Journal of Sustainable Bioenergy Systems, 2013, 3, 93-98

doi:10.4236/jsbs.2013.31013 Published Online March 2013 (http://www.scirp.org/journal/jsbs)

Study on Fuzzy Self-Adaptive PID Control

System of Biomass Boiler Drum Water*
Junran Jin1, Hengshuo Huang2, Junman Sun1, Yongchao Pang1
1
College of Electric and Information Engineering, Zhengzhou University of Light Industry, Zhengzhou, China
2
Zhumadian Power Supply Company, Zhumadian, China
Email: [email protected]

Received February 2, 2013; revised March 7, 2013; accepted March 18, 2013

ABSTRACT
The paper discusses the features of the Biomass Boiler drum water level. Conventional PID Control System can not
reach a satisfaction result in nonlinearity and time different from Biomass Boiler Drum Water Control System. In this
study, a kind of fuzzy self-adaptive PID controller is described and this controller is used in biomass boiler’s drum wa-
ter level control system. Using the simulink tool of MATLAB simulation software to simulate the fuzzy adaptive PID
and conventional PID control system, the result of the comparison shows that the fuzzy self-adaptive PID has the strong
anti-jamming, flexibility and adaptability as well as the higher control precision in Biomass Boiler Drum Water.

Keywords: Biomass Boiler; Fuzzy Adaptive PID; Fuzzy Control

1. Introduction volume is also high. The boiler will explode if it isn’t

controlled in time [3]. If the water level is too high, it
The biomass energy is the energy that can be transformed
will influence the steam water separation. Then steam
solar energy into chemical energy and stored in the or-
with water makes superheated fouling and destroys it. If
ganisms internal with the help of the green plant photo-
steam with water is the energy of the turbine, superheated
synthesis directly or indirectly. Modern biomass is seen
steam drops dramatically so that turbine blades are dam-
as one of the most promising renewable energy sources
aged and it causes the problems of safety and economy. So
in the near future. Using biomass to generate energy car-
it is not allowed that the water level is too low or too high.
ries like heat, electricity and gaseous and liquid fuels can
Because many confounding factors, such as biomass fuels’
contribute significantly to the reduction of greenhouse
humidity, density, type, as well as the change of boiler load
gas emissions, as shown in many studies and project in
are uncertain, time-varying, non-linear, biomass boiler com-
most scenarios about potential development of energy
bustion system is actually a fluctuation complex system,
and economic systems [1]. Biomass boiler can improve
showing strong coupling, nonlinear, large inertia, time va-
the utilization of biomass for centralized large-scale use
riation parameters and uncertainties. And thus it is very dif-
of biomass energy, adopting advanced combustion tech-
ficult to establish a precise mathematical model. Because
nology to transform the biomass into the boiler fuel [2].
the traditional PID control can not meet the control preci-
Development of biomass boiler technology can save coal
sion and the corresponding speed requirements, fuzzy adap-
and other fossil fuels which will help alleviate the energy
tive PID control theory is a new theory combining tradi-
crisis and meet the national energy saving policy.
tional PID control and fuzzy logic, which does not require
In the biomass boiler combustion system, the boiler
the exact mathematical model of controlled object and adapt
drum water level is an important parameter and it is a
to the fast, small overshoot, and the short transition. This
sign to measure whether boiler steaming-water system is
control method is suitable for biomass boiler drum water
balance. Maintaining the boiler drum water level in the
level control system [4].
specified range is a necessary condition and one of the
important indexes for the boiler safe operation. If the
2. The Design of Fuzzy Controller
water level is too low, the water of bubbles lessens at the
same time the load largeness. The results are the vapori- 2.1. The Control Difficulties of Biomass Boiler
zation speed is very fast and the change rate of water Drum Water Level
*
Supported by Foundation of Henan Educational Committee (2009B51 Biomass boiler drum water level is important monitoring
0017). parameters in the boiler operation. It indirectly reflects

Copyright © 2013 SciRes. JSBS

94 J. R. JIN ET AL.

the equilibrium relation between boiler steam load and

Fuzzy inference
water inflow. Keeping the right level of the steam pocket
is the essential condition to ensure the safety of boiler
and steam turbine. Because of the change of the steam Kp Ki Kd
load and the feeding water pressure, there is “false water Rin E(t) Yout
level” phenomenon appearing in the boiler. In the opera- PID
tion, when steam load increases, evaporation is greater Controller Object
than the feed water. Water level rises rapidly. It caused de/dt
by steam volume in the steam water mixture increasing Ec(t)
rapidly. This phenomenon called “false liquid level”.
When the steam load increases sharply, the water level
drops after rising for a while. When the steam load de- Figure 1. The structure of fuzzy self-adjusting PID control-
creases sharply, the water level rises after dropping for a ler.
while. “False liquid level” phenomenon appears when
this controlled variable is disturbance. If the feed water analects for the output of Kp is [−0.3, 0.3] and the range of
using single loop system which viewed water level as fuzzy analects for the output of Kd is [−0.3, 0.3], which are
controlled variable, the adjuster will opposite the direc- shown in Figure 3; the range of fuzzy analects for the out-
tion of the water flow and the direction of the change in put of Ki is [−0.06, 0.06], which is shown in Figure 4. The
load [5]. Consequently, it enlarges the imbalance be- analects linguistic values of input/output linguistic variable
tween input flow and output flow. are PB, PM, PS, Z0, NS, NM, NB. The following figures are
Conventional PID control can’t solve the “false liquid the membership function of linguistic variables’ definition.
level” phenomenon well. However, if the fuzzy control
arithmetic is put into the “false liquid level”, the water 2.4. The Control Rules of Fuzzy Controller
level of steam boiler can be controlled well. The three parameters of PID arithmetic will influence
stability, response speed, overshoot and steady precision
2.2. The Composition of Fuzzy Self-Adjusting of the system [7]. In the process operation system, the
PID Controller auto modification rules for E (different error), Ec (error
Fuzzy self-adjusting PID controller is composed of ad- rate), Kp, Ki and Kd are following items.
justable PID controller and fuzzy controller. The core is 1) When E is relatively big, we should take the bigger
fuzzy controller. It contains fuzzification, repository, Kp and the smaller Kd to make sure that the system has a
fuzzy inference, defuzzification, and input/output quanti- good tracking performance. We should also strictly take
fication and so on. Fuzzy logic is a rule which can map a an integral action to avoid the system has a larger over-
space-input to another space-output. In engineering ap- shoot. Usually take Ki = 0;
plication, fuzzy logic has the following characters: 1) 2) When E and Ec is quite suitable, we should make
Fuzzy logic is flexible; 2) Fuzzy logic is based on natural the Kp smaller to make sure the system has a smaller
language, and the requirement for intensive reading of overshoot. On this occasion, we should take the suitable
data is not very high; 3) Fuzzy logic can take full advan- Ki while Kd has important influence on system;
tage of expert information; 4) Fuzzy logic is easy to 3) When E is very small, we should take the bigger Kp
combine with traditional control technique [6]. and the bigger Ki to make sure that the system has a sta-
Fuzzy self-adjusting PID controller takes E (the error ble performance. We should also take a suitable Kd which
between feedback value and desired value of controlled is based on Ec to avoid the system oscillation appeared in
station) and Ec (error rate) as input. Using fuzzy reason- set value. When Ec is small, take a bigger Kd; When Ec is
ing method it adjusts the PID parameters (Kp, Ki, Kd) big, take a smaller Kd. We make some control-rule fig-
which can meet the requirements of E and Ec for PID ures in the following tables according to the setting prin-
parameter self-setting in different time. Change the PID ciple and the simulation test.
parameters on line by using the fuzzy rules; these func- It needs you to transform the output energy of fuzzy
tions form the self-setting fuzzy PID controller. Its con- controller into accurate quantity after the design of con-
trol system architecture is shown in Figure 1.
trol rules. Then the data can be sent to control actuator.
The design realized the organization of fuzzy quantity by
2.3. Membership Function of Linguistic Variable
using of method of weighted mean [8]. The Tables 1-3
According to the precision and control requirements, it is are showing the fuzzy control rules of Kp .In the online
appropriate that 7 levels are usually selected. The range operation, it works out the deviation and the deviation
of fuzzy analects for input variable E, Ec is usually [−3, rate in current time by constantly testing the output data of
3], which which is shown in Figure 2; the range of fuzzy the system. Then fuzzifier them and get E, Ec, you can get

Copyright © 2013 SciRes. JSBS

 J. R. JIN ET AL. 95

Table 1. Kp fuzzy control rules.

E
Ec
NB NM NS ZO PS PM PB

NB PB PB PM PM PS ZO ZO

NM PB PB PM PS PS ZO NS

NS PM PM PM PS ZO NS NS

ZO PM PM PS ZO NS NM NM

PS PS PS ZO NS NS NM NM

PM PS ZO NS NM NM NM NB

PB ZO ZO NM NM NM NB NB

Table 2. Ki fuzzy control rules.

E
Ec
NB NM NS ZO PS PM PB

NB NB NB NM NM NS ZO ZO

NM NB NB NM NS NS ZO ZO

NS NB NM NS NS ZO PS PS

ZO NM NM NS ZO PS PM PM

PS NM NS ZO PS PS PM PB

PM ZO ZO PS PS PM PB PB

PB ZO ZO PS PM PM PB PB

Table 3. Ki fuzzy control rules.

E
Ec
NB NM NS ZO PS PM PB

NB PS NS NB NB NB NM PS

NM PS NS NB NM NM NS ZO

NS ZO NS NM NM NS NS ZO

ZO ZO NS NS NS NS NS ZO

PS ZO ZO ZO ZO ZO ZO ZO

PM PB NS PS PS PS PS PB

PB PB PM PM PS PS PS PB

the adjusted quantity of the three parameters through con- using Simulink, Simulink extras, fuzzy logic and tool-
trol rules. All these steps are the adjustments to the con- box.
troller parameters.
3.1. The Editor of Fuzzy Controller
3. Simulation Research In MATLAB, a new FIS file is made by the FIS editor to
The model is constructed in the environment of Matlab/ confirm the structure and membership of fuzzy controller.
Simulink, which is composed of fuzzy controller, PID As shown in Figure 5.
control module, controlled member and input/output, Input the membership functions and fuzzy rule of E,

Copyright © 2013 SciRes. JSBS

96 J. R. JIN ET AL.

Figure 2. E, Ec membership function curve.

Figure 5. FIS editor window.

Figure 3. Kp, Kd membership function curve.

Figure 6. PID controlle rblock diagram.

10
G s  , justification factor Ke = 0.9, Kec =
2s 2  3s  1
Figure 4. Ki membership function curve. 0.1. Defuzzification factor K1 = 3, K2 = 1.2, K3 = 0.01,
PID initial value Kp = 4, Ki = 3, Kd = 1.5.
Ec, Kp, Ki, Kd in turns. And “and method” is regarded as
min, “or method” as max, “implication” as min, “aggre- 4. Analysis of Simulation Results
gation” as max and “defuzzification” as centroid. The file
In this section, the step response simulations by using
is saved and loaded into workspace when you want to
fuzzy PID control and PID control respectively. Further-
simulate the module.
more, authors make some comparisons between them.
The simulation results are shown by Figures 6-9.
3.2. The Simulation of Fuzzy Self-Adjusting PID
It can be seen that the fuzzy adaptive PID controller
Controller (Figure 6)
makes the speed of response much quicker, the overshoot
Select a boiler drum as controlled object. And the trans- much smaller, and the oscillation time much shorter such
fer function of feeding water flow and water level is that the states soon reach the level of stability than that of

Copyright © 2013 SciRes. JSBS

 J. R. JIN ET AL. 97

Figure 7. Fuzzy adaptive PID controlle rblock diagram.

the PID controller. In a word, the fuzzy adaptive PID

controller shows good dynamic and static performance
indicators. The fuzzy adaptive PID control algorithm merges
the advantages of the PID control and the fuzzy control,
and it provides a good control method for the complex
systems and the systems with higher demand dynamic
process.

REFERENCES
[1] Demibas, A. (2001) Biomass resource facilities and bio-
mass conversion processing for fuels and chemical. Ener-
gy Conversion and Management, 42, 1357-1378
Figure 8. The simulation wave forms by using the PID con- [2] Mahlia, T.M.I., Abdulmuin, M.Z. and Almsyah, T.M.I. (2003)
trol. Dynamic modeling and simulation of a palm wastes boi-
ler. Energy, 2003, 1235-1256.
[3] Zhu, S.H. and Li, Z. (2010) Study on control of water
level in boiler based on fuzzy self-tuning PID. Journal of
Shenyang Institute of Engineering (Natural Science), 6.
[4] Li, J. and Wang, D.Q. (2011) Design of fuzzy PID con-
troler and its simulation based on MATLAB. Industrial
Control Computer, 24.
[5] Wu, Q.E., Sun, J.M., Wang, J.F., et al. (2010) Fuzzy auto-
mata system with application. Journal of Computation
and Theoretical Nan Science, 10, 1911-1920.
[6] Sun, J.M., Xie, Z.H., Li, L.P. and Feng, G.J. (2010) Si-
mulation research on fuzzy self-adaptive PID control
systerm of biomass boiler. Boiler Technology, 41.
[7] Moon, U. and Lee, K.Y. (2003) A boiler-turbine system
Figure 9. The simulation wave forms by using the fuzzy control using a fuzzy auto-regressive moving average mo-
adaptive PID control. del. IEEE Transaction on Energy Conversion, 18, 165-

Copyright © 2013 SciRes. JSBS

98 J. R. JIN ET AL.

172. boiler. Journal of YUEYANG Vocational Technical Col-

[8] Chen, X., Ding, Y.J. and Rong J. (2011) The design and lege, 26.
simulation of fuzzy PID algorithm of temperature of

Copyright © 2013 SciRes. JSBS