JPH0460702A - Fuzzy control system - Google Patents

Abstract

PURPOSE:To execute fuzzy inference by means of plural characteristics without being restricted by means of a storage part by obtaining the position of a center of gravity in plural membership functions obtained by means of substituting the weighting value of the membership functions for the membership function and obtaining output corresponding to the center of gravity. CONSTITUTION:Data corresponding to the air-conditioning operation/heating operation modes of an air-conditioning unit 2 are read from storage parts 7-10. Then, a difference (e) between a room temperature T and a room temperature S which is set in a room setting unit 4 and the value of a change quantity DELTAe between a difference e' which is obtained last time and the difference (e) which is obtained this time are inputted and the realizing degree of the membership function is calculated. The membership function of a precedent part is weighted. The obtained weighting value of the membership function in the precedent part is set to be the constant B of the membership function and respective membership functions are composed by using constants A and C. The center of gravity is calculated by using the constants A, B (weighting value) and C of the membership functions. Then, a value is outputted to an invertor device 11 as the alteration quantity of the capacity of the air- conditioning unit 2, which corresponds to the center of gravity.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Application The present invention relates to a fuzzy control device used to control the operation of equipment so that a detected value becomes a target value.

(b) Conventional technology In general, as a conventional fuzzy control device, Japanese Patent Application Laid-Open No. 1-14
There was something that was completely described in the 2902 publication. What is described in this publication is 'A fuzzy control method for an automobile that is equipped with a fuzzy control section that performs fuzzy control of a controlled object such as a propulsion system of an automobile by a processor installed in the automobile. and a readout control unit of a storage control device, in which the storage device calculates the detected amount and temporal change in the controlled target portion in advance according to a predetermined rule, stores the results as a table, and stores the results in the form of a table. The read control section of the device reads out the table at a predetermined time and controls the controlled section.

By configuring the fuzzy control folds in this way,
When a processor is given a detected value, it can perform fuzzy control simply by reading out the value corresponding to the detected value from a table. That is, the processor does not need to perform fuzzy inference, which takes processing time, on each detected value, and has the advantage of being able to shorten the processing time that the processor spends on fuzzy control.

(C) Problems to be Solved by the Invention In the conventional technology configured as described above, it is possible to shorten the processing time of the processor by using a table, but in such a conventional technology, the processing time of the processor can be shortened by using a table. When fine control with resolution Jla < is required or when it is necessary to widen the input/output range of fuzzy control, it is necessary to increase the size of the table. For example, in order to double the output resolution or double the output range, the table must be set four times as large. This is not suitable for controlling general equipment due to the limited capacity of the storage unit, and has the problem that fuzzy inference based on a plurality of characteristics cannot be controlled accurately over a wide range.

To address these problems, the present invention provides a fuzzy control device that can perform fuzzy inference based on a plurality of characteristics without being limited by the storage unit.

(d) Means for Solving the Problems The fuzzy control method of the present invention uses a first
a second storage unit storing constants of a function that determines the membership function of the consequent to the control output value;
and a weighting section that weights the membership function of the consequent part based on the degree of establishment of the membership function of the antecedent part, and the membership function of the antecedent part is determined by assigning the detected value to the membership function of the antecedent part. The degree of validity of the function is calculated, the weighted value for the membership function of the consequent part is calculated from the calculated value of the degree of validity in the weighting part, and this weighted value is substituted for the membership function of the consequent part. The position of the center of gravity of the membership function is obtained, and an output corresponding to the center of gravity is obtained, and the output is stored in the first storage section and/or the second storage section.

(f) Effect When using the fuzzy control method configured in this way, the characteristics of the fuzzy inference can be changed by selecting and using the constants stored in the first and second storage units. The characteristics of the fuzzy inference can be changed depending on the operating condition of the object, for example, when control is started and when it is steady, and the control characteristics can be constantly improved.

(f) Examples Examples of the present invention will now be described based on the drawings. FIG. 1 is a schematic diagram when a fuzzy control device is applied to temperature control of a conditioned room. In this figure, 1 is a temperature-controlled room, and 2 is an air conditioning unit installed in this room. Cooling air or heated air is discharged from the outlet to cool or heat the room. This is what we do. This air conditioning unit can change its cooling capacity or heating capacity by using an inverter device, which will be described later. 3 is a temperature sensor attached to be able to detect the room temperature of the conditioned room 1, and uses a detection element such as a thermistor whose internal resistance value changes depending on the temperature. Reference numeral 4 denotes a room temperature setting device, which outputs a signal corresponding to the set temperature. There are various types of room temperature setting devices, such as those that use a variable resistor to change the resistance value, those that use a voltage generator to change the output voltage, and those that output a code or signal corresponding to the set value. Can be used. 5 is a detected value output section, which calculates the corresponding room temperature T from the change in the internal resistance value of the temperature sensor 3, and calculates the corresponding room temperature T.
and the room temperature S set in the room temperature setting device 4 “e=T−5
” (When performing heating operation, calculate “e--(
T-8)". Also, calculate the change "Δe=e'-e" between the difference e゛ found last time and the difference e found this time, and calculate these differences C and Δe in the microcomputer. 6. This microcomputer 6 calculates the output using the differences e and Δe and outputs it to the inverter device 11. This inverter device 11 operates the air conditioning unit based on the signal given from the microcomputer 6. For example, if a compressor using an induction motor as a drive source is used in the air conditioning unit 2, the operating capacity of the air conditioning unit 2 is controlled by changing the frequency of AC power supplied to the induction motor. The driving ability of the driver can be changed.

Alternatively, if a compressor using a DC motor as a drive source is used, the operating capacity of the air conditioning unit 2 can be changed by changing the voltage applied to the DC power supplied to the DC motor. Note that 7 to 10 are storage units, and in this embodiment, one memory element is divided into a plurality of areas, which are distributed to the storage units 7 to 10.

The operation of the microcomputer 6 will be explained below. The microcomputer 6 performs fuzzy inference using the C and ΔC. Figure 2 shows an example of a membership function (isosceles triangle), and this membership function is set as a function when the values of A and C on the horizontal axis and the maximum degree of establishment B on the vertical axis are given. be able to. That is, (1) when e<A, the degree of validity of the men/<-ship function is 0, (2) when A≦e<(A + C)
/ 2, the degree to which the membership function holds is (2BX
(e-A))/(C-A), (3)(A+C)/2≦e
In C, the degree of membership function is -(2BX
(e−C))/(C−A), (4) When C≦e, the degree to which the membership function is satisfied is O.

In this way, if the constants A, B, and C shown in FIG. 2 can be set, this membership function can be specifically expressed. Therefore, if the value of e is obtained, the degree to which the membership function holds can be determined. Note that membership functions are not limited to isosceles triangles, and can be arbitrarily set based on the object to be controlled and the five senses of the user, such as equilateral triangles, trapezoids, bell shapes, and exponential functions, but shapes that are easy to convert into functions are preferable.

FIG. 3 shows the membership function of the antecedent with respect to the difference e between the room temperature and the set value in A mode. For each membership function NB, NS, ZO, PS, FB, constants corresponding to A, B, and C in Fig. 2 are set,
A function representing each membership function is set from these constants. A list of constants representing these functions is shown in FIG. Furthermore, the membership function ZO is the difference e
The membership function PS represents the set for which the difference e is felt to be a little on the positive side, and the membership function PB represents the set for which the difference C is felt to be large on the positive side. The function NS represents a set where the difference e is felt to be slightly on the negative side, and the membership function NB represents a set where the difference e is felt to be large on the negative side. FIG. 3 shows the membership function of the antecedent part in B mode, and like the membership function shown in FIG. 3, a list of constants is shown in FIG.

Therefore, the degree to which each of the membership functions of the antecedent part holds true for the difference e can be calculated from the above equation using the A-mode or B-mode constant for each function of the membership function NB-FB.

Furthermore, the membership function of the antecedent part can be similarly set for the rate of change ΔC of the difference e.

In this case, the constant of the function representing the membership function may be changed between the A mode and the B mode, or an intermittent constant may be used. Generally, when setting a large variation range of Δe, it is preferable to change the constant between A mode 1 and B mode.
When the range of change in e is small, the same constant can be used, but it is not limited to this.The membership function for Δe is also the membership function for class 511 up to NB-FB in the same way as the membership function for e. Set.

Figure 6 is a relationship diagram of tuning that weights the membership function of the consequent (rules used in fuzzy inference).
The membership function of the consequent is weighted based on the degree to which the membership function holds true for e and the degree that the membership function holds true for Δe. Seven membership functions are set for the consequent part, from functions NB to PB. The membership function zO represents a set that does not change the operating capacity of the air conditioning unit 2 shown in FIG. 1, the membership function PS represents a set that slightly increases the operating capacity of the air conditioning unit 2, and the membership function PM Membership function FB represents a set that increases the operating capacity of air conditioning unit 2 to a medium degree, membership function NS represents a set that increases the operating capacity of air conditioning unit 2 to a large extent, and membership function NS represents a set that slightly decreases the operating capacity of air conditioning unit 2. , the membership function NM represents a set that reduces the operating capacity of the air conditioning unit 2 to a medium level,
The membership function NB represents a set that significantly reduces the operating capacity of the air conditioning unit 2. Incidentally, the horizontal axis of these membership functions sets the increase/decrease in the operating capacity of the air conditioning unit 2. The following 17 rules are set as fuzzy inference rules.

(1) If the membership function NB regarding e holds true and the membership function NB regarding Δe holds true, then the membership function NB of the consequent part holds true.

(2) If the membership function NS regarding e holds true and the membership function NS regarding Δe holds true, then the number NM of memberships in the consequent part holds true.

(17) If the membership function PB regarding e holds true and the membership function PB regarding Δe holds true, then the membership function FB of the consequent part holds true.

FIG. 7 is a diagram showing the membership function of the consequent part.

In this figure, the horizontal axis is the output, that is, the increase/decrease value of the operating capacity of the air conditioning unit 2. The function representing the membership function NB-PB of the consequent part is also determined in the same way as the membership function of the antecedent part, once the three constants are determined. 7th
A of this constant from the membership function of the consequent part shown in the figure.

C has been set in advance. Similar to the membership function of the antecedent part, these constants may also be stored in two types of constants, A mode and B mode, and may be selected and used according to the operating state of the air conditioning unit. Incidentally, the constant B represents a weight value and is a value that can be set according to the above rules (1> to (17)).

Next, the main operations will be explained using the flowchart shown in FIG. First, fuzzy inference is started from step So.

This fuzzy reasoning is used as part of the control of the air conditioning unit.
This is performed at predetermined intervals (about once every 30 seconds). Therefore, the operating capacity of the air conditioning unit is changed at predetermined intervals. Next, in step S1, it is determined whether to use the A-mode data or the B-mode data in FIG. 5. The distinction between mode A and mode B may be arbitrarily changed depending on various operating conditions, such as cooling operation/heating operation of the air conditioning unit, at the start of operation/at steady operation, or depending on the season. It goes without saying that the number of modes may be two or more. Next, in steps S2 and S3, data corresponding to the mode is read out from the storage section6.Next, in step S4, e, Δ
Input the value of e and calculate the degree of establishment of each member zip function in step S5.0For example, in A mode, e=
If el, Δe = Δc1, then by substituting el and ΔC1 into the membership function of the antecedent part (a function using A-mode data), respectively, regarding e, the membership function zO is obtained as shown in FIG. The degree of establishment of is E2,
The degree to which the membership function PS holds true is El, and the degree to which the membership functions NB, NS, and PB hold true are 0. Regarding Δe, as shown in FIG. 11, the degree to which the membership function zO is satisfied is ΔE2, the degree to which the membership function PS is valid is ΔE1, and the degree to which the membership functions NB, NS, and PB are satisfied is 0.

Next, in step S6, the membership function NB of the consequent part
- The membership functions for which the antecedent part holds true from FIGS. 10 and 11, which weight FB, are the membership function zO and the membership function PS for e, and the membership function zO for ΔC. and membership function PS. Therefore, the partial rule can be applied to the shaded area as shown in FIG. 12. This rule is the rule explained using FIG. That is, in the consequent part, membership function zO, membership function PS, and membership function PM are established. The respective weighting values are E2XΔE2, EIXΔE2(EIXΔE2
<Adopting the value of EIXΔE2 for E2XΔE1), EI
XΔE1.

In this example, the degree of establishment of the membership function of the antecedent part is represented by byte data of 0-FF (hexadecimal), and the weighting of the membership function of the consequent part is expressed as 0-FFFF (
Although the data is handled as word data in (hexadecimal) to facilitate calculations on the microcomputer 6, the degree of establishment and weighting may be converted to values between 0 and 1. At this time, a constant B stored in the storage section is also set to a value between 0 and 1.

In this way, the weighting value of the membership function of the consequent part obtained is set as the constant B of this membership function, and the respective membership functions are synthesized using constants A and C shown in Fig. 8. A composite figure as shown in FIG. 13 is obtained.

Next, in step S7, the center of gravity of the composite figure shown in FIG. 13 is calculated. The center of gravity is the constant A of the membership function before composition.
, B (weighting is a value), and C. This calculated center of gravity is G. Next, in step S8, a value Δf1 on the horizontal axis corresponding to this center of gravity is obtained. This Δf1 is output to the inverter unit [11] as the amount of change in the capacity of the air conditioning unit. When the frequency of the AC power source that supplies the operating capacity of the air conditioning unit 2 is changed, Δf1 corresponds to the increase/decrease value of the frequency.

As described above, the present invention stores the constants of the membership functions of the antecedent part and/or the consequent part in the storage unit, and when determining the degree of establishment of the membership function of the antecedent part, the membership function of the consequent part Since this constant is used for calculation when determining the center of gravity position of a composite figure by a function, the characteristics of fuzzy inference can be easily changed by changing the value of the constant stored in the memory unit. . Therefore,
If a plurality of combinations of these constants are set and stored in the storage unit, fuzzy control that is always suitable for the operating state can be performed by changing the constants depending on the operating state of the object to be controlled.

Note that when an exponential function is used as the membership function, a more optimal membership function can be set by changing the value of the index together with the constant of the membership function.

(G) Effects of the Invention As described above, the present invention provides a fuzzy control device including a first storage section storing constants of membership functions of antecedent parts to detected values, and membership functions of consequent parts to control output values. comprising a second storage unit storing constants of the function, and a weighting unit that weights the membership function of the consequent part based on the degree of establishment of the membership function of the antecedent part,
The detected value is assigned to the membership function of the antecedent part to calculate the degree of validity of the membership function of the antecedent part, and the weighting part calculates the weighted value of the membership function of the consequent part from the calculated value of the degree of validity. The center of gravity of the plurality of member functions obtained by substituting the weighted value into the membership function of the consequent part is determined, and an output corresponding to the center of gravity is obtained. 2.Fuzzy inference is performed by changing the constant value of the membership function stored in the storage unit according to the operating state of the controlled object, so the best fuzzy control can always be performed according to the operating state of the controlled object. be. In other words, it is possible to perform fuzzy inference with multiple characteristics for a single controlled object,
Even when the characteristics of the optimal fuzzy inference differ depending on the operating state of the controlled object, optimal fuzzy control can always be achieved.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram showing the outline of the present invention, Fig. 2 is a characteristic diagram showing an example of a membership function (isosceles triangle), and Fig. 3 is a characteristic diagram showing an example of a membership function (isosceles triangle).
The figure is a characteristic diagram of the membership function of the antecedent part of A mode.
Figure 4 is a characteristic diagram of the constants of the membership function of the antecedent part of B mode, Figure 5 is a correspondence diagram of the constants of the function representing the membership function of the antecedent part used in A mode and B mode, and Figure 6
The figure shows the relationship when weighting the membership function of the consequent part, Figure 7 shows the characteristics of the membership function of the consequent part, and Figure 8 shows the constants of the function representing the membership function of the consequent part. Correspondence diagram, Fig. 9 is a flowchart showing the operation of fuzzy control by a microcomputer, Fig. 10 is e.
Explanatory diagram showing the degree of establishment of the antecedent part when = e1, 1st
Fig. 1 is an explanatory diagram showing the degree of establishment of the antecedent part when Δea = Δc1, Fig. 12 is an explanatory diagram showing the membership function for weighting when e = el, Δe - Δe1,
FIG. 3 is a diagram of a figure in which the membership functions of the consequent parts that have been completely weighted according to FIG. 12 are synthesized. 1... Conditioned room, 2... Air conditioning unit, 3.
...Temperature sensor, 4...Room temperature setting device, 5...
Output section, 6...Microcomputer -7 to 10...
- Storage section, 11... Inverter IT.

Claims (1)

[Claims] (1) When controlling the operation of equipment so that the detected value of the controlled object becomes the target, an antecedent part consisting of multiple membership functions based on the detected values and the establishment of the membership function of the antecedent part It has a weighting part that weights the membership function of the consequent part according to the degree, and controls the amount of operation of the equipment and the capacity of the equipment based on a composite set of the weighted membership functions. In the fuzzy control device, the fuzzy control device includes a first storage unit that stores constants of a membership function of an antecedent part with respect to a detected value, and a constant of a membership function of a consequent part with respect to a control output value. A second storage unit and a weighting unit that weights the membership function of the consequent part based on the degree of establishment of the membership function of the antecedent part, and substitutes the detected value to the membership function of the antecedent part. calculates the degree to which the membership function of the antecedent part holds true, the weighting part calculates the weighted value of the membership function of the consequent part from the calculated value of the degree of validity, and this weighted value is used as the membership function of the consequent part. The barycentric positions of a plurality of membership functions obtained by substituting the above are obtained, and an output corresponding to the barycenter is obtained. A fuzzy control method characterized by performing fuzzy inference by selecting numerical values according to the operating state of the controlled object.