打印
[程序源码]

网上找的增量型PID控制,谁帮我解释一下!

[复制链接]
1792|5
手机看帖
扫描二维码
随时随地手机跟帖
跳转到指定楼层
楼主
网上找的PID控制代码,感觉挺好,就是有些地方看不懂,谁忙帮解释一下。


#include <stdio.h>
#include<math.h>

1.pid->pv 是什么意思?
2.pid->integral 又是什么意思。
3.pterm 应该是比例控制的中间值,可 为什么要判断它大于100或小于-100 if (pterm > 100 || pterm < -100)
4. pid->integral += pid->igain * ferror; 是什么意思。积分控制不就是等于KIe(n)吗。
  
  struct _pid {
   int pv; /*integer that contains the process value*/
   int sp; /*integer that contains the set point*/
   float integral;
   float pgain;
   float igain;
   float dgain;
   int deadband;
   int last_error;
  };
  
  struct _pid warm,*pid;
  int process_point, set_point,dead_band;
  float p_gain, i_gain, d_gain, integral_val,new_integ;;

/*------------------------------------------------------------------------
  pid_init
  
  DESCRIPTION This function initializes the pointers in the _pid structure
  to the process variable and the setpoint. *pv and *sp are
  integer pointers.
  ------------------------------------------------------------------------*/
  void pid_init(struct _pid *warm, int process_point, int set_point)
  {
   struct _pid *pid;
   
   pid = warm;
   pid->pv = process_point;
   pid->sp = set_point;
  }

/*------------------------------------------------------------------------
  pid_tune
  
  DESCRIPTION Sets the proportional gain (p_gain), integral gain (i_gain),
  derivitive gain (d_gain), and the dead band (dead_band) of
  a pid control structure _pid.
  ------------------------------------------------------------------------*/
  
  void pid_tune(struct _pid *pid, float p_gain, float i_gain, float d_gain, int dead_band)
  {
   pid->pgain = p_gain;
   pid->igain = i_gain;
   pid->dgain = d_gain;
   pid->deadband = dead_band;
   pid->integral= integral_val;
   pid->last_error=0;
  }
  
  /*------------------------------------------------------------------------
  pid_setinteg
  
  DESCRIPTION Set a new value for the integral term of the pid equation.
  This is useful for setting the initial output of the
  pid controller at start up.
  ------------------------------------------------------------------------*/
  void pid_setinteg(struct _pid *pid,float new_integ)
  {
   pid->integral = new_integ;
   pid->last_error = 0;
  }
  
  /*------------------------------------------------------------------------
  pid_bumpless
  
  DESCRIPTION Bumpless transfer algorithim. When suddenly changing
  setpoints, or when restarting the PID equation after an
  extended pause, the derivative of the equation can cause
  a bump in the controller output. This function will help
  smooth out that bump. The process value in *pv should
  be the updated just before this function is used.
  ------------------------------------------------------------------------*/
  void pid_bumpless(struct _pid *pid)
  {
  
   pid->last_error = (pid->sp)-(pid->pv);
   
  }
  
  /*------------------------------------------------------------------------
  pid_calc
  
  DESCRIPTION Performs PID calculations for the _pid structure *a. This function uses the positional form of the pid equation, and incorporates an integral windup prevention algorithim. Rectangular integration is used, so this function must be repeated on a consistent time basis for accurate control.
  
  RETURN VALUE The new output value for the pid loop.
  
  USAGE #include "control.h"*/

float pid_calc(struct _pid *pid)
  {
   int err;
   float pterm, dterm, result, ferror;
   
   err = (pid->sp) - (pid->pv);
   if (abs(err) > pid->deadband)
   {
           ferror = (float) err; /*do integer to float conversion only once*/
           pterm = pid->pgain * ferror;
           if (pterm > 100 || pterm < -100)
           {
                   pid->integral = 0.0;
           }
           else
           {
                   pid->integral += pid->igain * ferror;
                   if (pid->integral > 100.0)
                   {
                           pid->integral = 100.0;
                   }
                   else if (pid->integral < 0.0)
                        pid->integral = 0.0;
           }
           dterm = ((float)(err - pid->last_error)) * pid->dgain;
           result = pterm + pid->integral + dterm;
           }
   else
        result = pid->integral;

   pid->last_error = err;
   return (result);
  }

void main(void)
  {
   float display_value;
   int count=0;
  
   pid = &warm;
  
  // printf("Enter the values of Process point, Set point, P gain, I gain, D gain \n");
  // scanf("%d%d%f%f%f", &process_point, &set_point, &p_gain, &i_gain, &d_gain);

process_point = 30;
   set_point = 40;
   p_gain = (float)(5.2);
   i_gain = (float)(0.77);
   d_gain = (float)(0.18);

dead_band = 2;
   integral_val =(float)(0.01);

printf("The values of Process point, Set point, P gain, I gain, D gain \n");
   printf(" %6d %6d %4f %4f %4f\n", process_point, set_point, p_gain, i_gain, d_gain);
  
   printf("Enter the values of Process point\n");
  
   while(count<=20)
   {

scanf("%d",&process_point);
  
   pid_init(&warm, process_point, set_point);
   pid_tune(&warm, p_gain,i_gain,d_gain,dead_band);
   pid_setinteg(&warm,0.0); //pid_setinteg(&warm,30.0);
  
   //Get input value for process point
   pid_bumpless(&warm);
  
   // how to display output
   display_value = pid_calc(&warm);
   printf("%f\n", display_value);
   //printf("\n%f%f%f%f",warm.pv,warm.sp,warm.igain,warm.dgain);
   count++;
   
   }
  
  }

相关帖子

沙发
草民| | 2014-3-24 21:36 | 只看该作者
1、pid->pv 是指结构体里的变量,名字而已,pid是结构体名,pv是变量名。
2、同1
3、if (pterm > 100 || pterm < -100)
           {
                   pid->integral = 0.0;
           }
    应该是当pterm = pid->pgain * ferror这个计算结果大于100或者小于-100时, 积分项系数pid->integral就等于0,不需要积分控制了。
4、pid->integral += pid->igain * ferror; 是什么意思?pid->integral= pid->igain * ferror +pid->integral。积分控制当然就是误差的累积,不然和比例有什么区别?

使用特权

评论回复
板凳
草民| | 2014-3-24 21:38 | 只看该作者
位置PID的离散公式

使用特权

评论回复
地板
ichenqiang| | 2014-3-24 21:44 | 只看该作者
建议先去看看pid的原理推导,然后再来看算法实现

使用特权

评论回复
5
azurysky|  楼主 | 2014-3-26 11:08 | 只看该作者
草民 发表于 2014-3-24 21:36
1、pid->pv 是指结构体里的变量,名字而已,pid是结构体名,pv是变量名。
2、同1
3、if (pterm > 100 || pt ...

我当然知道是变量,想知道的是代表的是什么!
积分向的公式是KIe(n) 与 pid->integral += pid->igain * ferror 不符合,所以我才想知道为什么这么写。

使用特权

评论回复
6
azurysky|  楼主 | 2014-3-26 11:13 | 只看该作者
草民 发表于 2014-3-24 21:38
位置PID的离散公式

应该用的是增量公式
Δy=K[e(n)-e(n-1)]+KIe(n)+KD[e(n)-2e(n-1)+e(n-2)]
比例项与微分项都能看懂,可积分项与公式有出入。

使用特权

评论回复
发新帖 我要提问
您需要登录后才可以回帖 登录 | 注册

本版积分规则

32

主题

82

帖子

2

粉丝