CN1186972A

CN1186972A - Self regulating self correcting fuzzy controller

Info

Publication number: CN1186972A
Application number: CN 96117720
Authority: CN
Inventors: 欧阳恒刚
Original assignee: Individual
Current assignee: Individual
Priority date: 1996-09-17
Filing date: 1996-09-17
Publication date: 1998-07-08
Anticipated expiration: 2016-09-17
Also published as: CN1065345C

Abstract

The invented device is composed of basic control circuit, self-regulating control circuit and self-correction control circuit. The basic fuzzy control can be conducted to the sensed microsignal of industrial course by the process of differential amplifying, differentiating, clipping, fuzzifying and fuzzy calculation. The self correction control is conducted through the process of differential amplifying, clipping, fuzzifying, fuzzy calculation, discriminating and comparision amplifying, and the self regulating self correcting fuzzy control apparatus is composed from the above two control circuits. The advantages are resonable design, small size, reliable operation, convenient for using, low cost and wide range of application.

Description

Self regulating self correcting fuzzy controller

The present invention is the self regulating self correcting fuzzy controller that belongs to the industrial process control technology.

At present fuzzy control device is to rely on robot calculator and exist, carry out its purposes of fuzzy control with robot calculator and be subjected to certain limitation, televisor, washing machine, radio cassette player and refrigerator etc. such as household electrical appliance just are not easy to the fuzzy control device of being with microcomputer, and also there is the bigger shortcoming of volume in this device.Fuzzy logic circuit component of the present invention, the self regulating self correcting fuzzy controller of design just can be realized control to the industrial process that can't set up the coarse complexity of mathematical model and model of nonlinear system.

Purpose of design of the present invention is to be the fuzzy logic circuit component, and according to the fuzzy OR-gate and the AND gate of fuzzy logic principle, the design self regulating self correcting fuzzy controller is realized the industrial process of complexity is carried out fuzzy control.

Below with embodiment and as follows to the detailed description of the invention in conjunction with the accompanying drawings:

Fig. 1 is the basic control circuit figure of self regulating self correcting fuzzy controller;

Fig. 2 is the self-adjusting control circuit figure of fuzzy control device;

Fig. 3 be fuzzy control device from the Correction and Control circuit diagram;

Fig. 4 is the fuzzy discriminator circuit figure (17C) of fuzzy control device;

Fig. 5 is the fuzzy discriminating (17d) and the voltage comparator circuit figure of fuzzy control device;

Fig. 6 is cycle pulse generation circuit, D/A conversion and the storage circuit figure of fuzzy control device.

The present invention as shown in Figure 1, 2, 3, it is by basic control circuit, self-adjusting control circuit and becomes from the Correction and Control circuit bank, it is characterized in that having one by difference channel 1, differentiating circuit 2, amplitude limiter circuit 3, sample circuit 4, Fuzzy Calculation circuit 5a, b, fuzzy discriminator circuit 6a, and the manually operated actual value of industrial process and storage circuit 19 are formed basic control circuit to the industrial process fuzzy control through the original set-point of D/A change-over circuit 7; A sample period control circuit of forming the self-adjusting object by difference channel 8, differentiating circuit 9, amplitude limiter circuit 10, sample circuit 11, Fuzzy Calculation circuit 12a, b and fuzzy discriminator circuit 17a; One by difference channel 13, amplitude limiter circuit 14, sample circuit 15, Fuzzy Calculation circuit 16a, b and fuzzy discriminator circuit 17b, c, d form to the sample period of adjusting controlling object revise from the Correction and Control circuit, and above-mentioned control circuit and voltage comparator circuit 18, D/A change-over circuit 7, storage circuit 19 and cycle pulse generation circuit 20 and form the self regulating self correcting fuzzy controller that constitutes with the fuzzy logic circuit component.Wherein, basic control circuit as shown in Figure 1, be to deposit storage circuit 19 in by the output variable of manual observation controlling object and rate of change programming, the store digital amount converts voltage signal to as original set-point through D/A change-over circuit 7, input difference circuit 1 then during use, simultaneously to the measured parameter (temperature of industrial process, pressure, flow, liquid level) is transformed into the 0-100mA current signal through sensor, and become 0-5V level actual signal input difference circuit 1 difference through D/A change-over circuit 7 and amplify, amplifying signal input differentiating circuit 7 differential, differential signal input saturation circuit 3 is output as continuously accurately amount, be input to sample circuit 4 again and become fuzzy quantity, fuzzy quantity input Fuzzy Calculation circuit 5a, b calculates, result of calculation is input to fuzzy discriminator circuit and differentiates, forms to the needed controlling object of industrial control unit (ICU) at last.Its difference channel 1 as shown in Figure 1; be sensor feeble signal difference to be amplified with amplifier ic chip 1; 3,6 pin of IC1 connect protection ring and disturb to reduce; power supply is formed the secondary voltage stabilizing with D1,2 stabilivolts; R4 prevents that IC1 from producing controlled silicon effect, and R11 is a load circuit, and R12, CP form integrating circuit and filter high-frequency interferencing signal; R3 and R9,10,11 and R4 form backfeed loop, R7 and R8 adjust gain.Differentiating circuit 2 is to form second order damping differentiating circuit with high gain operational amplification integrated block IC 2 as shown in Figure 1.Amplitude limiter circuit 3 is the deviation variation range can be limited between the certain voltage with

ic chip

3,4,5 as shown in Figure 1, lays foundation for bluring later on.Sample circuit 4 is with eight path analoging switch ic chips 17 the accurate quantitative change of deviation variation rate to be become fuzzy set as shown in Figure 1, and the A of IC17, B, C pin connect cycle pulse generation circuit 20 (see figure 6)s.This cycle pulse generation circuit is by ic chip 120,121,122 form three element circuits, IC120 is the pulse enable unit, IC121, the 122nd, the monostable circuit of band Rc differential, output all is zero level at ordinary times, press start button SB1, IC120 triggers output switching activity and becomes 1 of IC120,7 pin, power supply is charged to 1 of IC121 to C48,7 pin are output as 0, pulse triggers IC121 through differentiating circuit, then 1 of IC121,7 pin are 1, arrive 1 of IC121 afterwards, 7 pin are output as 0 pulse and trigger IC122 by differentiating circuit and make 1 of IC120,7 pin are output as at 0 o'clock can make each control signal make independent control usefulness.Control time is the sampling time, to the period T that different physical adopted, is 1～5 second such as flow, pressure 6～10 seconds, liquid level 6～8 seconds, humidity 15～20 seconds, composition 15～20 seconds.Fuzzy Calculation circuit 5a, b as shown in Figure 1, be by IC6～16,18～20 constitute, and latch by four figures value comparison ic chip 15 and four and to trigger ic chip 14 and constitute maximum detecting circuit, when the measured data input, one the road delivers to the IC14 data relatively holds A, the data input pin D of IC15 is delivered on another road, when the POL end is identical with Cp end logic, its data are transferred to Q by D, the data of opposite Q are latched, R23,12 make the POL of IC14 hold moment identical with CP, logic is that the initial data of low level " 0 " is locked among the IC14, circuit A takes place by circulation in gating control end ST pulse signal, B, C supplies with, add high level " 1 " after, IC16 opens, import for the first time data＞0 this moment, comparison input end A＞B of IC15 then, the 13 pin output high level " 1 " of IC15 is through IC16 and IC15, R23, the differentiating circuit that D8 constitutes produces a ultra-narrow pulse makes IC14 latch new data.If the data of input are less than importing for the first time data for the second time, then the IC14 output terminal keeps data could upgrade up to there being big data to enter again, and the output terminal Q of IC14 can select maximum data thus.Minimum detecting circuit is to be made of IC6 and IC7 etc., and its circuit constitutes with the maximum data detecting circuit basic identical, and difference is the D that C11, R12 make IC7 after the energized through OR-gate ₄End is moment high level " 1 ", latch data is 1000, when importing＜1000, the 12 pin output high level " 1 " of IC6 make IC15 latch this data through the differentiating circuit that C9, R20, D7 constitute, circulation can make the output terminal Q of IC15 obtain minimum data thus, 16 pin of this IC circuit 7, IC14 connect positive supply, 8 pin ground connection.Fuzzy discriminator circuit 6a as shown in Figure 1, be to constitute by ic chip 21,22,23,24 etc., this circuit decomposes logical function, be f (X1, X2 ... Xn) decompose 0＜α n-1＜α n-2 ... α 1＜1, as long as input f drops on (a1, a1-10) when interior, have only 1 output terminal to export " 1 ".Storage circuit 19 and D/A change-over circuit 7 are as shown in Figure 6, be to constitute storage circuit by ic chip 123,124,125 etc., storage volume is 1KX8, putting counting unit is made of toggle switch K1-8 and two 4 reversible counting chips, execution instruction address can be inserted with K1-8, when PE end output this moment positive pulse was " 1 ", U/D was a plus coujnt, and low level is the subtraction counting; And constitute the D/A change-over circuit by IC126 and computing amplification integrated block IC 127 etc., can convert data-signal D0-7 to the 0-5V voltage signal and be output as set-point.

The self-adjusting control circuit as shown in Figure 2, be by difference channel 8, differentiating circuit 9, amplitude limiter circuit 10, sample circuit 11, Fuzzy Calculation circuit 12a, b and fuzzy discriminator circuit 17a form, its industrial process sensor output III signal with by storage circuit 19 and through the original set-point signal of D/A change-over circuit 7 conversions simultaneously ic chip 29 difference of input difference circuit 8 amplify, the IC30 differential of amplifying signal input differentiating circuit 9, differential signal is through the IC21 of amplitude limiter circuit 10,22 amplitude limits, the IC26 of clipped signal input sampling circuit 11,27 sampling and obfuscations, blurred signal input Fuzzy Calculation circuit 12a, b and by the IC33-47 Fuzzy Calculation, the IC48-55 of the fuzzy discriminator circuit 17a of result of calculation input differentiates Fuzzy Calculation.

From Correction and Control circuit such as Fig. 3,4, shown in 5, be by difference channel 13, amplitude limiter circuit 14, sample circuit 15, Fuzzy Calculation circuit 16a, b and fuzzy discriminator circuit 17b, c, d forms, its industrial process sensor is exported III signal and is amplified by reservoir 19 and through the original set-point signal IC56 computing of input difference circuit 13 simultaneously of D/A change-over circuit 7 conversions, the IC57 of amplifying signal input saturation circuit 14,58 amplitude limits, the IC59 of clipped signal input sampling circuit 15,60 obfuscations, blurred signal input Fuzzy Calculation circuit 16a, the IC61-74 of b makes Fuzzy Calculation, the Fuzzy Calculation result's of the signal of result of calculation and self-adjusting control circuit signal is imported fuzzy discriminator circuit 17b simultaneously, c, the IC75-119 of d differentiates, distinguishing signal output quantity ∑ is through the IC126 of IC119 and D/A change-over circuit 7, the 127 original set-point signals IC128 of input voltage comparator circuit 18 simultaneously relatively amplify, and last amplifying signal input actuating motor was controlled the sampling period.

Applicating example: such as full-automatic fuzzy washing machine, be arranged on the drain outlet place with infrared sensor, the transparency that detects laundry is to judge clothing dirt degree and laundry degree; Increase with the vibrating membrane pressure of level sensor after water filling, drive inductance L and the capacitor C that moves with the change coil unshakable in one's determination, thereby can reflect the water level height with the oscillation frequency of LC.The cloth consumption sensor can calculate clothes weight by laundry motor load effect (electric current or moment).The control of washing machine is water level in the cylinder, wash time and clothes weight etc., control device determines optimum level of water according to clothes weight, determine best wash time, best rinsing time and number of times according to dirt degree and character, and discharge water and the Best Times that dewaters is decided by water level and weight.

In a word, the self regulating self correcting fuzzy controller that manufactures and designs according to present embodiment, just can carry out the thick control that blurred signal amplifies, calculates and differentiate by fuzzy control to the fuzzy parameter signal that industrial process spreads out of, and the fuzzy control signal is adjusted amplification automatically, calculates and differentiated by the self-adjusting fuzzy control, simultaneously the fuzzy control signal is revised, calculated and differentiates, through comparison output signal the fuzzy parameter amount of industrial process is carried out fuzzy control at last by reviewing one's lessons by oneself positive fuzzy control.The fuzzy control device that the present invention adopts fuzzy logic circuit component design to form has that composite design is reasonable, volume is little, a reliable operation, easy to use and advantage that cost is low, and has the value of extensive practicality.

Claims

1. holotype phase control device is reviewed one's lessons by oneself in a self-adjusting, it is by basic control circuit, self-adjusting control circuit and becomes from the Correction and Control circuit bank, it is characterized in that having one by difference channel 1, differentiating circuit 2, amplitude limiter circuit 3, sample circuit 4, Fuzzy Calculation circuit 5a, b and fuzzy discriminator circuit 6a, and manually operated actual value and storage circuit 19 are formed basic control circuit to the industrial process fuzzy control through the set-point of D/A change-over circuit 7; A sample period control circuit of forming the self-adjusting object by difference channel 8, differentiating circuit 9, amplitude limiter circuit 10, sample circuit 11, Fuzzy Calculation circuit 12a, b and fuzzy discriminator circuit 17a; One by difference channel 13, amplitude limiter circuit 14, sample circuit 15, Fuzzy Calculation circuit 16a, b and fuzzy discriminator circuit 17b form to the sample period of adjusting controlling object revise from the Correction and Control circuit, and above-mentioned control circuit and fuzzy discriminator circuit 17c, d, voltage comparator circuit 18, D/A change-over circuit 7, storage circuit 19 and cycle pulse generation circuit 20 and form the self regulating self correcting fuzzy controller that constitutes with the fuzzy logic circuit component.

2. self regulating self correcting fuzzy controller as claimed in claim 1, the difference channel 1 usefulness amplifier ic chip 1 that it is characterized in that basic control circuit amplifies sensor feeble signal difference, differentiating circuit 2 usefulness computing amplification integrated block ICs 2 are formed second order damping differentiating circuit, amplitude limiter circuit 3 usefulness ic chips 3,4,5 variation ranges with deviation are limited between the certain voltage and lay foundation for sampling, sample circuit 4 usefulness eight path analoging switch ic chips 17 become fuzzy set to the fuzzy control deviation, the A of IC17, B, the C pin connects cycle pulse generation circuit 20, this circuit IC120,121,122 trigger pulses start, Fuzzy Calculation circuit 5a, b is by IC6-16,18-20 constitutes, and latch by four figures value comparison ic chip 15 and four and to trigger ic chip 14 and constitute maximum and detect, constituting minimum by ic chip 6 and IC8 detects, fuzzy discriminator circuit 6a constitutes with ic chip 21-24, this circuit decomposes logical function, storage circuit 19 usefulness ic chips 123～125 constitute, capacity is 1KX8, D/A change-over circuit 7 usefulness ic chips 126 and computing amplifier integrated block C127 constitute, and data-signal D0～7 can be transformed into 0～5V voltage signal and be output as set-point.

3. self regulating self correcting fuzzy controller as claimed in claim 1, it is characterized in that the self-adjusting control circuit be by industrial process sensor output III signal with by storage circuit 19 and through the original set-point signal of D/A change-over circuit 7 conversions simultaneously ic chip 29 difference of input difference circuit 8 amplify, the IC30 differential of amplifying signal input differentiating circuit 9, differential signal is through the IC21 of amplitude limiter circuit 10,22 amplitude limits, the IC26 of clipped signal input sampling circuit 11,27 sampling and obfuscations, this blurred signal input Fuzzy Calculation circuit 12a, the IC33-47 of b makes Fuzzy Calculation, and the IC48～55 pair Fuzzy Calculation of the fuzzy discriminator circuit 17a of result of calculation input is differentiated.

4. self regulating self correcting fuzzy controller as claimed in claim 1, it is characterized in that from the Correction and Control circuit it being to amplify through the original set-point signal IC56 computing of input difference circuit 13 simultaneously of D/A change-over circuit 7 by III signal and storage circuit 19 that industrial process sensor is exported, the IC57 of amplifying signal input saturation circuit 14,58 amplitude limits, the IC59 of clipped signal input sampling circuit 15,60 obfuscations, blurred signal input Fuzzy Calculation circuit 16a, Fuzzy Calculation is made in the IC61 of b～74, the Fuzzy Calculation result's of the signal of result of calculation and self-adjusting control circuit signal is imported fuzzy discriminator circuit 17b simultaneously, c, the IC75 of d～119 are differentiated, the original set-point signal of differentiating output quantity and D/A change-over circuit 7 relatively amplifies to the IC128 of voltage comparator circuit 18 simultaneously, and last amplifying signal input actuating motor was controlled the sampling period.