hardware configuration of this learning scheduler is disclosed in Japanese Patent Application No. 63-2278.
No. 25, “Learning method for network configuration data processing device”.

Next, with reference to the learning processing flow in FIG. 21, Section 22.
The learning process of the rule section pre-wire neuron according to the device configuration shown in FIG. 23 will be explained.

First, when the learning scheduler 52 of the learning instruction unit 50 is activated, the first step (hereinafter "step" is omitted) is to
At SL, a phase counter i indicating the progress of the learning phase is set to i=1, and the process proceeds to S2, where the neuron group number corresponding to the learning phase i=l is read out with reference to the phase/group correspondence table 56.

Next, the process advances to S3, and the connection group number corresponding to the neuron group number is read out with reference to the group/connection correspondence table 58.

Next, the process proceeds to S4, where the connection group number is output and the learning flag of the learning adjuster 80 provided to the connection belonging to the connection group number is turned on.

Specifically, the learning scheduler 52 of the learning instruction unit 50 outputs the connection group number to the learning instruction reading section 72 of the learning unit 70 and causes it to read it. The learning flag setting unit 74 that receives the reading result from the learning support reading unit 72 instructs the flag storage unit 82 of the learning adjuster 80 provided in the connection belonging to the connection group number for which the learning instruction has been received to turn on the flag. and set the learning flag.

Next, the process proceeds to S5, where the learning scheduler 52 issues a learning execution command to start the learning process. This learning process reads out the control force X prepared in the learning signal storage unit 54 and the teacher signal d which is a desired control output for the control force
0, and the teacher signal d is given to the weight changing unit 76 and the learning convergence determining unit 60. Then, the control output Y output from the rule section prewire neuro 100 that receives the learning human power X
is taken into the weight change unit 76 and learning convergence determination unit 60,
The weight changing unit 76 changes the weight values according to the pack propagation method, and the learning convergence determining unit 60 determines the end of learning when the error of the control output Y with respect to the teacher signal d becomes equal to or less than a specified value. Finish the process.

When the learning process ends based on the learning execution command in S5,
Increment the phase counter i by one in S6, and
Check the end of the learning phase at 7. Since the learning process of the present invention ends at phase 1, the series of processes ends as is.

[Principle of learning by pack propagation method 1 Next, the principle of weight learning by the pack propagation method of the present invention will be explained together with the hierarchical network.

FIG. 24 shows a general hierarchical neural network, which is composed of a type of node called uni-soto and internal connections with weights.

In Figure 24, the h layer is the input device, and the i layer is the intermediate layer.
layer, and the j layer as the output layer. Of course, there is only one i-layer as an intermediate layer, but even if there are multiple intermediate layers, they have the same name. Each unit has the configuration shown in FIG. 4, and the calculations performed in the unit are as shown in equations (1) and (2) above for the basic unit 20. In the pack propagation method, the weight value Wih and threshold value θi of the hierarchical network are adaptively and automatically adjusted and learned using error feedback. As is clear from equations (1) and (2) above, it is necessary to adjust the weight value Wih and the threshold value θi at the same time, but this work is a difficult work that interferes with each other.

Therefore, in addition to the normal unit, the applicant of this application has
A unit whose output value is always 1 is provided in the h layer on the input side,
Threshold value θi for input connection from this unit with output value 1
By assigning as a weight value, the threshold θi becomes the weight value W
A pack propagation method has been proposed in which the threshold value θi can be treated as a weight value by incorporating it into i h (Japanese Patent Application No. 62-333484).

By doing this, the above equations (1) and (2) become x pl
=Σ'J phW + h (I
f) y,, =: 1/ (1+eXp(Xpl))
(12), and the threshold value θi does not appear.

By expanding from equations (11) and (12) above, we get the following (13)
) (14) is obtained.

Xpl”ΣypIWth (13)y
,,= 1/ (1+exp(xp+) ) (
14) However, j: unit number W of the j layer, , : weight value x of internal connection between the i-j layers, , : sum of products Vp+'l of input from each unit of the i layer to the 1st unit of the j layer 1 of layer j for the input signal of )th pattern
The weight changing section 76 in FIG. 22 outputs the input cover turn X for learning.
When receiving the teacher pattern dpi, which is the signal that should be taken by the output cover turn Y+++ and the output cover turn "l+++" output from the output layer j when the output layer j is presented, the difference value (apt-yp+) between the two is calculated, and then δp+ =Yp+ (13'p+) (apt !-
+) Calculate (15). Subsequently, ΔW+I(tl =εΣδ,, y, l×ζΔW, (+
-+) However, ε: Learning constant ζ: Momentum t: The update amount ΔWI(1) of the weight value between the i layer and the j layer is calculated according to the number of learning times. Here, the reason for adding the weight value update amount determined in the previous update cycle in the second term on the right side of equation (I6) is to speed up learning.

Subsequently, the weight value changing unit 76 uses Δδ, 1 calculated by the above equation (15) to calculate δ−+=V pl(1”j pi)Σδo+WI+(
t-1) (17), then Δw+h(+) =εΣδp+Vph+ζΔwlh(
+-t), update amount Δw+h of weight values of h layer and i layer
Calculate (+).

Subsequently, the weight change unit 76 determines the weight value for the next update cycle according to the update amounts calculated by the equations (16) and (18). To update the above weight values, the output pattern y- for the input pattern pattern for learning is the teacher pattern apl.
The learning process is repeated so that a weight value that matches is obtained.

On the other hand, if the units constituting the hierarchical network in FIG. 24 are linear function units, the above equation (14) becomes V pi = X p
(20). The above equation (15) and (17) in this linear function unit
) formula is δ,, = d pi V pl
(21) δ, 1 = copy δ−+Wx (t 1)
It appears as (22).

[Weight Learning Process of the Present Invention] FIG. 25 is a process flow diagram of weight learning performed by the back propagation method for the rule section pre-wired neuron of the present invention shown in FIG. 5A.

The parameters of each part in this learning process flow are the 26th
It is defined as shown in the figure.

FIG. 26 schematically shows the hierarchical structure of the rule section pre-wire neuron in FIG. 5A. As shown, it has a six-layer structure from the first layer to the sixth layer.
The layer is made up of sigmoid function units, and the rest are made up of linear function units.

In addition, the fuzzy inference value output from the pre-wire neuron in the rule section, that is, the control output value, is y6, and the learning process in one phase is performed from the 6th layer to the 5th layer, the 4th layer, the 3rd layer, and the 2nd layer. This is done sequentially such as layer, first layer, and so on. The progress of learning for each layer is indicated by an i counter. That is, the i counter is set in the i layer 6 in the initial state. Furthermore,
In phase 1 in which the weight learning of the rule section 12 is performed, phase 1 ends when the learning of the third layer located before the rule section 12 is completed. In addition, in phase 2 of learning the weights of the antecedent membership function realization unit 10, the first
The process ends when the learning of the layer is completed.

Furthermore, to simplify the explanation, wl, l-1 and ΔW,
,, -1 indicates a matrix, and its size is (number of neurons in the i-th layer) x (number of neurons in the i-1th layer). Furthermore, δi and yi represent vectors whose size is the same as the number of neurons in the i-th layer.

Based on the parameters of each part shown in FIG.
The weight learning processing flow shown in the figure can be explained by dividing it into phase 1 and phase 2 as follows.

(1) Phase 1 In phase 1, the weights of the input connections of the neuron groups Gd and Ge in the third and fourth layers belonging to the rule section 12 are learned.

First, a counter i for setting a learning target layer is initialized using Sl. Specifically, since the device has a six-layer structure as shown in FIG. 26, the counter i is initially set to the i-layer 6. With this initial setting, the sixth layer is designated as the learning target.

Next, the process proceeds to 82, where it is determined whether the unit is a sigmoid function unit. Since the sixth layer is a linear function unit, the process proceeds to 814, where it is determined whether it is the final layer. Since the sixth layer is the final stage, the process proceeds to 815, and uses the teacher signal d and control output y6 obtained at that time to
) The difference value δ6 is determined according to a formula similar to the formula.

Next, the process proceeds to 86, where it is determined whether the unit is to learn weights. At this time, the 6th layer is excluded from the learning target, so
Specifically, since the learning flag of the learning adjuster 80 provided in the connection is off, the process proceeds to S8 and the weight update amount Δ
W6. is set to ΔWb, "=O, and the process proceeds to S9, and the weight value is not updated. Next, the process proceeds to S12, where it is determined whether the counter i is the final learning layer E. In phase 1, the final learning layer is the third learning layer. Since i=6 at the end of the sixth layer, the process proceeds to S13, where the counter i is decremented by one to make i=5, and the process returns to S2 to start learning the next fifth layer.

Regarding the fifth layer, since it is a linear function unit, S2
The process proceeds to 814, but since it is not the final stage layer, the process proceeds to 816.
Then, the difference value δ is calculated according to a formula similar to the formula (22) above.
, find.

Next, proceed to 86 to determine whether it is a learning target,
Since the fifth layer belonging to the consequent membership function realizing unit 14 is not a learning target, the weight value W54 is not updated by going through S9 88. Thereafter, similar processing is repeated until the final unit of the fifth layer is reached.

When the fifth layer processing is completed, the process proceeds to the fourth layer.

The fourth layer uses a sigmoid function unit, and since it is in the rule section 12, it is designated as a learning target. Therefore, from S2 to S14, in S16, the difference value δ
4 is calculated, the process proceeds from 36 to 87, where the weight update amount ΔW43 is calculated using the difference value δ3 calculated in S5 according to formulas similar to the formulas (16) and (18), and in S9, the weight update amount ΔW43 is calculated as described above (
The weight value W43 is updated according to a formula similar to formula 19).

Thereafter, the same process is repeated until proceeding to the final unit of the fourth layer, and when the process of the fourth layer is completed, the process proceeds to the third layer. Third
The layer is a sigmoid function unit, and the rule part 12
It is a subject of study because it belongs to . Therefore 3
2. Proceeding to S5 via S3, the difference value δ3 specific to the sigmoid function unit is calculated using the same calculation formula as the above-mentioned formula (17), and the process proceeds from S6 to 87, where the difference value δ calculated in S5 is calculated.
3 is used to calculate the weight update amount ΔW3□, and the weight value W, 2 is updated in S9. Thereafter, similar processing is repeated for the third layer until the final unit value is obtained in S10.

When the final unit of the third layer is determined in 810, the process proceeds to step S12.
At the final layer E = 3, the value of the i counter indicates the end of the phase.
Therefore, the series of learning processing in phase 1 ends.

(2) Phase 2 In Phase 2, the weight values of the connection group numbers Cab and Cbc of the neuron groups Gb and Gc in the first and second layers included in the antecedent membership function realizing unit 10 are learned. As in the case of Phase 1, this learning process is performed sequentially from the sixth layer of the final layer in FIG. 26 to the previous layer.

Since the 6th to 3rd layers are excluded from the learning target, the difference value δi at 84, 35, 815, and 816 is calculated, but the weight update amount ΔW at S7 is not calculated, and ΔW = 0 and continue processing.

Then, when proceeding to the processing of the second layer, since the second layer is a linear function unit, the difference value δ obtained in 816
2 is used to determine the weight update amount in S7, and the weight value is updated in S9.

When the weight learning of the second layer is completed and the process proceeds to the first layer, the difference value δ is calculated in S5 since the first layer is a sigmoid function unit, the weight update amount is calculated in S7, and the amount of weight update is calculated in S9. Update the weight values.

When the processing of the last unit of the first layer is completed, 810 to 8
The process proceeds to step 12, and since the ending hierarchy E in S12 is set to E=1 in phase 2, i=E, and the process of phase 2 ends.

[Initial Setting of Weights Using Random Numbers] Next, another embodiment of the learning method of the present invention will be described with reference to the learning operation flow diagram of FIG. 27.

The learning process shown in FIG. 27 is characterized in that the weight values are randomly initialized using random numbers as the initial setting of the weight values for the rule section 12 in step S2. Furthermore, if there is some knowledge that is previously held, settings may be made based on this knowledge.

In this case, the initial setting of the weight values for the antecedent membership function realization unit 10 and the consequent membership function realization unit 14 is based on the introduction of pre-existing knowledge, similar to the first embodiment in FIG. is being carried out. Other steps S
4. S5. The processing at S6 is the same as the first embodiment shown in FIG.

[Others] Note that the above embodiment takes as an example the rule section pre-wired neuro that implements fuzzy rules targeting two control inputs X, , and X2 as shown in FIG. 5A. Of course, the number of control inputs may be one or three or more.

In addition, in the above embodiment, y=1/(1+e−wx”)(23
) was taken as an example of the membership function, but
Two sigmoid function units 9 as shown in FIG.
The unit output of 4-1.94-2 may be applied to the subtraction unit 96 to realize the sigmoid function shown in FIG. 4(b). In this case, the unit output y is y = 1/(1-e −”x-” ) −1/(]
−e −w2x′”).

Furthermore, since the consequent membership function realization unit 14 in FIG. 5A sets the initial value fixedly and does not perform weight learning, it may be configured with ordinary hardware instead of pre-wired neurons. .

[Effects of the Invention] As explained above, according to the present invention, the initial setting of the weight value of the rule part is performed by introducing pre-existing knowledge or using random numbers, and The initial setting of the weight values of the subject membership function realization part is done based on the introduction of the knowledge that is held in advance, and after the initial setting of the weight values, the weight learning of the rule part is performed first, and then the antecedent part membership function realization is performed. By learning the weights of the rule part and the previous part, while making use of the weights of the antecedent part membership function realization part and the consequent part membership function realization part, which have been initialized by introducing knowledge that they have in advance, It is possible to efficiently perform learning to adjust each weight of the subject membership function realization part to the optimal value suitable for the system, and when realizing an arbitrary control system using fuzzy rules,
A more efficient connection weight analysis can be achieved.

In addition, the initial setting of weights that incorporates pre-existing knowledge facilitates learning and improves learning efficiency.

Furthermore, by initializing the weights of the rule section using random numbers or by using a combination of pre-existing knowledge and random numbers, it is possible to maximize the neuro-based learning function.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of the principle of the present invention; Fig. 2 is a learning operation flow diagram of the present invention; Fig. 3 is a configuration diagram of an embodiment of the rule section pre-wired neuron of the present invention; Fig. 4 is an explanatory diagram of the basic unit; FIG. 5A is a specific example configuration diagram of the rule section prewire neuron of the present invention; FIG. 5B is an explanatory diagram of the neuron group and connection group of FIG. 5A; FIG. 6 is an antecedent section membership of the present invention. FIG. 7 is an explanatory diagram of the membership function of the antecedent part; FIG. 8 is an explanatory diagram of the initial setting of the rule section of the present invention; FIG. 9 is an explanatory diagram of the rule section unit of the present invention; Figure 10 is an explanatory diagram of fuzzy rule L1 in Figure 9; Figure 11 is an explanatory diagram of fuzzy rule L2 in Figure 9.
Explanatory diagram; FIG. 12 is an explanatory diagram of fuzzy rule L3 in FIG. 9; FIG. 13 is an explanatory diagram of fuzzy rule L4 in FIG. 9;
4 is an explanatory diagram of fuzzy rule L5 in FIG. 9; FIG. 15 is an explanatory diagram of fuzzy rule L6 in FIG. 9; FIG. 16 is an explanatory diagram of fuzzy rule L7 in FIG. 9; An explanatory diagram of rule L8; FIG. 18 is an explanatory diagram of the initial settings of the consequent membership function implementation unit of the present invention; FIG. 19 is an explanatory diagram of the initial settings of the center of gravity calculation implementation unit of the invention; An explanatory diagram of the settings of the phase/group correspondence table and the group/connection correspondence table; Fig. 21 is a flowchart of the learning process of the present invention; Fig. 22 is a diagram of the device configuration of the learning process of the present invention; Fig. 23 is the same as that of Fig. 22. A configuration diagram of the learning adjustment station provided for each connection; Figure 24 is an explanatory diagram of a general hierarchical neural network; Figure 25 is a flow diagram of the weight learning process of the present invention; Figure 26 is an explanation of the parameters of each part in Figure 25. FIG. 27 is an operational diagram showing another embodiment of the present invention; FIG. 28 is an explanatory diagram of another membership link used in the present invention. In the figure, 10: 12 = 14: 16: 18: Antecedent part Membership function realization part Rule part Consequent part Membership function realization part Input part Centroid calculation realization part 20: Basic unit 22: Multiplication part 24: Addition part 26 :Threshold value processing unit 28-1.28-2.32-1 to 32-4.38-]
~38-'1.4L1.40-2 Linear Function Units 30-1 ~30-2.34-1 ~34-8.94-19
4-2: Sigmoid function unit 40 = center of gravity determining element output device 42: center of gravity calculation device 50: learning instruction unit 52: learning scheduler 54: learning signal storage section 56: phase/group correspondence table 58 niger loop/phase correspondence table 60: learning Convergence determination section 70: Learning unit 72: Learning instruction reading section 74: Learning flag setting section 76: Weight changing section 80: Learning adjuster 82/flag storage section 84: Weight change information reading section 86: Weight change amount adjustment section 88: Weight calculation section 90: Weight storage section 92: Weight change amount storage section 96: Subtraction unit 100: Rule section pre-wire neuro

Claims (6)

[Claims] (1) It has a hierarchical network consisting of an antecedent part membership function realization part (10), a rule part (12) consisting of one or more layers, and a consequent part membership function realization part (14), and the rule part (12) does not internally connect all units with the antecedent membership function realization unit (10) in the previous stage and/or the consequent membership function realization unit (14) in the subsequent stage, but only some parts according to the control rule. In a learning method of a rule section pre-wired neuro that is internally coupled and outputs one or more control operation quantities (Y) corresponding to input control state quantities (X_1, X_2,...X_n), The antecedent membership function realization unit (10), the rule unit (12), and the consequent membership function realization unit (14)
a first step of setting initial weight values for each of the; a second step of learning only the weights of the rule section (12) using the learning data; and learning using the learning data after the second step is completed. A method for learning a rule section pre-wire neuro, comprising: a third step of learning only the weight of the antecedent membership function section (10) using data. (2) The setting of the initial weight values in the first step is carried out by using the initial weight values created based on knowledge held in advance by the antecedent membership function realization unit (10), the rule unit (12),
and the consequent membership function realization unit (14), respectively. (3) To set the initial weight values in the first process, initial weight values created using random numbers are set in the rule section (12), and the antecedent membership function realization section (10) and the post 2. The method for learning a rule section prewire neuron according to claim 1, wherein initial weight values created based on knowledge held in advance are set in the subject membership function realization section (14). (4) In setting the initial weight values of the rule section (12) in the first step, weights that can be set based on knowledge held in advance are initialized using this knowledge, and other weights are initialized using random numbers. A learning method for a rule section pre-wired neuron according to claim 1. (5) At the start of learning in the second process, the rule section (1
In addition to turning on the learning flag provided for each connection in 2), the learning flag provided for each connection of the antecedent membership function realization unit (10) and the consequent membership function realization unit (14) is turned on. 2. A learning method for a rule section pre-wired neuron according to claim 1, wherein only the weight values of connections having learning flags that are turned off and turned on are optimized through learning processing. (6) At the start of learning in the third step, turn on the learning flag provided for each connection of the antecedent membership function realization unit (10), and set the rule unit (12) and consequent membership The rule according to claim 1, characterized in that a learning flag provided for each connection of the function realization unit (14) is turned off, and only the weight value of a connection having a learning flag in an on state is optimized by a learning process. How to learn part prewire neuro.