Introduction to Industrial Automation

Figure 9.2  Example of parameters to be assigned in a PID FB program (Siemens).

---

378

 

◾

  Introduction to Industrial Automation

Analog input 1

Analog input 2

Analog input 4

Analog output 1

Analog output 2

Analog output 4

PID

Controller 1

PID

Controller 2

PID

Controller 4

Figure 9.3  An independent and separate module of PID control offers multiple control loops.

SP

PV

PID

Controller 4

PID

Controller 1

PID

Controller 2

PID

Controller 3

SP 1

SP 2

SP 3

PV 1

PV 2

PV 3

Constant 1

Constant 2

Constant 3

Process 1

Process 2

Process 3

+

+

−

×

×

×

−

−

−

Figure 9.4  Combination of PID controllers for controlling three interrelated sub-processes 

(Siemens, FM 355 PID Module).

SP 1

SP 2

PV 1

PV 2

PID

Controller 1

PID

Controller 2

Process 1

Process 2

−

×

−

Figure 9.5  Sequential control of two processes with two PID control loops.

---

PID Control in the Industry 

◾

 

379

controller 2 operation will follow, and so on, as shown in Figure 9.6. The result of the PID algorithm 

on each controller is transferred directly to the corresponding analog output without waiting for the 

operation of the next controller. If a controller is not used by the user, then its function is bypassed 

and therefore does not contribute to the total time of the operation cycle of a PID control module. In 

this case, special virtual panels provide the user with a form of menu, the ability to easily define the 

number and type of controller terms (P, PI, PD, and PID), the combination of inner loops to introduce 

the values of the PID controller’s gain and, in general, to organize the control task of the process, as 

happens in the FM-355 closed-loop control module of Siemens, for example.

In some PLCs and in their corresponding software for a PID FB, a method for self-regulation 

(auto-tuning) of the PID controller’s gains has been integrated. This facilitates the work of the 

control engineer, provided that the basic characteristics of the controlled process and especially 

the slow or fast response of the process under control are known a priori. One method for the 

self-regulation of the PID controller that is widely used from control engineers is the so-called 

technique of “feedback with relay” by Aström and Hägglund (1984). Over the last 25 years, this 

technique has been used by a large number of manufacturers, as well by developing proper FBs. 

The central idea of the technique is the supply of a small and sustained oscillation of the controlled 

process, which is generally stable. Based on the period of oscillations and the size of variations that 

are observed in the process variable, the fundamental frequency and the gain of the controlled 

process can be calculated. Based on these two values, the auto-tuning algorithm is able to calculate 

the values of the gains P, I, and D of the controller, which obviously would not be the optimum 

but would still be very close. Finally, it should be noted that special attention should be followed 

in this method not to destroy or create fatigue in the mechanical parts of the controlled process by 

the application of the oscillating inputs.

Run

Analog input 1

Analog input 2

Analog input 3

Analog input 4

Analog output 1

Analog output 2

Analog output 3

Analog output 4

PID

Controller 1

PID

Controller 2

PID

Controller 3

PID

Controller 4

Figure 9.6  Sequential execution of PID control in four controllers with a direct update of ana-