自動化專業(yè)畢業(yè)論文外文翻譯--對循環(huán)流化床鍋爐燃燒系統(tǒng)的多變量解耦控制系統(tǒng)的研究（英文）

1、 1Abstract--The combustion system of Circulating Fluidized Bed Boiler (CFBB) is a complex object with multivariable, large delay, tightly coupling, nonlinear and slow time-varying. It is difficult to build the precis

2、e mathematic model of the object and to accurately control the object with the traditional control methods. In this paper, the object is dynamically decoupled by a feed-forward compensator. Then, the object is respect

3、ively controlled by three controllers, PID controller, fuzzy controller and self-tuning parameter fuzzy–PID controller. The simulation experiments are carried out through contrast with former controllers in MATLAB si

4、mulation environment. The simulation results show that the self-tuning parameter fuzzy-PID controller is superior to the general PID controller and the general fuzzy controller in speediness, stability, adaptability,

5、robustness and ability of anti-disturbance. Index Terms--Circulating Fluidized Bed Boiler; Decoupling; Fuzzy Control; Multivariable; Self-tuning Parameter Fuzzy-PID Control I．INTRODUCTION HE circulating fluidized bed

6、 boiler (CFBB) has been widely used as a clean coal combustion technique for its high combustion efficiency, wide adaptability to coal ranks, load adjusting performance and low pollution [1-4]. However, due to its sp

7、ecial structure and the complexity of combustion mechanism, the combustion process has complex features, such as highly nonlinear, time-varying, large delay and multivariable decoupling, etc. it is very difficult to e

8、stablish its precise mathematical model [5-8], and to control the object with the traditional control methods. At present, the common control method is that the main steam pressure is focally controlled, meanwhile,

9、the primary air flow is adjusted according to the best air-coal ratio, and the bed temperature is controlled in the required range [9]. The method can’t maintain the bed temperature in the best range while keeping th

10、e main steam pressure. In this paper, on the basis of decoupling model of the main steam pressure and the bed temperature [10], the self-tuning fuzzy-PID controller [11-18] which has better adaptability and This wor

11、k was supported by Leading Academic Discipline Project of Shanghai Municipal Education Commission (Project Number: J51301) and Nature Science Key Foundation of Shanghai Municipal Education Commission (Project Number:

12、06ZZ69). All of the authors are with College of Electric Power and Automation, Shanghai University of Electric Power, Yangpu District, Shanghai, 200090, China(email: chengqiming@sina.com). better robustness, is used t

13、o control the steam pressure and the bed temperature to get the better control effects. II．THE CHARACTERISTICS OF CFBB AND THE STRUCTURE OF THE DECOUPLED CONTROL SYSTEM OF CFBB The key reasons that CFBB combustion sys

14、tem is difficult to control, are the strong coupling relations between multi- inputs (coal, primary air, secondary air, drawing wind, recycle material) and multi-outputs (bed temperature, main steam pressure, the furn

15、ace negative pressure, oxygen content), the most important relation in these coupling relations is the coupling relation between bed temperature and main steam pressure [5-6]. In China, CFBB are usually designed wit

16、hout external heat exchanger to ensure simple structure and low cost. The main steam pressure and the bed temperature with strongly coupled relation are controlled by regulating the coal amount and the primary air am

17、ount. This control method is widely used to the actual control of CFBB combustion system in China [12]. Therefore, in this paper, the bed temperature is controlled by the primary air amount, and the main steam pressur

18、e is regulated by the coal amount. In this paper, the domestic 75 t/h CFBB is selected as the object of simulation experiments. The transfer function matrix of the system at the load range of 70% to 100% is gotten [

19、16]: ? ?? ? ??? ? ? ???? ? ? ???+ ++?+ =? ?? ? ?? ? ?? ? ?? = ? ?? ? ????1 260223021 22 2112 110) 1 160 (036 . 0) 1 260 (0066 . 0 ) 1 163 (8767 . 8) 1 180 (386 . 0) ( ) () ( ) (QBs e ss e sQBs G s Gs G s GPTssb(1) where

20、 Tb, P0 are respectively bed temperature and main steam pressure; B, Q1 are respectively coal amount and primary air amount; G11, G12 G21, G22 are respectively input-output transfer functions of B-Tb, B-P0, Q1-Tb and

21、Q1-P0. From Eq.(1) to see, that time delays exist in both coal - main steam pressure loop and coal - bed temperature loop and there are seriously coupling relations in CFBB system. Therefore, dynamic feed-forward co

22、mpensation is needed for dynamic decoupling system to control the system better. Dynamic feed-forward compensation is commonly used for system decoupling [10]. Fig.1 is the structure of the Research on Multivariable D

23、ecoupling Control System for Combustion System of Circulating Fluidized Bed Boiler Qi-Ming Cheng, Rui-Qing Guo, and Xu-Feng Du T3trial-and-error method. IV．THE DESIGN OF SELF-TUNING FUZZY-PID CONTROLLER FOR COMBUSTI

24、ON CONTROL SYSTEM OF CFBB A. The Structure of Control System The PID controller has many advantages, such as simple structure, mature theoretical basis, wide applicability, convenient parameter tuning, much engineerin

25、g application. Therefore, the PID controller occupies a dominant position in the actual control system. But the linear characteristics of the conventional PID controller with fixed control parameters have good contro

26、l performance only when working near operating point. When the system is farther out of the operating point, non-linear control characteristics of the object is difficult to maintain the dynamic quality of PID control

27、. Therefore, fuzzy reasoning is introduced to solved the problem, the parameters of PID controller based on the initial PID control parameters are corrected with adding fuzzy reasoning to improve the system dynamic p

28、erformance. In self-tuning fuzzy-PID controller [16-18], the conditions and the operations of control rules are expressed by fuzzy set on the basis of PID control, and these fuzzy control rules as well as other infor

29、mation are stored into computer knowledge bases, then, according to actual response of the control system, fuzzy reasoning is carried out by computer to achieve best adjustment of PID control. The structure of self-tu

30、ning parameter fuzzy-PID control system is shown in Fig.5. Firstly, the self-tuning fuzzy-PID controller is designed to find the fuzzy relations between the three control parameters (namely, proportional coefficient

31、Kp, differential coefficient Kd and integral coefficient Ki ) and the two system variables (that is, error e, error deviation ec); Then, ΔKp, ΔKi, ΔKd, the changes of Kp, Ki, Kd PID control parameters are online ame

32、nded on fuzzy theory with measuring e and ec during the system operation; Finally, the control system has good dynamic and static performance. The digital PID controller usually can be expressed as u(k) = Kp e(k) + K

33、i Σe(i) + Kd ec(k) （4） In fuzzy reasoning, the input variables are e and ec, and the output variables are ΔKp, ΔKi, Δ

34、Kd. The PID control parameters are given as Kp= Kp0+ΔKp ，Ki= Ki0+ΔKi，Kd=Kd0+ΔKd （5） Where, Kp0, Ki0, Kd0 are the initial setting values of Kd, Ki, Kd PID control parameters. B. The Rule Tables of Fuzzy Control

35、From the characteristics of PID control to know, the strong integral action I leads to big overshoot, fast response; the strong derivative action D has good stability, small overshoot, small ability of anti- interfere

36、nce. According to e and ec at different stages, the setting principle of PID control parameters are given [11,18] ： (1) When the controlled variables are close to setting values, the integral action with the same sign

37、 as ec can avoids overshoot and oscillation, and is beneficial to the control. When the controlled variables are far from setting values, the integral action with the opposite sign as e can reduce overshoot and avoid

38、 oscillation. (2) At initial stage of PID parameter adjustment , it can avoid overshoot and increase response speed that Kp is properly large and Ki is small or zero; At middle stage, let the values of Kp and Ki be m

39、oderate while taking account to stability and control precision; At last stage, it can eliminate the errors and reduce overshoot that Kp is reduced and Ki is increased properly. (3) Differential coefficient Kd can in

40、hibit change of controlled variable, shorten steady time, reduce steady-state error. It is a supplement to Kp, Ki. The fuzzy subsets of input and output variables of fuzzy controller are respectively e, ec, ΔKp, ΔKi,

41、ΔKd. The language values of these fuzzy variables are in {NB, NM, NS, ZO, PS, PM, PB}, and the membership functions of these fuzzy variables are all sensitively trigonometric functions with the value field in [-3, +

42、3]. By fuzzy reasoning and test modification based on the above setting principle, the fuzzy control rules of the self-tuning parameter fuzzy-PID control system are obtained and shown in Table III. TALLE II FUZZY CO