(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 13, No. 1, 2022

Design and Performance Analysis of Anti-Surge

Control Mechanism for Compressor System using
Neural Networks
Divya M.N1 S L Gangadhariah3
Research Scholar Dept. of Electronics and Communication
VTU Research Centre, MSRIT M.S. Ramaiah Institute of Technology
VTU, Belagavi, India Bangalore, India

Narayanappa C.K2 V Nuthan Prasad4

Dept. of Medical Electronics Dept. of Electronics and Communication Organization
M.S. Ramaiah Institute of Technology, Bangalore, India M.S. Ramaiah Institute of Technology, Bangalore, India

Abstract—The compressor system is caused by the surge, Similarly, a surge will occur with the mass flow and
which is an instability occurrence in most gas-process and oil pressure oscillations if the' B' factor is higher. The spool
industries. These issues are solved by using a recycle valve that dynamics are introduced in the 'B' parameter to improve the
avoids the surge and provides higher mass flow in the speed control [4]. The Full-range Surge and rotating-stall
compressor system. An advanced controller-based anti-surge avoidance by Badmus et al. are presented in the compressor
control mechanism is a need in the compressor system to improve system. The model uses open-loop feed-forward and closed-
the stability and surge issues. In this manuscript, an efficient, loop feedback control mechanisms in the throttle valve area to
Neural-network predictive controller (NNPC) based variable control the Pressure and mass flow. These control mechanisms
speed compressor recycle system is modeled with an anti-surge
are scheduled and use in rotating stall and surge avoidance [5].
control mechanism. When the mass flow is deficient, the recycle
system is introduced, acts as a safety system, and feeds the
The calculation of operating location between surge and the
compressed gas back to the upstream system. The different stable condition is necessary to avoid the surge in Compressor.
controllers like Proportional Integral Derivative (PID) controller, To counteract the surge, the invariant coordinate system is
Fuzzy logic controller (FLC), and Neuro-fuzzy controller (NFC) introduced by changing molecular weight [6]. Gravdahl et al.
based anti-surge control mechanism are also used in Compressor present the drive-torque actuation system to regulate the active
recycle system to compare the stability and performance metrics surge in the centrifugal Compressor. The active surge
with NNPC. The NNPC based compressor system provides a stabilization utilizing control law [7] causes the throttle line to
better operating position and dynamic response with less error appear left to the surge line.
than other controllers-based compressor systems.
In contrast, the compressor system for active surge control
Keywords—Anti-surge; fuzzy logic controller (FLC); neuro- is presented by bohagen et al. using drive torque as a control
fuzzy controller (NFC); compressors; neural-network input. The desired operating point stabilization is achieved with
static and dynamic feedback mechanisms [8]. The Fuzzy logic-
I. INTRODUCTION based surge control mechanism [9] is introduced for the
The centrifugal Compressor is used in most industries like Moore-Greitzer model with constant speed in the compressor
oil, aero-space, and gas-plant to increase pressure and oil system, stabilizing the different operating conditions and
production. The gas compressors are generally divided into extending the stable line next to the surge line. The active surge
four types: Axial-flow, centrifugal, rotary, and Reciprocating- control with variable speed mechanism [10] in centrifugal
type. The rotary and Reciprocating-type compressors reduce Compressor is designed to analyze the different performance
the gas volume's occupancy and later apply the higher Pressure metrics with better improvement than the Gravdahl et al. [1]
to discharge the gas. The axial and centrifugal compressors act work. The Contribution of the research work is organized as
as turbo-compressor with continuous flow [1-2]. The Axial follows:
compressors with rotating stall and surge are presented with • The complete variable speed-based Compressor system
theoretical and mathematical concepts by Greitzer et al. [3]. with a recycling loop is modeled to ensure higher mass
The non-dimensional parameter (B) factors define whether the flow and avoid the surge.
Compressor is in surge or rotating stall conditions. If the 'B'
factor is lower, then a rotating stall will occur, which causes • The Neural-network predictive controller (NNPC)
the instabilities by reducing the mass flow by rising the based anti-surge control mechanism is modeled for the
Pressure. Compressor recycle system to improve stability and
prevent active surges.

710 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

• The different controllers like PID controller, Fuzzy an anti-surge controller to avoid the surge and improve the
logic controller (FLC), and Neuro-fuzzy controller robustness against the mass flow and pressure-based
(NFC) are also modeled in anti-surge control disturbances. Khsheem et al. [22] present the Surge compressor
mechanism to analyze the stability and performance system with an active control mechanism. The PID controller
comparison. controls the surge line and provides the dynamic response to
use in the control valve. Guan et al. [23] present the centrifugal
In this manuscript, an efficient Compressor recycle system compressor surge control mechanism. The work analyzes the
with an anti-surge control mechanism using NNPC is modeled variable –tip clearance mechanism is introduced to control the
and compared with other controllers to evaluate performance surge and analyzes the compressor performance. The Surge
metrics. The manuscript is organized as follows: Section II control with different speeds and also with different throttle
explains the Compressor-based system's current works for opening sizes are analyzed with simulation results. Aribi et al.
different applications. The simple compressor system with [24] present the Recycle compressor system for active surge
mathematical expressions is described in Section III. The control using a hybrid adaptive controller. The PID with auto
Compressor recycle system with an anti-surge control tuning Fuzzy interface system (FIS) controls the operating
mechanism using other controllers is explained in Section IV. points under the Surge line at different speeds. The artificial
The results and discussion is carried out in Section V. neural network (ANN) based Centrifugal compressor is
Section VI concludes the overall work with performance modeled by Ebrahimi et al. [25] to predict operational
improvement and also suggest the future scope. parameters.
II. RELATED WORK III. COMPRESSOR MODEL
This section analyzes the recent works on Compressor The surge control and instabilities are the main issues in
based systems for different application requirements. Budinis any of the axial and centrifugal compressors. In this work,
et al. [11-12] describe the Centrifugal compressor's control variable-speed Centrifugal-based Compressor is discussed. The
mechanism using a Model predictive controller (MPC), which overview of the Compressor system is represented in Fig. 1. It
can control the Compressor's Pressure during surge operation. mainly contains a Torque drive, Compressor with duct,
The hot and cold recycle line is configured with an anti-surge Plenum, and Throttle valve. The Compressor model produces
control mechanism to analyze the performance metrics and the mass and Pressure using torque drive and plenum units.
power consumption. Cortinovis et al. [13-14] present the The Throttle valve delivers the mass throttle value and
Linearize MPC-based centrifugal gas compressor system with feedback to the Plenum unit and Compressor mass flow.
an anti-surge and process control mechanism. These
mechanisms are adopted in safe and electric-driven When the mass is applied to the Plenum unit, which
applications with improved surge 11% by maintaining distance produces the Pressure, it is represented in Equation (1). The
and 50% reduction in process control time (settling). Sheng et momentum applied on the Compressor with duct produces the
al. [15] describe the Throttle and closed-coupled Valve (CCV) mass flow is described in Equation (2). The torque drive
actuators based Compressor system for the instability control delivers the angular speed using the angular momentum feature
mechanism. The double actuators scheme is introduced to represented in Equation (3). The work was initially presented
improve the control performance in CCV by tuning the B- by Gravdahl et al. [1], and it is extended by Fink et al. [2]. The
parameter in the Throttle valve. Gritli et al. [16] present the Pressure, mass flow, and speed is represented in Equation (1-3)
Electronic throttle valve (ETV) optimization using a genetic as follows:
algorithm (GA) by tuning the PID-based Fuzzy factors. The
GA with PID-based FLC provides better disturbance rejection a 2p
P p = ( m − mt )
and trajectory tracking points than other conventional ETV Vp
(1)
approaches. Azeem et al. [17] present the ETV with FLC based
super-twisting-Sliding mode control (STSMC) scheme. The A
STSMC scheme improves the controller's dynamic  =
m (ψ c (m, ω ) P01 − Pp ) (2)
L
performance in ETV without compromising the tracking
accuracy and stability features. Saeed et al. [18] present the 1
Hybrid electric vehicles (HEVs) speed control mechanism ω = (τ d − τ c ) (3)
J
using different controllers. The state observer controller (OBC)
and linear quadratic regulator (LQR) based scheme provide Torque Drive
better dynamic control response than the conventional PID
Controller scheme in HEV. τd τc

The Neuro-Fuzzy controller (NFC) based Grid-connected ω

gas turbine is modeled by Iqbal et al. [19] to improve the Pp, Vp, Tp
Duct
transient response and maintain stable operations. Liying et al.
[20] present the non-linear compensation control mechanism P01 P02 A m mt
using a PID controller for ETV to analyze the response time L
and performance. Zaibari et al. [21] explain the constant-speed- Plenum Throttle
Compressor
based centrifugal Compressor using Tube-MPC to control the
surge instability. The Adaptive fuzzy logic with MPC acts as Fig. 1. An Overview of the Compressor System.

711 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

The above Equations (1-3) uses few of the notations as affects the operating point and speed features in the compressor
follows: where ṁ is mass flow, Ṗ p is Plenum Pressure, ω is system. Suppose the pressure P 1 value is too high in the
Torque speed of the impeller, V p is Plenum Volume, a p suction unit, which fails to operate in the compressor system.
Plenum's speed of the gas, m t is Throttle's mass flow, A is the The Suction unit with Pressure (P 1 ), and Volume1 (V 1 ), are
cross-sectional area of the duct, L is the length of the duct, P 01 part of the upstream piping operation. Similarly, The Plenum
is the Pressure (Ambient), ψ c is Compressor Characteristic, J unit with Pressure (P 2 ), and Volume2 (V 2 ), are part of the
represents the moment of inertia of torque, τ d is Torque drive, downstream piping operation. The modeling of the Compressor
and τ c is impeller blades torque. recycle system is represented in Fig. 3.
The predicting surge is possible using Equations (1-3) and The amount of feed flow (m f ) with mass applied on the
is one-dimensional in centrifugal compressors. When Suction unit, which generates the Pressure (P 1 ) and is
Compressor is in the deep surge point, and the impeller blade's represented in Equation (7). The mass flow is used on the
torque is represented in Equation (4) as follows: Plenum unit, which produces the Pressure (P 2 ) and is
represented in Equation (8). The momentum applied to the
τ c = m r22 µω Compressor creates the mass flow (m), described in Equation
(4) (9). The throttle mass flow of the compressor recycle system is
The Compressor system is used to determine the pressure represented in Equation (10) as follows:
ratio using ψ c . The compressor characteristic is expressed in
a 2p
Equation (5) as follows [1]:
P1 = ( m f − mr − m )
k V1 (7)
 1  k −1
 µr22ω 2 − r12 (ω − αm) 2 − k f m 2 
ψ c (m, ω ) = 1 + 2  a 2p


a pT1 


P2 = (m − mt − mr ) (8)
  (5) V2
The throttle mass flow is represented in Equation (6) as
A
follows: m = (ψ c (m, ω ) P1 − P2 ) (9)
L
mt = tanh(ς ( Pp − P01 )) At ( Pp − P01 ) tanh(ς ( Pp − P01 ))
(6)
mt = tanh(ς ( P2 − P01 ))C t ( P2 − P01 ) tanh(ς ( P2 − P01 )) ς >> 0
The above Equations (4-6) uses a few of the notations as (10)
follows: r 1 is inducer radius, and r 2 is impeller exit radius, µ is
flow co-efficient, k is heat capacity ratio, T 1 is inlet The feed flow mechanism is dependent more on the type of
temperature, A t is Orifice Opening area, and ζ is zeta (>>1). Compressor used and is not constant while performing the
The Compressor characteristic provides the performance simulation process. The feed flow (m f ) operation is represented
metrics of the Compressor using mass flow and Pressure. in Equation (11), and recycle flow (m r ) operation is described
When the mass flow decreases, the Pressure will increase until in Equation (12) and are as follows:
its instability point, Then Surge or rotating stall problems arise
in the compressor system. m f = C f P01u − P1 (11)
IV. COMPRESSOR WITH ANTI-SURGE CONTROL USING
DIFFERENT CONTROLLERS mr = Cr P2 − P1 (12)
The oscillations, stalls/surge, and lower pressure rises will C f and C r are coefficients of Feed flow and recycle flow,
always affect the centrifugal Compressor's performance. The respectively; P 01u is upstream ambient Pressure.
operating point is always far away from the surge line (SL),
which is the best option at low pressure and high mass flow The Anti-Surge Control mechanism using PID Controller
conditions. But the efficiency of the Compressor always lacks with different Controllers like Fuzzy logic controller (FLC),
in this region. So for better efficiency, the Pressure will rise at Neuro-fuzzy Controller (NFC), and Neutral network Predictive
its highest, which is close to the SL. Surge avoidance is also Controller (NNPC) are represented in Fig. 4. In this work, the
known as an anti-surge control mechanism. Most commonly, PID controller is used as the industry standard and other
the recycle valves are used to avoid the surge in industry controllers to analyze the anti-surge control operations. The
applications. The Compressor Recycle system based Anti- Compressor system produces the mass flow (m) and
surge Control Mechanism using different Controllers is compressor characteristic (ψ c ) outputs and inputs the anti-surge
represented in Fig. 2. mechanism. The different controller's output (asc in ) is input the
recycle value unit. The Surge control line (SCL) is considered
Using an anti-surge control mechanism, the Compressor along with the surge line (SL), which is linear and horizontal to
recycle system contains a Suction unit, Compressor with duct, the surge margin. The compressor characteristic SCL with the
Torque drive unit, Plenum and recycle valve units. The feed mass flow is represented in Equation (13) as follows:
flow mechanism is essential to analyze the compressor system
performance if the feed flow value is high or too low, which ψ scl (m) = im + j (13)

712 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

Torque Drive
τd τc

ω
mf P1, V1, T1 Duct P2, V2, T2

P1 P2 A m
mt

Suction Compressor
Plenum
Controller
mr

Recycle Value

Fig. 2. Compressor Recycle System based Anti-surge Control Mechanism in using different Controllers.

Compressor Recycle System

mf
P1
P01u Feed Flow Recycle mr
Suction ascin
Valve Valve

wm P2

Torque
Compressor m Throttle
Drive Plenum P01a
Td with Duct Valve mt

ψc

Fig. 3. Modeling of the Compressor Recycle System.

Anti-Surge Control Mechanism

m Opid
Surge Control Er PID
FLC/ NFC/ NNPC
ψc Mecahnsim Controller
ascin

Fig. 4. Anti-Surge Control Mechanism using different Controllers.

Where i and j are the coefficients for SCL, when the di = m − mscl (ψ c ) (14)
operating point (OP) is right to the SCL, the control
mechanism is not activated. The OP is calculated using mass (m) and ψ c . The Ψ scl (m)
Only when the operating point (OP) is left to the SCL, the is same as ψ c , when Pressure rises and consider the inverse of
control mechanism is activated. The PID controller and other Equation (13) to calculate the mscl (ψ c ), which is represented
controllers (FLC/ NFC/NNPC) are used to getting the OP back in Equation (15) as follows:
to the right, with its controlling mechanism. To analyze the
ψc − j
controlling mechanism, first, find the distance (d i ), as the m scl (ψ c ) =
difference (horizontal) between the SCL and OP, which is i (15)
represented in Equation (14) as follows:

713 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

Use Equation (15) to calculate the distance value in The two Fuzzy inputs (O pid and CO pid ) and FLC output (Oflc)
Equation (14). The error value is calculated based on the are normalized to the display range [0 to 0.6] using fuzzy set
horizontal distance using OP and SCL. When the distance membership functions for the given universe of discourse. The
value is +ve, the error value (E r ) is set to zero, and no need to Fuzzy Input and output use the triangular-shaped membership
perform any controlling operation, and the OP is located to the functions. The Three fuzzy variables are used as Opid and
right position of SCL. When the distance value is -ve, its +ve COpid, which includes Negative Big (NB), Zero (ZO), and
value is used to calculate the error value (E r ) using controlling Positive Big (PB). Human knowledge is used to create the
operation. The error value (E r ) calculation is represented in decision logic for fuzzy rule sets. The Fuzzy rule set for
Equation (16) as follows: Compressor recycle system is tabulated in Table I. The fuzzy
variables like NB are in the range of [-0.3 to 0.3], ZO is in the
 0 di > 0 field of [0 to 0.6], and PB is in the range of [0.3 to 0.6]. The
Er =  (16) nine- rules are used in the fuzzy inputs (O pid and CO pid ) with
− d i else three fuzzy variables. The rule set converts these two fuzzy
inputs into single FLC output using the centroid technique in
This error value (E r ) is input to the PID controller and it is
the de-fuzzification process.
represented in Equation (17) as follows:
t
B. Neuro-Fuzzy Controller (NFC)
O pid = K p Er + K i ∫ Er dt The FLC is extended with NFC using an Artificial Neuro-
0 (17) fuzzy interface system (ANFIS). The NFC automatically
Where K p and K i are the proportional gain and integral realizes the FIS using Neural networks (NNs). The NFC
gain of the PID controller, the PID Controller output (O pid ) is provides a conversion mechanism (Crisp to fuzzy and vice-
used directly to control the flow percentage in recycling valve versa) efficiently. The NFC uses the NN to optimize the FIS
for the anti-surge control mechanism. To improve the anti- using the ANFIS system. The operation of the NFC is
surge control with OP, a different controlling mechanism is represented in Fig. 6. The NN has a learning skill set and is
introduced. capable of training and testing the FIS in NFC. The NFC uses
two inputs like PID controller output (O pid ) and change in PID
A. Fuzzy Logic Controller (FLC) controller output (CO pid ) for training. Once the training data is
The Fuzzy logic Controller (FLC) provides a better loaded in NFC, generate the FIS using grid partition using the
dynamic response than the conventional PID controllers and is Gaussian membership function. The seven MF's are considered
used in most industrial control applications. The FLC mainly with constant output in NFC.
contains five blocks: the Compressor recycle system: Next, Train the FIS using the hybrid optimization method.
Fuzzification, De-fuzzification, database (Knowledge base), The error tolerance is set to 3 epochs to complete the ANFIS
Ruleset, and evaluation process. The Fuzzification converts training. Lastly, test the training data to analyze the NFC
absolute (crisp) data into Fuzzy (linguistic) data. The PID output.
controller output is input to the fuzzification process as crisp
data and produces the fuzzy data. These fuzzy data values are Knowledge
considered as Fuzzy-sets in FLC. The membership functions Base
analyze these fuzzy sets and estimate whether they are low,
high, or oversized. The evaluation process provides the
decision to control and study the fuzzy rules stored in the rule Fuzzification Evaluation Defuzzification
Fuzzy Fuzzy
base. The FLC process performance is analyzed by rule base.
The de-fuzzification process converts back the Fuzzy data into Rule Base
crisp data. The Knowledge-based is used to store the
membership functions of both the Fuzzification and de- Compressor Non-Fuzzy
fuzzification processes. The evaluation of fuzzy rules is done Processed (Recycle
Actual control
by basic fuzzy sets with AND, OR, and NOT operations. The Output Valve)
de-fuzzification output is the final FLC output and is processed Fig. 5. Fuzzy Logic Process with Compressor Recycle System.
in Compressor recycle system. The function of the Fuzzy logic
process with the Compressor recycle system is represented in TABLE I. THE FUZZY RULE SET FOR COMPRESSOR RECYCLE SYSTEM
Fig. 5.
E/CE NB ZO PB
The Mamdani Fuzzy interface system (FIS) is used to NB NB NB ZO
create the FLC. The two fuzzy variables are used in the FLC:
ZO NB ZO PB
one is PID output (O pid ), and another is Change in PID output
(CO pid ) = O pid (i) – O pid (i-1). Where i value is set to 1 to 3. PB ZO PB PB

Input Output
Training Data FIS Generation Train FIS Test FIS
data data

Fig. 6. Operation of NFC.

714 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

V. RESULTS AND DISCUSSION

C. Neutral Network Predictive Controller (NNPC)
The NNPC mainly contains the NN model, Optimization The Compressor recycle system with an anti-surge control
unit, and Compressor recycle system. The NNPC predicts the mechanism using a PID controller with different controllers is
future performance of the Compressor recycle system using the presented in this section. The Simulation results of NNPC
NN model. The NNPC Based Compressor recycle system for based compressor recycle system are introduced and compared
the anti-surge controlling mechanism is represented in Fig. 7. with other controllers like PID controller, FLC, and NFC to
The NNPC is used to calculate the NN model input (m) and realize the performance metrics. The NNPC based Compressor
recycle system is simulated with the following assumption. The
further optimize the Compressor recycle system's performance
mass flow (m) and compressor speed (ω) is set to zero initially;
over a defined time horizon.
two volumes of Pressure (V 1 and V 2 ) are defined as ambient.
PID Opid The higher mass flow will be obtained by opening the recycle
Er
Controller valve. When more mass flow occurs, the operating point (OP)
Optimization
positions are shifted to the right position in ψc. The simulation
ascin
results of the Compressor recycle system are represented in
mm Fig. 8. The compressor output mass flow (m), throttle valve
asc`in Compressor
Recylce system output (m t ), recycles valve output (m r ), and feed flow valve
output (m f ) is represented on the left-hand side. The torque
Neural Network(NN) drive's speed (ω), suction pressure (P 1 ), and Plenum pressure
m Model (P 2 ) are presented on the right side of Fig. 8.

Fig. 7. NNPC based Compressor Recycle System for Anti-surge Control.

The NNPC first identifies the Compressor recycle system

(plant) and then uses this model to predict the future
performance using a controlling mechanism. In the first stage,
the NN model is trained by a predictive control mechanism to
analyze the dynamics of the Compressor recycle system. The
NN training signal is generated by calculating the predictive
error between the NN mode output and Compressor recycle
system output. The NN model [26] uses the previous inputs
(Optimization) and the previous compressor system's output to
analyze the Compressor recycle system's output values. In the
second stage, the optimization algorithm [27-28] is used for
precision and to calculate the control signal. This control signal
minimizes the below Equation's (18) performance over a
defined time horizon.
N2 Nu
J= ∑ (O
j = N1
pid (t + j ) − mm (t + j )) 2 + ρ ∑ (asc`in (t + j − 1) − asc`in (t + j − 2)) 2
j =1 (18)
Where asc' in is a temporary control signal, O pid is PID
controller's output acts as a desired response, m m is the
response of NN model, ρ is used to calculate the performance
index's sum with squares of control increments. N 1 , N 2 , and N u
values define cost and control horizon.
The J value is minimized using the optimization unit by
calculating the asc' in and inputting the ascin to the Compressor
recycle system. The Recycle valve output is represented after
the anti-surge controlling mechanism using NNPC controller is
described in Equation (19) as follows:

mr = O pid .ascin . P2 − P1
(19)
The NNPC uses four input layers, five hidden layers, and 1-
output layer for training the data. The 100 training samples
with a sample interval of one second are selected for training
the data in the NN training tool. The NNPC obtains the six
iterations (epoch) with 8.44e-6 performance and six validation
checks while training the data. Fig. 8. Simulation Results of Compressor Recycle System.

715 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

The mass flow is set to zero initially and varied based on compared with SL and SCL. The SL and SCL are defined
speed, Throttle, and feed flow response. The throttle output initially and perform the simulation of the Compressor recycle
(mt) is changed based on plenum pressure (P 2 ) and ambient system using different controllers to obtain the OP's position.
condition (P 01a ). The recycle output is not activated until the
controller responds and varies based on NNPC controlling The SL and SCL are linear and defined with i and j values
mechanism. The feed flow (m f ) is zero initially and increases using Equation (15). The OP is in the left position during the
to some extent and gradually decreases, which indicates the initialization of the Compressor recycle system. Later shifted to
system enters into surge line (SL). When the feed flow (m f ) the right next to the SCL and moving around it. The OP is very
output is steady, the OP's are shifted from SL to SCL. The near to SL and turns back to SCL using PID Controller,
torque drive provides 3000 rad/s—speed output (ω) after the whereas, in FLC and NFC, The OP is flat and back to SCL. In
controller response. The two pressures (P 1 and P 2 ) outputs are contrast, The NNPC provides a better OP position, and OP is
decreased eventually and reach steady-state after NNPC located next to the SCL. So, the NNPC provides a better OP
response. position than PID, FLC, and NFC in Compressor recycle
system.
The operating point (OP) position and SL and SCL using
different controllers in the Compressor recycle system is The different controller's output response for Compressor
represented in Fig. 9. The current OP position is identified recycle system is represented in Fig. 10. The output response
using other controllers (PID, FLC, NFC, and NNPC) and of all the controllers is initially zero and later varied based on
the controlling mechanism in Compressor recycle system.

Fig. 9. Different Controller's Operating Points Position in Compressor Recycle System.

716 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

NNPC ITAE IAE ISE

NFC

Controllers
FLC

PID

0 0.05 0.1 0.15

Error Values

Fig. 10. Dynamic Response of different Controllers for Compressor Recycle Fig. 12. Error Calculation of different Controllers for Compressors Recycle
System. System.

The FLC takes more time to provide the output response, VI. CONCLUSION AND FUTURE WORK
whereas NNPC takes less time to provide output response than In this manuscript, an efficient NNPC based anti-surge
PID and NFC. The dynamic output response values of different control mechanism is modeled for variable–speed centrifugal
controllers concerning time for the Compressor recycle system Compressor recycle system. The NNPC provides higher mass
are represented in Fig. 11. The output response of the PID flow in the recycle valve to the compressor system and avoids
controller takes a rise time of 0.976 sec, settled at 4.27 sec with the surge with better stabilization. The Compressor is working
an overshoot (%) of 0.263 sec. Similarly, the FLC controller at a more increased mass flow with low Pressure to avoid the
takes a rise time of 1.905 sec, settled at 3.87 sec with an surge. The simulation results of NNPC based Compressor
overshoot (%) time of 0.077 sec. The NFC takes 0.1112 for recycle systems like mass flow, speed, and Pressure waveforms
rising time, settled at 0.236 sec with overshoot (%) time of are discussed. The operating point's position with the Surge
zero. In contrast, the output response of the NNPC takes a rise line and surge control line for different controllers is
time of 0.013 sec, settled at 0.2369 sec with an overshoot (%) represented. The NNPC based mechanism provides dynamic
of zero. Overall, the NNPC provides a better dynamic output response than PID, FLC, and NFC controller-based
response with less rise time, settled early, and zero overshoot mechanisms. The dynamic responses (Rise time, settling time,
time than PID, FLC, and NFC. and % overshoot time) of the NNPC based Compressor recycle
The error analysis of different controllers for the system reduces the ISE of 14.49%, 4.83%, and 11.86% in PID,
Compressor recycle system is represented in Fig. 12. The error FLC, and NFC based Compressor recycle system. Similarly,
analysis includes Integral Square Error (ISE), Integral Absolute NNPC also reduces the IAE and ITAE values than PID, FLC,
Error (IAE), and integral time absolute error (ITAE) and NFC-based controller mechanisms. In future, incorporate
parameters. The error response of the NNPC based Compressor the proposed work with real-time gas plants to realize the
recycle system improves 14.49 % in ISE, 51.9 % in IAE, and performance metrics. Also extend the recycle system with new
95.6 % in ITAE than the PID-based Compressor recycle approach to analyze the stability of the compressor system.
system. Similarly, the 4.83 % in ISE, 34.8 % in IAE, and 91.04 REFERENCES
% in ITAE are improved compared to FLC. Compared to NFC, [1] Gravdahl, Jan Tommy, and Olav Egeland. Compressor surge and
the 11.86 % in ISE, 34.92 % in IAE and 95.53 % in ITAE are rotating stall: Modeling and control. Springer Science & Business
improved in NNPC based compressor recycle system. Media, 1999.
[2] Fink, David Alan, Nicholas A. Cumpsty, and Edward M. Greitzer.
4.5 "Surge dynamics in a free-spool centrifugal compressor system." (1992):
4 Rise Time 321-332.
Settling Time [3] Greitzer, Edward M. "Surge and rotating stall in axial flow
3.5
% Overshoot compressors—Part II: experimental results and comparison with
Response Time

3 theory." (1976): 199-211.

2.5 [4] Gravdahl, Jan Tommy, and Olav Egeland. "A Moore-Greitzer axial
compressor model with spool dynamics." In Proceedings of the 36th
2
IEEE Conference on Decision and Control, vol. 5, pp. 4714-4719. IEEE,
1.5 1997.
1 [5] Badmus, O. O., C. N. Nett, and F. J. Schork. "An integrated, full-range
surge control/rotating stall avoidance compressor control system." In
0.5
1991 American Control Conference, pp. 3173-3180. IEEE, 1991.
0 [6] Batson, Brett W. "Invariant coordinate systems for compressor control."
PID FLC NFC NNPC In Turbo Expo: Power for Land, Sea, and Air, vol. 78729, p.
Controllers V001T01A070. American Society of Mechanical Engineers, 1996.
[7] Gravdahl, Jan Tommy, Olav Egeland, and Svein Ove Vatland. "Drive
Fig. 11. Output Response Values of different Controllers for Compressor torque actuation in active surge control of centrifugal compressors."
Recycle System. Automatica 38, no. 11 (2002): 1881-1893.

717 | P a g e
www.ijacsa.thesai.org
 (IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 13, No. 1, 2022

[8] Bøhagen, Bjørnar, and Jan Tommy Gravdahl. "Active surge control of [19] Mustafa, Mohamed Iqbal Mohamed, Joseph Xavier Rayappan, and
compression system using drive torque." Automatica 44, no. 4 (2008): Kanakaraj Jagannathan. "A neuro-fuzzy controller for grid-connected
1135-1140. heavy-duty gas turbine power plants." Turkish Journal of Electrical
[9] Shehata, Raef S., Hussein A. Abdullah, and Fayez FG Areed. "Fuzzy Engineering & Computer Sciences 25, no. 3 (2017): 2375-2387.
logic surge control in constant speed centrifugal compressors." In 2008 [20] Liying, Zhang, Xu Ting, Li Xuejun, Li Liyan, Guo Pan, and Zhu Yubao.
Canadian Conference on Electrical and Computer Engineering, pp. "Study on Nonlinear Compensation Control Method for Electronic
000653-000658. IEEE, 2008. Throttle Valve." In 2018 Chinese Automation Congress (CAC), pp.
[10] Albehigi, Abdulkareem A. Wahab, and Rasha Hyder Hashim. 1085-1089. IEEE, 2018.
"Experimental and Theoretical Analysis of the Surge in a Centrifugal [21] Taleb Ziabari, Masoud, Mohammad Reza Jahed-Motlagh, Karim
Compressor." In Modern Methods of Construction Design, pp. 317-328. Salahshoor, Amin Ramezani, and Ali Moarefianpur. "Robust adaptive
Springer, Cham, 2014. control of surge instability in constant speed centrifugal compressors
[11] Budinis, S., and N. F. Thornhill. "Control of centrifugal compressors via using tube-MPC." Cogent Engineering 4, no. 1 (2017): 1339335.
model predictive control for enhanced oil recovery applications." IFAC- [22] Abo-khsheem, K. A. "Active Control of Surge Compressor System."
papers online 48, no. 6 (2015): 9-14. Journal of Electrical & Electronic Systems 7, no. 3 (2018): 1000267.
[12] Budinis, Sara, and Nina F. Thornhill. "Supercritical fluid recycle for [23] Guan, Xudong, Jin Zhou, Chaowu Jin, Yuanping Xu, and Hengbin Cui.
surge control of CO2 centrifugal compressors." Computers & Chemical "Influence of different operating conditions on centrifugal compressor
Engineering 91 (2016): 329-342. surge control with active magnetic bearings." Engineering Applications
[13] Cortinovis, Andrea, Hans Joachim Ferreau, Daniel Lewandowski, and of Computational Fluid Mechanics 13, no. 1 (2019): 824-832.
Mehmet Mercangöz. "Safe and efficient operation of centrifugal [24] ARIBI, Yacine, Razika Zammoum Boushaki, and Hocine Loubar.
compressors using linearized MPC." In 53rd IEEE Conference on "Active Surge Control of the Recycle Compression System By Hybrid
Decision and Control, pp. 3982-3987. IEEE, 2014. Adaptive Controller." In 2019 4th International Conference on Power
[14] Cortinovis, A., H. J. Ferreau, D. Lewandowski, and M. Mercangöz. Electronics and their Applications (ICPEA), pp. 1-9. IEEE, 2019.
"Experimental evaluation of MPC-based anti-surge and process control [25] Ebrahimi, Seyed Hossain, and Ahmad Afshari. "An Artificial Neural
for electric driven centrifugal gas compressors." Journal of process Network Model for Prediction of the Operational Parameters of
control 34 (2015): 13-25. Centrifugal Compressors: An Alternative Comparison Method for
[15] Sheng, Hanlin, Wei Huang, Tianhong Zhang, and Xianghua Huang. Regression." Journal of Sciences, Islamic Republic of Iran 31, no. 3
"Compressor instability active control via a closed-coupled valve and (2020): 259-275.
throttle actuators." International Journal of Turbo & Jet-Engines 32, no. [26] Asgari, Hamid, XiaoQi Chen, Mohammad B. Menhaj, and Raazesh
3 (2015): 257-264. Sainudiin. "Artificial neural network-based system identification for a
[16] Gritli, Wafa, Hajer Gharsallaoui, and Mohamed Benrejeb. "PID-type single-shaft gas turbine." Journal of Engineering for Gas Turbines and
fuzzy scaling factors tuning using genetic algorithm and Simulink Power 135, no. 9 (2013).
design optimization for an electronic throttle valve." In 2016 [27] Zhong, Lulu, Yang Liu, Jun Zhao, and Wei Wang. "Deep predictive
International Conference on Control, Decision and Information controller designed for centrifugal compressor system anti-surge." In
Technologies (CoDIT), pp. 216-221. IEEE, 2016. 2020 Chinese Automation Congress (CAC), pp. 6554-6559. IEEE, 2020.
[17] Azeem, Mohammad Fazle, and Abdul Kareem. "A fuzzy logic-based [28] Shestopalov, Mikhail Yu, Ruslan I. Smirnov, and Damir H. Imaev.
super-twisting sliding mode control scheme for electronic throttle "Approximation of the Natural Gas Pumping Compressor
control." In 2016 IEEE International Conference on Fuzzy Systems Characteristics using a Multi-layer Neural Network." In 2021 IEEE
(FUZZ-IEEE), pp. 485-492. IEEE, 2016. Conference of Russian Young Researchers in Electrical and Electronic
[18] Saeed, Muhammad Ahsan, Nisar Ahmed, Mujahid Hussain, and Adnan Engineering (ElConRus), pp. 1088-1091. IEEE, 2021.
Jafar. "A comparative study of controllers for optimal speed control of
the hybrid electric vehicle." In 2016 International Conference on
Intelligent Systems Engineering (ICISE), pp. 1-4. IEEE, 2016.

718 | P a g e
www.ijacsa.thesai.org
 © 2022. This work is licensed under
https://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding
the ProQuest Terms and Conditions, you may use this content in accordance
with the terms of the License.