Hindawi Publishing Corporation

Mathematical Problems in Engineering

Volume 2015, Article ID 120975, 9 pages
http://dx.doi.org/10.1155/2015/120975

Research Article
Dynamic Modeling of Steam Condenser and Design of
PI Controller Based on Grey Wolf Optimizer

Shu-Xia Li1 and Jie-Sheng Wang1,2

1
School of Electronic and Information Engineering, University of Science & Technology Liaoning, Anshan 114044, China
2
National Financial Security and System Equipment Engineering Research Center, University of Science & Technology Liaoning,
Anshan 114044, China

Correspondence should be addressed to Jie-Sheng Wang; wang [email protected]

Received 15 September 2015; Revised 16 November 2015; Accepted 18 November 2015

Academic Editor: Salvatore Alfonzetti

Copyright © 2015 S.-X. Li and J.-S. Wang. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.

Shell-and-tube condenser is a heat exchanger for cooling steam with high temperature and pressure, which is one of the main
kinds of heat exchange equipment in thermal, nuclear, and marine power plant. Based on the lumped parameter modeling
method, the dynamic mathematical model of the simplified steam condenser is established. Then, the pressure PI control system
of steam condenser based on the Matlab/Simulink simulation platform is designed. In order to obtain better performance, a new
metaheuristic intelligent algorithm, grey wolf optimizer (GWO), is used to realize the fine-tuning of PI controller parameters.
On the other hand, the Z-N engineering tuning method, genetic algorithm, and particle swarm algorithm are adopted for tuning
PI controller parameters and compared with GWO algorithm. Simulation results show that GWO algorithm has better control
performance than other four algorithms.

1. Introduction Therefore, through the computer simulation experi-

ments, the establishment of the dynamic model and under-
The condenser is one of the important kinds of equipment standing the dynamic characteristics of the condenser have
in thermal power plant, nuclear power plants, and marine a great significance on improving the safety and economic
power plant. The reliability of condenser running directly operation level of the steam condenser [3]. A hybrid modeling
affects the safety and economic operation of the entire power approach is proposed to describe the dynamic behavior of
plant or power system [1]. A steam condenser is a piece of the two-phase flow condensers used in air-conditioning and
machinery that turns steam into water. Many steam-based refrigeration systems based on fundamental energy and mass
systems use a circuit of water to maximize their efficiency. balance governing equations and thermodynamic principles.
Water is heated into steam, the steam provides motivation for The model validation studies on an experimental system
a process, a steam condenser turns it back into water, and the show that the model predicts the system dynamic well [4].
cycle begins again [2]. The failure of the condenser may cause A method to improve the load change capacity is proposed
the boiler or steam turbine unit to overheat, which endangers for the water cooled power plants through controlling the
the safety of the whole generating unit or power plant. cooling water flow. Then, the CCWCS (condenser cooling
The power plant has strict requirements on the reliability water control system) is put forward to execute this method
and the tightness of the condenser. In addition to safety on the premise of unit safety [5]. A robust strategy for online
considerations, the condensation process of steam in the fault detection and optimal control of condenser cooling
condenser is an important part of the system thermodynamic water systems is proposed. The optimal control strategy is for-
cycle, which greatly affects the economic performance of the mulated using a model-based approach, in which simplified
system. models and a hybrid quick search (HQS) method are used to
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 Mathematical Problems in Engineering

optimize the performance of the overall system by changing

Cooling water inlet Steam inlet
the settings of the local process controllers [6]. In the field
of computer simulation, the simulation technology is used
to simulate the operation of the condenser system and study
its working performance. The optimization of the parameters
of PID controller can not only satisfy the static accuracy but
also make the system have better stabilization. The accurate
mathematical model can accurately and comprehensively Cooling Condensate
represent all working conditions in the real running process water outlet water outlet
of steam condenser system (normal conditions, fault condi- Figure 1: Structure diagram of shell-and-tube condenser.
tions, and transient conditions) and make them reappear. On
the same time, it can be used to carry through the research
on the dynamical characteristic to meet the requirements of
different engineering projects. GWO algorithm is able to greatly reduce the probability
At present, people by simulating biotic population and of producing premature phenomenon and falling into local
their evolution process in nature have developed a variety of optimum.
intelligence algorithms: particle swarm optimization (PSO) Aiming at the steam condenser pressure control problem,
algorithm, ant colony optimization (ACO) algorithm, artifi- based on the Matlab/Simulink simulation platform and the
cial bee colony (ABC) algorithm, shuffled frog leaping (SFL) established mathematical model, a closed-loop condenser
algorithm, cuckoo search (CS) algorithm, Dolphin partner pressure control system is designed. The GWO algorithm is
optimization (DPO), and Firefly algorithm (FA). adopted to optimize the parameters of the PI controller. Sim-
GA is the most typical algorithm of evolutionary branch. ulation results show the effectiveness of the proposed control
This algorithm simulates Darwinian evolution concepts [7]. strategy. The paper is organized as follows. In Section 2, the
Each new population is created by the combination and technique and dynamic modeling of the shell-and-tube steam
mutation of the individuals in the previous generation. condenser are introduced. The GWO algorithm is illustrated
Particle swarm optimization (PSO) was inspired from the in Section 3. The parameter optimization of PI controller
social behavior of birds flocking [8]. The PSO algorithm based on GWO algorithm is described and the simulation
employs multiple particles that chase the position of the best experiments and results analysis are introduced in detail in
particle and their own best positions obtained so far. Artificial Section 4. Finally, the conclusion illustrates the last part.
bee colony (ABC) algorithm mimics the collective behavior
of bees in finding food sources [9]. There are three types 2. Dynamic Modeling of Steam Condenser
of bees in ABC: scout, onlooker, and employed bees. The
scout bees are responsible for exploring the search space, 2.1. Structure Characteristics and Working Principle of Steam
whereas onlooker and employed bees exploit the promising Condenser. The shell of shell-and-tube type condenser is
solutions found by scout bees. Ant colony optimization usually cylindrical or elliptical as shown in Figure 1, which is
(ACO) algorithm was inspired by the social behavior of ants connected with end closures for constituting the water cham-
in an ant colony. In fact, the social intelligence of ants in bers. Between the end closures and the shells, a perforated
finding the shortest path between the nest and a source of tube plate is fixed, in which a lot of cooling water pipes are
food is the main inspiration of ACO [10]. Cuckoo search (CS) arranged hierarchically. The entrance pipe of steam is located
algorithm was put forward by Yang and Deb in 2009 [11]. in the upper part of the condenser shell, which is directly or
This algorithm is mainly based on two aspects: the cuckoo’s indirectly connected with the exhaust equipment through the
parasitic reproduction mechanism and Levy flights search compensator. In the lower part of the shell, there is a gathering
principle. Assume each cuckoo only lays one egg at a time tank (or a hot well water tank) of the condensed water. The air
and randomly chooses bird’s nest to hatch the egg. outlet port is positioned at the lower part of shell and the air
The grey wolf optimizer (GWO) as a new swarm intel- is drawn through this nozzle.
ligent algorithm is put forward by Seyedali, Mirjalili, and so The working principle of the steam condenser is shown in
forth in 2014, which mainly mimics wolf leadership hierarchy Figure 2.
and hunting mechanism in nature. Seyedali and Mirjalili, and Steam goes into steam field of the condenser through
so forth, have proved that the search performance of basic steam admission pipe. Steam gets in touch with the condenser
wolf algorithm is superior to that of PSO, GSA, DE, and FEP tube wall to begin the radiate condensation; at the same time,
algorithm. Due to the wolves’ algorithm with the advantages the latent heat is transferred to the cooling water through the
of being simple in principle, having fast seeking speed and surface of the cooling water pipe. Cooling water with inlet
high search precision, and being easy to realize, it is more temperature is fed into water chamber through the cooling
easily combined with the practical engineering problems. water pipe, where the cooling water is assigned to all pipes
Many kinds of swarm intelligence are inspired by hunt- of the first procedure in the lower part of the condenser
ing and search behaviors of a population. However, GWO shell. The cooling water flows into another water chamber
algorithm simulates internal leadership hierarchy of wolves; along the first six cooling water pipes and then enters the
thus, in the searching process, the position of best solution next flow pipes and carries through heat exchange with
can be comprehensively assessed by three solutions. So the the steam. Through such several procedures in return, the
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Mathematical Problems in Engineering 3

Steam inlet Gh1 Th1 enthalpy of saturated water corresponding to the condenser
pressure (kJ/kg).

Cooling Cooling (ii) Vapor pressure equation is as follows:

water inlet water outlet
𝑑𝑃𝑠 𝑅𝑠 (𝑑𝐺𝑠 /𝑑𝑡)
Gc Tc1 Gc Tc2 = (𝑇𝑠 + 273.15) , (3)
𝑑𝑡 𝑉
where 𝑃𝑠 is the internal steam pressure of condenser (Pa),
Condensate water outlet Gh2 Th2 𝑅𝑠 is the steam gas constant (0.4615 kJ/(kgK)), 𝑉 is the
space volume of gas in the condenser (m3 ), and 𝑇𝑠 is the
Figure 2: Schematic diagram of condenser. temperature of saturated gas (∘ C).

(iii) Average enthalpy of steam in the condenser is as follows:

cooling water with the outlet temperature is discharged from 𝑑𝐺𝑠 𝐻𝑠

= 𝐺st × 𝐻st + 𝐺ost × 𝐺ost − (𝐺𝑐 + 𝐺ss ) × 𝐻𝑠 , (4)
the outlet pipes. Due to the lack of system sealing property, 𝑑𝑡
the air is drawn out from the condenser constantly to ensure where 𝐻𝑠 is the average enthalpy of steam (kJ/kg), 𝐻st is the
the requirements of the system’s vacuum degree. The drawn enthalpy of steam turbine exhaust (kJ/kg), and 𝐻ost is the
gas contains the air and steam. In the beginning of the other inlet enthalpy (kJ/kg).
condensation, the air volume is very smaller than the total
amount of steam. With the steam and air flowing toward the (2) Air Zone
exhaust port, steam is continuously condensed down. Then,
the steam quality in the mixture gradually decreases. On the (i) Air mass equation is as follows:
contrary, the relative content of the air increases gradually.
Until the relative content of air fed into the cooling zone air 𝑑𝐺𝑎
= 𝐺vb + 𝐺𝑛 + 𝐺𝑔 − 𝐺𝑎 , (5)
has reached a great numerical extent, the steam condensation 𝑑𝑡
process terminates. where 𝐺vb is the air quantity of the condenser from the
vacuum break valve (kg/s), 𝐺𝑛 is the air volume of the normal
2.2. Mathematical Model drain condenser (kg/s), 𝐺𝑔 is the air amount from the seal
leakage of the condenser (kg/s), and 𝐺𝑎 is the air quantity
2.2.1. Mathematical Model of Shell Side of Condenser from air extractor (kg/s).

(1) Steam Zone (ii) Air pressure equation is as follows:

(i) Steam mass equation is as follows: 𝑑𝑃𝑎 𝑅𝑎 (𝑑𝐺𝑎 /𝑑𝑡)

= (𝑇𝑠 + 273.15) , (6)
𝑑𝑡 𝑉
𝑑𝐺𝑠
= 𝐺st + 𝐺ost − 𝐺𝑐 − 𝐺ss , (1) where 𝑃𝑎 is the air pressure in the condenser (Pa) and 𝑅𝑎 is
𝑑𝑡 the gas constant of the air (0.287 kJ/(kgK)).
where 𝐺𝑠 is the steam content in the shell side of the (iii) Absolute pressure of the condenser is as follows:
condenser (kg/s), 𝐺st is the exhaust volume of steam turbine
(kg/s), 𝐺ost is the other steam inlet of the condenser (kg/s), 𝐺𝑐 𝑃𝑐 = 𝑃𝑠 + 𝑃𝑎 , (7)
is the main steam condensate (kg/s), and 𝐺ss is the amount of
steam drawn out by vacuum pumping equipment (kg/s). where 𝑃𝑐 is the absolute pressure in the condenser (Pa).
𝐺ss and 𝐺𝑐 are calculated by the following equations:
(3) Hot Water Area
𝐺ss = 𝐺ao (1 − 𝑅) , (i) Hot well water level equation is as follows:
𝑀𝑎 𝑃𝑎 𝑅𝑠 𝐺𝑊
𝑅= = , 𝐿𝑐 = , (8)
𝑀𝑠 + 𝑀𝑎 𝑃𝑠 𝑅𝑎 + 𝑃𝑎 𝑅𝑠 𝜌𝐴 𝑤
(2)
𝑄𝑐 = 𝐺𝑐 (𝐻𝑠 − 𝐻cw ) , where 𝐿 𝑐 is the hot well water level (m), 𝐺𝑊 is the hot well
water quality (kg/s), 𝜌 is the hot well water density (kg/m3 ),
𝑈𝐴Δ𝑡𝑚
𝐺𝑐 = , and 𝐴 𝑤 is the hot well cross-sectional area (m2 ).
(𝐻𝑐 − 𝐻cw )
(ii) Hot water quality equation is as follows:
where 𝐺ao is the quality of the gas mixture pumped by the
pumping unit (kg/s), 𝑅 is the share air ratio in the condenser, 𝑑𝐺𝑤
= 𝐺𝑐 + 𝐺gp − 𝐺wo , (9)
𝐻𝑠 is the average enthalpy of steam (kJ/kg), and 𝐻cw is the 𝑑𝑡
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 Mathematical Problems in Engineering

where type 𝐺gp is the bubbling oxygen exhaust volume (kg/s) X∗ − X

and 𝐺wo is the condenser water outlet quantity (kg/s). (X∗ , Y)
(X∗ − X, Y)
(iii) Enthalpy equation of hot well water is as follows:
(X, Y)
𝑑𝐺𝑤 𝐻𝑤
= 𝐺𝑐 ∗ 𝐻cw + 𝐺gp ∗ 𝐻gp − 𝐺wo ∗ 𝐻𝑤 , (10)

Y∗ − Y
𝑑𝑡
(X∗ , Y∗ )
where 𝐻gp is the bubbling oxygen exhaust steam enthalpy
(kJ/kg), 𝐻𝑤 is the enthalpy of hot well water (kJ/kg), and (X∗ − X, Y∗ )
𝐻cw is the enthalpy of saturated water corresponding to the (X, Y∗ )
condenser pressure (kJ/kg).

2.2.2. Mathematical Model of Condenser Tube Side. The

dynamic heat balance equation of the circulating water is
described as follows:
(X∗ − X, Y∗ − Y) (X, Y∗ − Y)
𝑑𝑇 ∗ ∗
(X , Y − Y)
𝑀𝑤 𝐶𝑤 2 = 𝑄 − 𝑄𝑤 = 𝑈𝐴Δ𝑡𝑚 − 𝐹cw 𝐶𝑝 (𝑇 − 𝑇cw ) , (11)
𝑑𝑡
Figure 3: Position vector of wolf and next moving position in two-
where 𝑀𝑤 is the circulating water quality (kg), 𝐶𝑝 is the dimensional space.
circulating water heat capacity (kJ/(kg ∗ C)), 𝑄 is the steam
outlet heat (kJ), 𝑄𝑤 is the circulating water heat absorption
quantity (kJ), 𝑈 is the condenser heat transfer coefficient to the principle of the wolves hunting, the hunt process is
(W/(m2 ∗ ∘ C)), Δ𝑡𝑚 is the logarithmic mean temperature divided into three stages: searching prey, surrounding prey,
difference (∘ C), 𝐴 is the condenser heat transfer area (m2 ), and attacking prey. In the four groups of wolves, 𝛼, 𝛽, and 𝛿
𝐹cw is the circulating water flow (kg/s), 𝑇cw is the circulating are seen as the three best wolves; they guide the other wolf (Ω)
water inlet temperature (∘ C), and 𝑇 is the circulating water trending in the search space in the best region. In the iterative
outlet temperature (∘ C). searching process, 𝛼, 𝛽, 𝛿, and Ω wolves are used to realize
The logarithmic heat transfer temperature difference is the assessment of prey possible positions in the optimization
calculated by process. The positions of wolves are updated in accordance
with the following equations:
𝑇 − 𝑇cw
Δ𝑡𝑚 = . (12) 󵄨󵄨 󳨀󳨀→ 󵄨󵄨
ln ((𝑇𝑠 − 𝑇cw ) / (𝑇𝑠 − 𝑇)) 𝐷⃗ = 󵄨󵄨󵄨𝐶⃗ ⋅ 𝑋𝑃 (𝑡) − 𝑋⃗ (𝑡)󵄨󵄨󵄨 ,
󵄨 󵄨
(15)
The overcooling of condenser is calculated by 󳨀󳨀→
𝑋⃗ (𝑡 + 1) = 𝑋𝑃 (𝑡) − 𝐴⃗ ⋅ 𝐷,⃗
Δ𝑡𝑤 = 𝑇𝑐 − 𝑇𝑤 , (13)
where 𝑡 is the current iteration number, 𝐴⃗ and 𝐶⃗ are the
󳨀󳨀→
where 𝑇𝑐 is the saturated water temperature of the vapor coefficient vector, 𝑋𝑃 is the position vector of the prey, and 𝑋⃗
pressure in the condenser (∘ C) and 𝑇𝑤 is the condenser hot is the position of the wolf. The vectors 𝐴⃗ and 𝐶⃗ are expressed
well water temperature (∘ C). as follows:
The heat transfer error of the condenser is calculated by
𝐴⃗ = 2𝑎 ⋅ →
󳨀
𝑟1 − 𝑎,⃗
𝛿𝑡 = 𝑇𝑠 − 𝑇, (14) (16)
𝐶⃗ = 2 ⋅ →
󳨀
𝑟2 ,
where 𝑇𝑠 is the saturated gas temperature corresponding to
saturation pressure in condenser (∘ C). where the coefficient 𝑎⃗ linearly increases from 2 to 0 with the
increase of the iteration number and→ 𝑟1 and→
󳨀 󳨀
𝑟2 are the random
vector located in the scope [0, 1].
3. Grey Wolf Optimizer The principle and concept of the position update (15) are
The grey wolf optimizer is a new metaheuristic intelligent described in Figure 3 [7].
algorithm proposed by Mirjalili et al. in 2014 [12], which is It can be seen from Figure 3 that the wolf in the position
successfully applied in many fields, such as security smart (𝑋, 𝑌) can be arranged a new location on the basis of the
grid power system management [13], parameter estimation above formula. While Figure 3 shows only the 7 possible
[14], reactive power dispatch problem [15], flow shop schedul- positions of the wolf, the randomly adjusting parameters 𝐴⃗
ing problem [16], combined heat and power dispatch [17], and and 𝐶⃗ can make the wolf move to anywhere in a continuous
automatic control [18]. For immigrating the wolves’ internal space. In GWO algorithm, the positions of wolves 𝛼, 𝛽, and
leadership hierarchy, the wolves are divided into four types: 𝛿 are likely the prey (optimal) position. In the searching
alpha (𝛼), beta (𝛽), delta (𝛿), and omega (Ω). According process, the previous three best solutions are assumed to be
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Mathematical Problems in Engineering 5

Alpha

Beta
Explored
Unexplored

| ≥1
1 If | A
D𝛼 If | A| <

D𝛽

Move
Explored
Unexplored

D𝛿

Figure 5: Exploration and development of wolves.

Omega
Delta

Figure 4: Sketch map of position update of the wolves.

It can be seen from Figure 4 that the random and adaptive
vectors 𝐴⃗ and 𝐶⃗ can be used to balance the exploration
𝛼, 𝛽, and 𝛿, and then the others are regarded as the Ω wolves. and development capabilities of the GWO algorithm. When
The positions of 𝛼, 𝛽, and 𝛿 are used to update their positions. |𝐴|⃗ > 1, the wolf has detection ability. On the other hand,
The following mathematical formulae are used to adjust the when the value of vector 𝐶⃗ is greater than 1, it can also
position of Ω wolf again, and the position update schematic promote the enhancement of the detection ability of the wolf.
graph is shown in Figures 4 and 5: In contrast, when |𝐴|⃗ < 1 and 𝐶 < 1, the wolf ’s information
mining capacity is enhanced. In order to enhance the ability
󳨀→ 󵄨󵄨󵄨󳨀 → 󳨀→ 󵄨󵄨
𝐷𝛼 = 󵄨󵄨𝐶1 ⋅ 𝑋𝛼 − 𝑋⃗ 󵄨󵄨󵄨 , of the wolf, with the increase of the iteration number, 𝐴⃗ is
󵄨 󵄨
decreased linearly. However, 𝐶⃗ is randomly generated in the
󳨀→ 󵄨󵄨󳨀󵄨 → 󳨀→ 󵄨󵄨
𝐷𝛽 = 󵄨󵄨𝐶2 ⋅ 𝑋𝛽 − 𝑋⃗ 󵄨󵄨󵄨 , (17) whole optimization process, which can make the detection
󵄨 󵄨 and exploitation ability of the wolf to reach equilibrium at any
󳨀→ 󵄨󵄨󵄨󳨀 → 󳨀→ 󵄨⃗ 󵄨󵄨 stage, especially in the final stage of the iteration, and prevent
𝐷𝛿 = 󵄨󵄨𝐶3 ⋅ 𝑋𝛿 − 𝑋󵄨󵄨 ,
󵄨 󵄨 the algorithm from falling into local optimum.
The procedure of the GWO algorithm is described as
󳨀→ 󳨀→ 󳨀→
where 𝑋𝛼 , 𝑋𝛽 , and 𝑋𝛿 are the position of the wolves 𝛼, 𝛽, follows.
→ 󳨀
󳨀 → 󳨀
→
and 𝛿, respectively, 𝐶1 , 𝐶2 and 𝐶3 are random vectors, and 𝑋⃗ Step 1. Initialize the wolves. Randomly initialize the position
represents the position of the current solution. Equations (17) of the wolves 𝑋𝑖 (𝑖 = 1, 2, . . . , 𝑛) and parameters 𝑎, 𝐴, and 𝐶.
are used to calculate the approximate distance between the
current position and 𝛼, 𝛽, and 𝛿, respectively. After defining Step 2. Calculate the fitness of each wolf, and choose the three
the distance between them, the final position of the current wolves with best fitness as wolves 𝛼, 𝛽, and 𝛿.
solution is calculated by the following:
Step 3. Update positions. Based on (17)–(21), update the
󳨀→ 󳨀→ 󳨀→ 󳨀→
𝑋1 = 𝑋𝛼 − 𝐴 1 ⋅ (𝐷𝛼 ) , (18) positions of the other wolves, that is to say, the positions of
the Ω wolves.
󳨀→ 󳨀→ 󳨀→ 󳨀→
𝑋2 = 𝑋𝛽 − 𝐴 2 ⋅ (𝐷𝛽 ) , (19)
Step 4. Update parameters 𝑎, 𝐴, and 𝐶.
󳨀→ 󳨀→ 󳨀→ 󳨀→
𝑋3 = 𝑋𝛿 − 𝐴 3 ⋅ (𝐷𝛿 ) , (20) Step 5. Judge whether to meet the termination conditions or
not. If satisfied, the position of 𝛼 wolf and the fitness value
󳨀→ 󳨀→ 󳨀→ are the optimal output. If the termination condition is not
⃗ 𝑋 + 𝑋2 + 𝑋3
𝑋 (𝑡 + 1) = 1 , (21) satisfied, return to Step 2.
3
󳨀→ 󳨀→ 󳨀→
where 𝐴 1 , 𝐴 2 , and 𝐴 3 are random vectors and 𝑡 represents 4. Parameter Optimization of PI Controller
the number of iterations. Based on GWO Algorithm
Seen from the above equations, (17) define the step size
of the wolf Ω tending to the wolves 𝛼, 𝛽, and 𝛿. Equations 4.1. Dynamic Model of Steam Condenser Based on Mat-
(19)–(21) define the final position of Ω wolf. lab/Simulink Simulation Software. The establishment of the
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 Mathematical Problems in Engineering

dynamic mathematical model of the condenser is based on Table 1: Parameters of steam condenser.
the following assumption that the total amount of condensa- Parameter Parameter value Unit
tion, the circulating water flow, and the condenser volume are
𝑅 0.461526 kJ/kgK
certain. So it is set up based on the dynamic heat balance and
mass balance of the condenser water. 𝑉 3 m3
𝜆 2265.65 kJ/kg
4.1.1. Dynamic Heat Balance. In the dynamic heat balance, it 𝑈𝐴 356.972 kW/K
is assumed that the total amount of condensation is certain 𝑀cw 6500 kg
and that the input steam and the output condensate are 𝐶𝑝 4.2 kJ/(kgK)
saturated. Therefore, the heat from steam to the circulating 𝛼 0.3162 K/kPa
∘
water and the steam potential heat are equal. So the steam 𝛽 68.0958 C
released heat can be approximated, calculated by the follow- 𝛼1 8.7292e − 2
ing equations: 𝛼2 7.3787e − 4
𝑄 ≈ 𝑈𝐴Δ𝑡𝑚 ,
Table 2: Variables of steam condenser.
𝑇 − 𝑇cw (22)
Δ𝑡𝑚 = . Variable Meaning Variable value Unit
ln ((𝑇𝑠 − 𝑇cw ) / (𝑇𝑠 − 𝑇)) 𝐹𝑠 Steam flow 4 kg/s
The heat transfer coefficient 𝑈 and the heat transfer area 𝐹𝑐 Condensate water flow 4 kg/s
𝐴 can be replaced by the following exponential equation 𝐹cw Cooling water flow 107.8881 kg/s
approximately: 𝑃 Condenser pressure 90 kPa
∘
𝑇 Circulating water outlet temperature 80 C
1
= 𝛼1 𝐹cw −0.8 + 𝛼2 , (23) 𝑇cw Circulating water inlet temperature 60 ∘
C
𝑈𝐴 ∘
𝑇𝑐 Saturated water temperature 96.5538 C
where 𝛼1 and 𝛼2 are constants. 𝑄 Steam heat 9062.6 kW
When 𝐹cw → ∞ , 𝛼2 is determined by 𝑈𝐴. In this case, 𝑈𝐴
is determined by the heat transfer ratio between the steam
and the tube wall and the thermal resistance of the tube wall. are eight variables (𝐹𝑠 , 𝐹𝑐 , 𝐹cw , 𝑃, 𝑇, 𝑇cw , 𝑇𝑐 , and 𝑄) and ten
Thus, by assuming the outlet temperature of the circulating parameters (𝑅, 𝑉, 𝜆, 𝑈𝐴, 𝑀cw , 𝐶𝑝 , 𝛼, 𝛽, 𝛼1 , and 𝛼2 ). The values
water, 𝛼2 can be determined. Based on the above assumptions of ten parameters are shown in Table 1, where 𝛼1 and 𝛼2 are
and equations, the dynamic heat balance equation of the determined under 𝑇 = 90∘ C (𝐹cw → ∞). The values of the
condenser is described as follows: eight variables are shown in Table 2 under the assumption
𝑑𝑇 𝐹 𝑄 that the system is stable.
= cw (𝑇cw − 𝑇) + . (24) Based on the above model equations and the software
𝑑𝑡 𝑀cw 𝑀cw 𝐶𝑝
Matlab/Simulink, a simulation model of PI controller for
condenser pressure closed-loop control system is established
4.1.2. Mass Balance. The mass balance of steam and conden- as shown in Figure 6, which includes a first-order delay unit
sate is based on the assumption that the space 𝑉 is constant, used to represent the actuator with a time constant 𝜏 = 10 (s)
and the volume of the steam and air is constant. That is to and a lag unit caused by the pressure sensor with the time
say, in order to maintain the vapor condensation level of constant 𝜏 = 5 (s).
the condenser (certain vacuum degree), the output flow of
condensate water needs to be controlled in a certain range. So,
in order to simplify the model, we assume that the inlet and 4.2. Encoding and Fitness Function. Because the design of
outlet of the condensate are saturated. Therefore, the ideal gas the PID controller is actually a multidimensional function
model equation is expressed as optimization problem, the GWO algorithm adopts the real
number coding. So, for the parameters optimization of the PI
𝑑𝑃 𝑅𝑇𝑐 controller, each wolf can be directly coded as (𝐾𝑝 , 𝐾𝑖 ):
= (𝐹𝑠 − 𝐹𝑐 ) , (25)
𝑑𝑡 𝑉
𝑋 = {𝐾𝑝 , 𝐾𝑖 } . (27)
where 𝐹𝑠 is the steam flow (kg/s) and 𝐹𝑐 is the condensate
water flow (kg/s).
The control parameter optimization is designed to make
Among them, the condensation water temperature 𝑇𝑐 and
the control error tend to zero and has a faster response speed
the condenser pressure 𝑃 have a unique relationship. In order
and smaller overshoot. So the evaluation of the performance
to simplify the model, it is approximated by the following
of each set of control parameters is good or bad; the integral
linear relationship equation:
of the product of absolute error and time is selected as the
𝑇𝑐 = 𝛼𝑃 + 𝛽. (26) fitness function:
∞
The steam condenser model given above has five equa- ITAE = ∫ 𝑡 |𝑒 (𝑡)| 𝑑𝑡. (28)
tions, among which two are dynamic equations. Here, there 0
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Mathematical Problems in Engineering 7

⟨Fcw ⟩
4 Fs 145
⟨T⟩
Fs
60 Tcw y = ⟨Fcw ⟩ 140
⟨Q⟩
Tcw ⟨P⟩ 135
Fcw Scope
Steam condenser 130

++ PID
−
+ 125

Amplitude
PID controller 120
Input step test Pressure setpoint 115
Reaction curve
PID tuning
110
Click here to tune the PID controller
105
Figure 6: Matlab/Simulink simulation model of PI controller for
100
condenser pressure closed-loop control system.
95
0 10 20 30 40 50
⟨T⟩ Time (s)
80
ZN PSO
GA GWO

Figure 8: Output response curves of circulating water flow under

79.5 different algorithms.
Amplitude

⟨Q⟩
10200
79
10000
9800
9600
78.5 9400
Amplitude

0 10 20 30 40 50
Time (s) 9200
9000
ZN PSO
GA GWO 8800
Figure 7: Output response curves of outlet temperature of cooling 8600
water under different algorithms.
8400
8200
Table 3: Parameters of PI controller. 0 10 20 30 40 50
Time (s)
PID parameters ZN GA PSO GWO
𝐾𝑝 10.8 8.53 7.07 4.47 ZN PSO
GA GWO
𝐾𝑖 4.22 0.91 1.04 0.89
Figure 9: Output response curves of steam heat output under
different algorithms.
4.3. Simulation Experiments and Results Analysis of PI Con-
troller. On the basis of the above established model of steam
condenser, the GWO algorithm is adopted to optimize the As seen from the above simulation results, the PI con-
parameters of the adopted PI controller. The self-tuning troller under the optimization by the proposed GWO algo-
performances are compared with the Z-N engineering tun- rithm has the best control performance, that is to say, small
ing method, genetic algorithm (GA), and particle swarm overshoot and short rise time and adjustment time. The Z-N
optimization (PSO) algorithm. Respectively, run GWO, PSO, engineering self-tuning method has the worst performance,
and GA algorithm 30 times, and then select the best PID where the overshoot is the largest, and the rise time and
parameters of each algorithm. The output response curves adjustment time are the longest. The GWO algorithm can
of cooling water outlet temperature, circulating water flow, effectively improve the system control quality and achieve the
steam discharge heat, and condenser pressure are shown in desired effect.
Figures 7–10. The parameters of PI controllers are listed in Because the Z-N method belongs to engineering setting
Table 3. method, setting the PID parameters depends on experience
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 Mathematical Problems in Engineering

⟨P⟩ Conflict of Interests

90

89 The authors declare that there is no conflict of interests

regarding the publication of this paper.
88

87 Authors’ Contribution
Amplitude

86
Shu-Xia Li participated in the data collection, analysis, algo-
85 rithm simulation, draft writing, and critical revision of this
paper. Jie-Sheng Wang participated in the concept, design,
84
and interpretation and commented on the paper.
83

82 Acknowledgments
81 This work is partially supported by National Key Technolo-
0 10 20 30 40 50
Time (s)
gies R & D Program of China (Grant no. 2014BAF05B01),
Project by National Natural Science Foundation of China
ZN PSO (Grant no. 21576127), Program for Liaoning Excellent Talents
GA GWO in University (Grant no. LR2014008), Project by Liaoning
Figure 10: Output response curves of pressure under different Provincial Natural Science Foundation of China (Grant no.
algorithms. 2014020177), Program for Research Special Foundation of
University of Science and Technology of Liaoning (Grant
no. 2015TD04), and Opening Project of National Financial
Security and System Equipment Engineering Research Cen-
value, so the effect of PID control is poor. The GA algorithm ter (Grant no. USTLKEC201401).
and PSO algorithm are intelligent method, so control effect
of PID controller whose parameters are optimized by these References
two methods has been greatly improved, relative to the Z-N
setting method. But because of their inherent search mode [1] T. Tahri, S. A. Abdul-Wahab, A. Bettahar, M. Douani, H. Al-
and the defects, the search accuracy is not high enough; thus, Hinai, and Y. Al-Mulla, “Simulation of the condenser of the
result of parameters setting is poorer than that of grey wolf seawater greenhouse. Part I: theoretical development,” Journal
optimizer. It can be known from searching way of GWO of Thermal Analysis and Calorimetry, vol. 96, no. 1, pp. 35–42,
2009.
algorithm above that the grey wolf optimizer is going to
search solutions in the comprehensive evaluation by three [2] P. K. Bansal and T. C. Chin, “Modelling and optimisation of
wire-and-tube condenser,” International Journal of Refrigera-
wolves. Therefore, the GWO algorithm has a high search
tion, vol. 26, no. 5, pp. 601–613, 2003.
precision, which can make it sure to search a better solution
value. Thus, the GWO algorithm increases the probability of [3] M. R. Malin, “Modelling flow in an experimental marine
condenser,” International Communications in Heat and Mass
searching a better group of PID parameters.
Transfer, vol. 24, no. 5, pp. 597–608, 1997.
[4] X. Ding, W. Cai, P. Duan, and J. Yan, “Hybrid dynamic modeling
5. Conclusions for two phase flow condensers,” Applied Thermal Engineering,
vol. 62, no. 2, pp. 830–837, 2014.
In this paper, a dynamic model of steam condenser and the [5] W. Wang, D. Zeng, J. Liu, Y. Niu, and C. Cui, “Feasibility analysis
closed-loop control system are established on the basis of the of changing turbine load in power plants using continuous
lumped parameter model of the condenser. For the pressure condenser pressure adjustment,” Energy, vol. 64, pp. 533–540,
closed-loop control system, in order to improve the system 2014.
performance, the GWO algorithm is adopted to optimize the [6] Z. Ma and S. Wang, “Online fault detection and robust control
PI parameters. Through the simulation experiments and per- of condenser cooling water systems in building central chiller
formance comparison of cooling water outlet temperature, plants,” Energy and Buildings, vol. 43, no. 1, pp. 153–165, 2011.
circulating water flow, steam discharge heat, and condenser [7] D. Whitley, “An executable model of a simple genetic algo-
pressure, the introduction of the GWO algorithm makes the rithm,” Foundations of Genetic Algorithms, vol. 2, no. 1519, pp.
steam condenser PI controller have better control effect. In 45–62, 2014.
order to have a better control effect of steam condenser, we [8] J. Kennedy and R. Eberhart, “Particle swarm optimization,”
can try to improve the design of PID controller. Since the in Proceedings of the IEEE International Conference on Neural
research of grey wolf optimizer is just in its infancy, the Networks, pp. 1942–1948, December 1995.
GWO algorithm should be improved to obtain better PID [9] D. Karaboga and B. Basturk, “On the performance of artificial
parameters in the further study. Other superior algorithms bee colony (ABC) algorithm,” Applied Soft Computing, vol. 8,
should be exploited for PID parameter optimization. no. 1, pp. 687–697, 2008.
 2629, 2015, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1155/2015/120975 by GEBZE TEKNIK UNIVERSITESI, Wiley Online Library on [14/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Mathematical Problems in Engineering 9

[10] M. Dorigo, M. Birattari, and T. Stützle, “Ant colony optimiza-

tion,” IEEE Computational Intelligence Magazine, vol. 1, no. 4,
pp. 28–39, 2006.
[11] X.-S. Yang and S. Deb, “Cuckoo search via Lévy flights,” in Pro-
ceedings of the World Congress on Nature & Biologically Inspired
Computing (NABIC ’09), pp. 210–214, IEEE, Coimbatore, India,
December 2009.
[12] S. Mirjalili, S. M. Mirjalili, and A. Lewis, “Grey wolf optimizer,”
Advances in Engineering Software, vol. 69, pp. 46–61, 2014.
[13] B. Mahdad and K. Srairi, “Blackout risk prevention in a smart
grid based flexible optimal strategy using Grey Wolf-pattern
search algorithms,” Energy Conversion and Management, vol.
98, pp. 411–429, 2015.
[14] X. Song, L. Tang, S. Zhao et al., “Grey Wolf Optimizer for
parameter estimation in surface waves,” Soil Dynamics and
Earthquake Engineering, vol. 75, pp. 147–157, 2015.
[15] M. H. Sulaiman, Z. Mustaffa, M. R. Mohamed, and O. Aliman,
“Using the gray wolf optimizer for solving optimal reactive
power dispatch problem,” Applied Soft Computing, vol. 32, pp.
286–292, 2015.
[16] G. M. Komaki and V. Kayvanfar, “Grey Wolf Optimizer algo-
rithm for the two-stage assembly flow shop scheduling problem
with release time,” Journal of Computational Science, vol. 8, pp.
109–120, 2015.
[17] N. Jayakumar, S. Subramanian, S. Ganesan, and E. B.
Elanchezhian, “Grey wolf optimization for combined heat
and power dispatch with cogeneration systems,” International
Journal of Electrical Power & Energy Systems, vol. 74, pp.
252–264, 2016.
[18] Y. Sharma and L. C. Saikia, “Automatic generation control
of a multi-area ST—thermal power system using Grey Wolf
optimizer algorithm based classical controllers,” International
Journal of Electrical Power & Energy Systems, vol. 73, pp. 853–
862, 2015.