PID控制器自动整定：继电器反馈法

5星 · 超过95%的资源需积分: 10 52 浏览量更新于2024-07-22 6 收藏 8.09MB PDF 举报

"《PID控制器的自动整定：继电器反馈法》由Cheng-Ching Yu博士撰写，旨在帮助工程师和研究人员提升控制器性能。本书提供了一种简单而有效的PID控制器自动整定方法，同时考虑了不同过程条件，如负载扰动、非线性和噪声。书中尽量减少了数学理论的深度，重点在于实验设计，以获取与调参规则相关的实际过程信息。通过大量实例和案例研究，阐述了自动整定过程和闭环性能。本书作为独立学习工具，涵盖了自动整定的多个方面，鼓励读者将其应用于工业控制实践。" 在自动化控制领域，PID（比例-积分-微分）控制器是广泛应用的控制算法，因为它能够对各种系统进行稳定和精确的控制。然而，手动调整PID参数往往需要专业知识和经验，且可能无法应对复杂的过程变化。因此，自动整定技术应运而生，以优化控制器的性能。本书“Autotuning of PID Controllers: A Relay Feedback Approach”介绍了一种基于继电器反馈的自动整定方法。继电器反馈法是一种通过短暂断开控制器和被控对象之间的连接，利用系统的振荡来获取系统动态特性的方法。这种方法相对简单，且在实践中易于实施。通过这种方法，可以获取关于系统响应速度、稳定性和非线性行为的关键信息，从而合理地设定PID参数。作者Cheng-Ching Yu博士在书中详细解释了如何使用继电器反馈法进行自动整定，并提供了必要的实用工具来处理各种过程挑战。书中包含的140个图形有助于读者直观理解理论和实践操作。此外，大量实例和案例研究使理论知识更加具体，让读者能够了解如何将这些方法应用到实际的工业控制系统中。书中的内容不仅适用于有经验的控制工程师，也适合初学者。它没有过分依赖复杂的数学描述，而是强调实验设计和实际操作，使得读者能够快速掌握自动整定技术。此外，书中还提到了MATLAB，这是一个广泛用于控制系统设计和仿真的软件，表明本书也关注于现代控制工程中的数字化工具。总而言之，"Autotuning of PID Controllers: A Relay Feedback Approach" 是一本深入浅出的教材，涵盖了PID控制器自动整定的各个方面，对于希望提高控制器性能的工程师和研究人员来说，是一本不可多得的参考资料。

Introduction 5

Table 1.1. Autotuners from different vendors

Manufacturer Identification

method

Gain scheduling

ABB Step/relay Yes

Emerson Process Management Relay Yes

Foxboro Step No

Honeywell Step Yes

Siemens Step Yes

Yokogawa Step Yes

1. Control tuning can improve the performance, but it should be recognized that

good tuning can only solve part of the problem.

2. The experimental design for system identification becomes rather important,

since we are not able to keep all the control valves in perfect condition.

3. The system identification step should be time efficient. This is rather useful for

many slow industrial processes.

1.3 Relay Feedback Identification

System identification plays an integral part in automatic tuning of the PID control-

ler. Based on the information obtained, the methods for identification can be classi-

fied into the frequency-domain and time-domain approaches.

The time-domain approaches generate responses from step or pulse tests [2,3].

The characteristics of the process response are then utilized to back-calculate the

parameters of an assumed process model [19]. The step tests can be performed in

open-loop (manual mode) or closed-loop mode (while controller is working). The

open-loop step test is fairly straightforward. However, it is vulnerable to load dis-

turbances, especially for systems with large time constants. Moreover, the behavior

of the control valve is not fully tested in the experiment. The closed-loop step tests,

on the other hand, can shorten the time for experiment. But we have to choose a set

of controller parameters in order to generate oscillatory (underdamped) responses

[19]. The process model is then approximated from the damping behavior. The pat-

tern recognition controller [20,21] is a typical example. Since step-like change is

involved, it is not expected to work well for highly non-linear systems, (e.g. high

purity distillation columns [22]).

Another category is the Ziegler–Nichols type of experimental design. Probably

the more successful part of the Ziegler–Nichols method is not the tuning rule itself.

Rather, it is the identification procedure: a way to find the important process in-

formation, ultimate gain

K and ultimate frequency



. This is often referred to

as the trial-and-error procedure [2,3]. A typical approach can be summarized as

follows:

6 Autotuning of PID Control lers

1. Set the controller gain

K at a low value, perhaps 0.2.

2. Put the controller in the automatic mode.

3. Make a small change in the set point or load variable and observe the response.

If the gain is low, then the response will be sluggish.

4. Increase the gain by a factor of two and make another set point or load change.

5. Repeat step 4 until the loop becomes oscillatory and continuous cycling is ob-

served. The gain at which this occurs is the ultimate gain

K ,andtheperiodof

oscillation is the ultimate period

P (



/2= ).

This is a simple and reliable approach to obtain

K and



. The disadvantage is

also obvious: it is time consuming. The present-day version is the relay feedback

test proposed by Åströmand Hägglund [23]. First, a continuous cycling of the con-

trolled variable is generated from a relay feedback experiment and the important

process information,

K and



, can be extracted directly from the experiment.

The information obtained from the relay feedback experiment is exactly the same

as that from the conventional continuous cycling method. It should be noticed that

the relay feedback is an old and useful technique for feedback control, as can be

seen from earlier results [24,25], and, here, a new meaning is assigned to the relay

feedback. However, an important difference is that the sustained oscillation is gen-

erated in a controlled manner (e.g. the magnitude of oscillation can be controlled)

in the relay feedback test. Moreover, in virtually all cases, this is a very efficient

way, i.e. a one-shot solution, to generate a sustain oscillation. Applications of the

Åströmand Hägglund autotuner are found throughout process industries using sin-

gle-station controllers or a DCS (Table 1.1). The success of this autotuner is due to

the fact that the identification and tuning mechanism are so simple that operators

understand how it works. It also works well in slow and highly nonlinear processes

[22]. Over the past two decades, extensive research has been done on relay feed-

back tests. Refinements on the accuracy and improvements on the experimental de-

sign have been made. Discussions about potential problems, extensions to multi-

variable systems and incorporation of gain scheduling have also been reported.

Luyben brings the autotuner to another level in which the “shape” of relay feed-

back can be utilized to identify the model structure. This motivates us for the revi-

sion. It is our view that the relay-feedback-based autotuners now can provide the

necessary tools to improve control performance in a reliable way.

1.4 Conclusion

In this chapter we clearly define the scope of process control, and one should real-

ize that the controller tuning only constitutes a fraction of the process operation

problems. Surveys indicate that the PID controller is the major controller in proc-

ess industries. After many years of experience, the control loops, often thought too

simple, do not perform as well as one might expect. The failure comes from the

Introduction 7

lack of the required knowledge to maintain the control loops, to tune the control-

lers, to design an appropriate process for control and to design a suitable control

configuration for a given process. Poor control performance may have many dif-

ferent causes. However, obtaining good tuning is always the most cost-effective

way to improve control. You should recognize that controllers are working with

imperfect valves, noisy sensors and frequent load disturbances. These factors have

to be taken into account when you are designing the experiment to find controller

parameters.

1.5 References

1. Ogunnaike BA, Ray WH. Process dynamics, modeling, and control. New York: Oxford

University Press; 1994.

2. Luyben WL, Luyben ML. Essentials of process control. New York: McGraw-Hill;

1997.

3. Seborg DE, Edgar TF, Mellichamp DA. Process dynamics and control; 2nd ed. New

York: John Wiley & Sons; 2004.

4. Luyben WL, Tyreus BD, Luyben ML. Plantwide control. New York: McGraw-Hill;

1999.

5. Seferlis P, Georgiadis MC. The integration of process design and control. 2nd ed. Am-

sterdam: Elsevier; 2004.

6. Morari M, Zafiriou E. Robust process control. Prentice Hall: Englewood Cliff; 1989.

7. Desborough L, Miller R. Increasing customer value of industrial control performance

monitoring-Honeywell experience. In: 6th Int. Conf. Chemical Process Control, AIChE

Symp. Series 326; Rawlings JB, Ogunnaike BA, Eaton JW. Eds. AIChE: New York;

2002.

8. Ziegler JG, Nichols NB. Optimum settings for automatic controllers. Trans. ASME

1942;12:759.

9. Bialkowski WL. Dream vs Reality: A view from both sides of the gap. Pulp Paper Can

1993;94:19.

10. Ender DB. Process control performance: Not as good as you think. Control Eng

1993;40:180.

11. McMillan GK. Tuning and control loop performance. Instrument Society of America:

Research Triangle Park; 1994.

12. Hersh MA, Johnson MA. A study of advanced control systems in the workplace. Con-

trol Eng. Prac. 1997;5:771.

13. Cheng YC, Yu CC. Relay feedback identification for actuators with hysteresis. Ind.

Eng. Chem. Res. 2000;39:4239.

剩余266页未读，继续阅读

ganalog

粉丝: 10
资源: 16

PID控制器自动整定：继电器反馈法

automatic sql tuning in oracle 10g

Autotuning with PID Controller:autotuning with pid controller-matlab开发

Rectangle Pulse Response Method AutoTuning

藏经阁-Spark Autotuning.pdf

ck-autotuning:CK自动化操作使用户能够实现可移植，可自定义和可重复使用的程序工作流，以实现可再现，协作和多目标的基准测试，优化和SWHW协同设计

PID参数调试

labview的Pid工具包

matlab开发-PID控制器标签代码

士研电机PID数位控制器TF系列说明书.pdf

Spark Autotuning: 藏经阁文件优化手册

最新资源