如何实现rbf神经网络的模糊pid控制仿真

如何实现rbf神经网络的模糊pid控制仿真,第1张

我的毕设只用把PID和模糊PID相比较

常规PID,用Matlab里的Simulink模块仿真,建立你要做的动力学模型的传函或者状态空间。PID参数调节可用临界比度法。

模糊PID就麻烦了,打开Matlab中FIS模块,一般都用二阶模糊?输入E,EC的隶属函数,一般为高斯,和输出模糊Kp,Ki,Kd,一般为三角。还要整定模糊规则,再加载到Simulink里。调节模糊因子Gu,Ge,Gec,设置模糊PID的参数。

总之,你这个问题在白度知道里很难说清楚。

PIC单片机的你看一下

//*********************************************************************************

#include <pic.h>

#include <pic16f684.h>

#include <math.h>

#include <stdlib.h>

void Init()

void PID()

void Set_Constants()

bit flag1,do_PID,int_flag

signed char en0, en1, en2, en3, term1_char, term2_char, off_set

unsigned char temp

short int temp_int

unsigned short int ki, kd, kp

signed int SumE_Min, SumE_Max, SumE, integral_term, derivative_term, un

signed long Cn

// __CONFIG _CP_OFF &_CPD_OFF &_BOD_OFF &_MCLRE_ON &_WDT_OFF &_INTRC_OSC_NOCLKOUT &_FCMEN_ON

//***************************************************************************

// Positional PID 256 Hz

//***************************************************************************

//***************************************************************************

//Main() - Main Routine

//***************************************************************************

void main()

{

Init() //Initialize 12F629 Microcontroller

Set_Constants() //Get PID coefficients ki, kp and kd

while(1) //Loop Forever

{

if(do_PID){

PID()

}

}

}

//***************************************************************************

//Init - Initialization Routine

//***************************************************************************

void Init()

{

PORTA = 0

TRISA = 0b00101101 // Set RA4 and RA2 as outputs

PORTC = 0

TRISC = 0b00000011 // Set RC0 and RC1 as inputs, rest outputs

CMCON0 = 0x07 // Disable the comparator

IRCF0 = 1 // Used to set intrc speed to 8 MHz

IRCF1 = 1 // Used to set intrc speed to 8 MHz

IRCF2 = 1 // Used to set intrc speed to 8 MHz

CCP1CON = 0b01001100 // Full bridge PWM forward

ECCPAS = 0 // Auto_shutdown is disabled for now

PR2 = 0x3F // Sets PWM Period at 31.2 kHz

T2CON = 0 // TMR2 Off with no prescale

CCPR1L = 0 // Sets Duty Cycle to zero

TMR2ON = 1 // Start Timer2

ANSEL = 0b00110101 // Configure AN0,AN2,AN4 and AN5 as analog

VCFG = 0 // Use Vdd as Ref

ADFM = 1 // Right justified A/D result

ADCS0 = 1 // 16 TOSC prescale

ADCS1 = 0

ADCS2 = 1

CHS0 = 0 // Channel select AN0

CHS1 = 0

CHS2 = 0

ADON = 1 //Turn A/D on

en0 = en1 = en2 = en3 = term1_char = term2_char =0

ki = kd = 0

kp = off_set = 0

temp_int = integral_term = derivative_term = un =0

SumE_Max = 30000

SumE_Min = 1 - SumE_Max

do_PID = 1 // Allowed to do PID function

T0CS = 0 // Timer0 as timer not a counter

TMR0 = 10 // Preload value

PSA = 0 // Prescaler to Timer0

PS0 = 0 // Prescale to 32 =>256 Hz

PS1 = 0

PS2 = 1

INTCON = 0

PIE1 = 0

T0IE = 1 // Enable Timer0 int

GIE = 1

return

}

void PID() // The from of the PID is C(n) = K(E(n) + (Ts/Ti)SumE + (Td/Ts)[E(n) - E(n-1)])

{

integral_term = derivative_term = 0

// Calculate the integral term

SumE = SumE + en0 // SumE is the summation of the error terms

if(SumE >SumE_Max){ // Test if the summation is too big

SumE = SumE_Max

}

if(SumE <SumE_Min){ // Test if the summation is too small

SumE = SumE_Min

} // Integral term is (Ts/Ti)*SumE where Ti is Kp/Ki

// and Ts is the sampling period

// Actual equation used to calculate the integral term is

// Ki*SumE/(Kp*Fs*X) where X is an unknown scaling factor

// and Fs is the sampling frequency

integral_term = SumE / 256 // Divide by the sampling frequency

integral_term = integral_term * ki // Multiply Ki

integral_term = integral_term / 16 // combination of scaling factor and Kp

// Calculate the derivative term

derivative_term = en0 - en3

if(derivative_term >120){ // Test if too large

derivative_term = 120

}

if(derivative_term <-120){ // test if too small

derivative_term = -120

} // Calculate derivative term using (Td/Ts)[E(n) - E(n-1)]

// Where Td is Kd/Kp

// Actual equation used is Kd(en0-en3)/(Kp*X*3*Ts)

derivative_term = derivative_term * kd // Where X is an unknown scaling factor

derivative_term = derivative_term >>5 // divide by 32 precalculated Kp*X*3*Ts

if(derivative_term >120){

derivative_term = 120

}

if(derivative_term <-120){

derivative_term = -120

}

// C(n) = K(E(n) + (Ts/Ti)SumE + (Td/Ts)[E(n) - E(n-1)])

Cn = en0 + integral_term + derivative_term // Sum the terms

Cn = Cn * kp / 1024 // multiply by Kp then scale

if(Cn >= 1000) // Used to limit duty cycle not to have punch through

{

Cn = 1000

}

if(Cn <= -1000)

{

Cn = -1000

}

if(Cn == 0){ // Set the speed of the PWM

DC1B1 = DC1B1 = 0

CCPR1L = 0

}

if(Cn >0){ // Motor should go forward and set the duty cycle to Cn

P1M1 = 0 // Motor is going forward

temp = Cn

if(temp^0b00000001){

DC1B0 = 1

}

else{

DC1B0 = 0

}

if(temp^0b00000010){

DC1B1 = 1

}

else{

DC1B1 = 0

}

CCPR1L = Cn >>2 // Used to stop the pendulum from continually going around in a circle

off_set = off_set +1 // the offset is use to adjust the angle of the pendulum to slightly

if(off_set >55){ // larger than it actually is

off_set = 55

}

}

else { // Motor should go backwards and set the duty cycle to Cn

P1M1 = 1 // Motor is going backwards

temp_int = abs(Cn) // Returns the absolute int value of Cn

temp = temp_int // int to char of LS-Byte

if(temp^0b00000001){

DC1B0 = 1

}

else{

DC1B0 = 0

}

if(temp^0b00000010){

DC1B1 = 1

}

else{

DC1B1 = 0

}

CCPR1L = temp_int >>2 // Used to stop the pendulum from continually going around in a circle

off_set = off_set -1

if(off_set <-55){

off_set = -55

}

}

en3 = en2 // Shift error signals

en2 = en1

en1 = en0

en0 = 0

do_PID = 0 // Done

RA4 = 0 // Test flag to measure the speed of the loop

return

}

void Set_Constants()

{

ANS2 = 1 // Configure AN2 as analog

ANS4 = 1 // Configure AN4 as analog

ANS5 = 1 // Configure AN5 as analog

ADFM = 1 // Right justified A/D result

CHS0 = 0 // Channel select AN4

CHS1 = 0

CHS2 = 1

temp = 200 // Gives delay

while(temp){

temp--

}

GODONE = 1

while(GODONE){

temp = 0 // Does nothing.....

}

ki = ADRESH <<8 // Store the A/D result to Integral Constant

ki = ki + ADRESL

CHS0 = 1 // Channel select AN5

CHS1 = 0

CHS2 = 1

temp = 200 // Gives delay

while(temp){

temp--

}

GODONE = 1

while(GODONE){

temp = 0 // Does nothing.....

}

kd = ADRESH <<8 // Store the A/D result to Differential Constant

kd = kd + ADRESL

CHS0 = 0 // Channel select AN2

CHS1 = 1

CHS2 = 0

temp = 200 // Gives delay

while(temp){

temp--

}

GODONE = 1

while(GODONE){

temp = 0 // Does nothing.....

}

kp = ADRESH <<8 // Store the A/D result to Proportional Constant

kp = kp + ADRESL

CHS0 = 0 // Channel select AN0

CHS1 = 0

CHS2 = 0

}

void interrupt Isr()

{

if(T0IF&&T0IE){

TMR0 = 10 // Preload value

T0IF = 0 // Clear Int Flag

// flag1 = (!flag1)

RA4 = 1

temp_int = 0

temp_int = ADRESH <<8 // Store the A/D result with offset

temp_int = temp_int + ADRESL - 512

en0 = temp_int + off_set/8 // Store to error function asuming no over-flow

do_PID = 1 // Allowed to do PID function

GODONE = 1 // Start next A/D cycle

}

else

{

PIR1 = 0

RAIF = 0

INTF = 0

}

if(temp_int >180){ //Check if error is too large (positive)

DC1B0 = DC1B1 = 0 // Stop PWM

CCPR1L = 0

en0 = en1 = en2 = en3 = term1_char = term2_char = off_set = 0 // Clear all PID constants

Cn = integral_term = derivative_term = SumE = RA4 = 0

do_PID = 0 // Stop doing PID

}

if(temp_int <-180){ //Check if error is too large (negative)

DC1B0 = DC1B1 = 0 // Stop PWM

CCPR1L = 0

en0 = en1 = en2 = en3 = term1_char = term2_char = off_set = 0 // Clear all PID constants

Cn = integral_term = derivative_term = SumE = RA4 = 0

do_PID = 0 // Stop doing PID

}

}


欢迎分享,转载请注明来源:内存溢出

原文地址: http://outofmemory.cn/yw/11998915.html

(0)
打赏 微信扫一扫 微信扫一扫 支付宝扫一扫 支付宝扫一扫
上一篇 2023-05-20
下一篇 2023-05-20

发表评论

登录后才能评论

评论列表(0条)

保存