537

1 INTRODUCTION

On the sea fleet ships, the use of alternating current in

the electric power plant is dominant. The energy

efficiency of this power plants depends and provides

by using the rational methods of electric energy

control. The process of transmitting and converting

electric energy into other types of energy is

accompanied by periodic exchange and the transition

of reactive power between its inductive and capacitive

elements. With such energy metabolic processes,

energy losses are inevitable [1, 2].

Active-inductive consumers of alternating current

require generation of additional reactive power for

their operation, the sources of which on ships are

usually synchronous generators. However, alternating

current capacitors can also be sources of reactive

power. Cosine polypropylene capacitors with

aluminium-coated plates are an order of magnitude

lighter, smaller and cheaper than electric machines.

Therefore, their wider use as reactive power

compensators will increase the energy efficiency of the

ship's power plant. [3-5].

Until now, the industrial reactive power

compensation units (RPCU) are operating at a steady-

state load power factor, i.e. they are not dynamic. This

explains their insufficiently widespread use on ships,

where short-term and transient operating modes of

mechanisms predominate. The use of capacitor

reactive power sources on ships will be justified with

the successful scientific and technical development of

high-speed (dynamic) systems for measuring the

reactive power of consumers and controlling the

necessary compensating capacity RPCU [5, 6-8].

The operation modes of ship multi-generator

electrical power installations are characterized by

sharply variable loads with frequent start-ups of

electric drives with the commensurate power of an

induction electric motors and generators. As is known,

in the starting mode the motor consumes a large

reactive current, significantly reducing the overall

power factor of the generating network.

Dynamic Compensation of Reactive Power in Ship

Power Plants

L. Vyshnevskyi, M. Mukha, O. Veretennik, A. Drankova, I. Kozyryev & O. Vyshnevskyi

National University Odessa Maritime Academy, Odessa, Ukraine

ABSTRACT: In this article, the authors analysed the starting modes of the powerful marine electric drives with

asynchronous electric motors and capacitor reactive power compensators. The article considers reactive power

control in the class of linear-pulse regulators associated with the unstressed switching of the capacitors in the

network. The analysis of reactive power parameters sensors which are used in the discrete-pulse control system

showed that the most effective sensor is the reactive conductivity sensor, the value of which is defined as the

average value during the switching period and the current sensor, where measurements occur in the moments

when voltage transition through zero. The technical implementation of the second sensor is quite simple, does

not require the execution of division and multiplication operations, i.e. use of controllers.

http://www.transnav.eu

the International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 19

Number 2

June 2025

DOI: 10.12716/1001.19.02.24

538

Approximately the same power of drive engines

and generators aggregates, different slopes of loading

characteristics of primary engines and the speed

performance of their regulators, the modes of

generators parallel operation with different type of

driven mechanisms affect a transitional energy

processes of a ship autonomous multi-generator power

plant. In addition, the commensurate power of

generators and load, the need for generators

autonomous operation with high-power electrical

drives lead to more severe transitional modes with the

danger of disconnecting the electric power plant [5,9].

The use of condenser installations of dynamic

compensation for reactive power in the ship's

autonomous power plants opens the following

possibilities: firstly, to solve the problem of reactive

power compensation during the starting of the

powerful induction motor from the synchronous

generator and reduce the total current consumed from

the electric grid, and secondly, to increase the speed of

the voltage control channel for the synchronous

generator by joint regulation of reactive current in its

stator circuit. The effectiveness of this increasing

energy efficiency method requires the use of a high-

speed system for measuring and controlling the

consumed reactive power [5,9,10].

2 ANALYSIS OF STARTING AN

ASYNCHRONOUS MOTOR FROM THE

SYNCHRONOUS DIESEL GENERATOR WITH A

REACTIVE POWER COMPENSATION DEVICE

To compensate for the consumed inductive current of

the electric motor, we will consider the use of the

capacitive current adjustable source of switched

capacitors connected in parallel to the asynchronous

motor stator windings, Fig. 1 [9].

Figure 1. Modelling scheme of starting an asynchronous

motor from the synchronous diesel generator with a reactive

power compensation device: SG - synchronous generator; D

- diesel engine; AM - asynchronous motor; DM - driven

mechanism; RPCU – reactive power-controlled compensator;

VR, RR, RPR – voltage regulator, rotation regulator of diesel,

reactive power regulator

The calculation of the starting processes will be

performed on the developed mathematical model of

the system. The simulated autonomous power plant

contains automatic voltage control systems for the

synchronous generator, the rotation frequency control

drive engine, the starting asynchronous motor, and a

device for reactive power compensating load.

Let us consider two processes: direct starting of an

electric motor without compensation (Fig. 2 - Fig. 4)

and with compensation of the motor reactive current

(Fig. 5 - Fig. 6). The starting electric motor power is 40%

of the diesel generator power. Parameters of the

simulated asynchronous motor are: R1=0.02; R2=0.018;

X1=0.11; X2=0.073; Xm=3.3

The generator load before asynchronous motor

starts was 10% at power factor: cos



=1. The nominal

values of the generator corresponding parameters, as

well as the nominal rotation frequency of the starting

electric motor, are taken as the basic values of the

electrical and electromagnetic quantities in the figures.

Figure 2. Electromagnetic starting torque Md of the electric

motor and the resistance torque Ms of the driven mechanism

Figure 3. Rotation speed of the starting electric motor



Figure 4. Starting current Id of an electric motor for direct

starting

The regulation of the consumed reactive power in

the model is carried out by a controlled reactive power

compensator (RPCU) by changing the capacitive

current of the connected capacitors, which ensures a

close-to-zero angle



and a practically unitary load

power factor cos



. The current Ig of the generator in the

asynchronous motor start mode with reactive power

compensation is shown in Fig. 5.

539

Figure 5. Generator current Ig when starting an electric motor

with reactive current compensation

The reactive current of the generating network total

load is measured during each current period. The

capacitance of the compensating capacitors also

changes during each period, its change is shown in Fig.

Figure 6. Compensating capacitance Ck changing when

starting an electric motor

Comparison of transient processes in Fig. 4 and Fig.

5 shows that with reactive power compensation the

maximum starting current of the generator decreases

approximately twofold. At the same time, the starting

torques, and acceleration time of the asynchronous

motor remain practically unchanged.

2.1 Reactive power control with the unstressed inclusion

of the capacitors in the network

This article discusses technical solutions for the third

problem: an analysis of several discrete control laws for

three-phase sections of AG excitation capacitors is

carried out. The authors of the article consider the

further development of the previously described

controller [8, 9], which implements the integral

discrete-pulse law of voltage stabilization of the AG

with capacitor excitation.

In this article, we consider the reactive power

control in the class of linear-pulse regulators [9], in

which the control is quantized by time, and the control

signal amplitude linearly depends on the input signal.

Discrete time control is associated with

synchronization of the unstressed inclusion moments

of capacitors in the network [8-13].

The switching processes in circuits with capacitors

contribute to the occurrence of recharge currents,

which can be unacceptable for the normal operation of

thyristor switches.

If the difference in the voltage of the network and

the switched capacitor is more than a few volts ∆U, and

the resistance of the open key Rk is not enough, then the

charging current IRC may exceed the permissible values

of the thyristor current Imax. At the beginning of the

recharge, the current is equal, IRC=∆U/Rk, i.e. IRC>Imax. In

RC–circuits, the accumulation rate of the capacitor

charge during switching dIRC/dt is determined by the

maximum of opening speed the pn-transition of the

thyristor switch. The rate of voltage increases in the

RC-circuit is determined by constant time

T=RC:dURC/dt=∆U/RC. The maximum values of the

pulse current Imax, current dI/dt growth rate and the

voltage dU/dt growth rate of the thyristors is limited,

excesses of which leads to their destruction.

Operational restrictions on the parameters of switches

elements can be solved by turning on the limiting

throttle or synchronizing the moments of inclusion of

capacitors in the network, when the voltage on the

condenser and the network are equal, ∆U=0. In this

case, the charging current of the capacitor is zero.

An example of the synchronous inclusion of the

capacitor in the system for regulating the reactive

power of the synchronous generator's load is shown in

Fig. 7.

Figure 7. Transition processes during synchronization of the

inclusion of the capacitor to the network with equality

∆U=Ua-Ub=0

The opening moments of the switches in Fig.7

coincide with the moments when the difference

between the voltages of the network and the capacitor

is equal to zero ∆U=Ua-Ub=0. In the open state of the

thyristor switch, the voltage on the capacitor Uc

coincides with the network voltage Ua, and when

turned off, the capacitor is discharged according to an

exponential law through a discharge resistor. There are

no charging currents, and the value of the capacitor

current after opening the switch is determined by the

derivative of the network current: Ic=C



dUa/dt.

Figure 8. Thyristor switch for shockless capacitor switching

from EPCOS [3]

540

When a powerful load of the ship's network is

turned on during the transient process, the voltage

decreases significantly. The voltage on the capacitor

may not have time to drop to the mains voltage. If at

this moment the control circuit should turn on the

capacitor, then the synchronization circuit will not

allow this to happen and a delay in control will occur.

To discharge capacitors during the off state, special

discharge resistors are used, which are installed in

parallel with the capacitors, Fig. 8 and Fig. 9. These

schemes show examples of unstressed synchronization

of the inclusion of capacitors to the AC network [8,14].

Figure 9. Simistor switch for shockless capacitor switching

using a MOC 3083 type simistor with zero crossing circuit

The use of thyristor (simistor) circuit of unstressed

inclusion leads to a delay in switching of three phases

capacitors by at least 120 electric degrees. When using

synchronization of switching with the network, the

minimum period of discretion of control in time will be

180 electric degrees or a half-period of the network.

2.2 Discrete sensors of reactive power parameters

In this work, we limit ourselves to the consideration of

discrete control systems with a period of switching T0,

a multiple of the network, which for a network of 50

Hz is 20 ms. During this time, switching processes end

in the scattering chains of the motor load of the

shipboard network.

For discrete-pulse reactive power control, there are

several options for using information about the

adjustable parameter in synchronized moments of

time. We can use instant (1), filtered (2) or averaged for

the switching period (3) the value of the adjustable

parameter. As previously considered [5, 9], such

parameters can be reactive conductivity Y, reactive

current Ir or reactive power of the load Q.

The simplest implementation is the use of the

instantaneous value of the adjustable parameter - the

reactive current of the load Ir, because at the time of the

transition of phase voltage ug through zero, the instant

value of the phase current is equal to the value of the

reactive current, Ir=il0, see Fig. 10. It is enough to

measure this value, preserve and use for unstressed

control of capacitors at the time of the transition of their

currents through zero.

To use the instantaneous value of reactive

conductivity Y or reactive power Q for control, the

measured value of the reactive current Ir can be divided

and multiplied respectively by the voltage amplitude

Um:Y=Ir/Um, Q=Ir



Um. During the management period

T0, it is possible to perform additional operations of

filtering and averaging the control signals.

The computing process of multiplication or division

involves the use of more complex and expensive

controllers. However, the use of synchronized

instantaneous load current il0 values allow the use of

simple microcontrollers.

Figure 10. Voltage and load currents in the alternating

current circuit with a frequency of 50 Hz

The amplitude values of signals in discrete-pulse

systems change only in quantum moments of time. To

determine the current value of the reactive current

load, the measurement should be made at the moment

of the supply voltage transition through zero value,

Fig. 11.

Figure 11. Measurement of the current reactive current value

at the beginning of the direct start of the asynchronous

electric motor with a capacity of 36 % of the generator power

at the time of the phase voltage transition through zero value

If there are sources of nonlinear distortion in the

generator load, then the generator voltage and the load

current can be filtered with the same frequency

properties filters [10-17]. In this paper, we do not

consider such a load, so the role of the filters of

interference will be performed by the sensors of the

average for the period of the measured signal value,

Fig. 12:



dat n



dat rn



dat n



dat mn

where N is the number of measurements of the

adjustable parameter during the T0 period of the

discrete-pulse control system of reactive power.

541

Figure 12. Sensors signals of adjustable parameters - average

values for the switching period T0: a - reactive conductivity

sensor signal Ydat; b - reactive current signal Idat; c - voltage

sensor signal Udat

Figure 13. Comparison of reactive current sensors

3 CONCLUSIONS

1. The reactive power dynamic compensation by

connecting corresponding capacitors capacitance when

starting powerful electrical drives of the ship

mechanisms reduces the ship generator current of two

times.

2. This method of the reactive power compensation

allows to double of the ship power plant capabilities in

the start-up modes of powerful asynchronous motors,

which in turn will make it possible to increase the

power of asynchronous electric motors to 35-40 %

compared to the generators power.

3. Analysis of the measuring processes of reactive

power parameters in the discrete-pulse control system

shows that the most effective are the sensors of the

average value of reactive conduction and reactive

current.

4. The reactive current sensor can be implemented by

using a simple controller, since there is no need to

calculate nonlinear functions.

REFERENCES

[1] Energy efficiency handbook. Toward zero emission ship

operations. Helsinki: ABB Marine & Ports,

9AKK107680A8255. (2020). https://new.abb.com/marine/

energy-efficiency-handbook

[2] Power Quality Solutions. Product Profile. Edition 03/2022

– TDK Electronics AG, - 128 p. www.tdk-

electronics.tdk.com

[3] Film Capacitors – Power Factor Correction: Switching

Devices – Thyristor Modules for Dynamic PFC, TSM-LC-

I. – TDK: EPCOS AG 2015, - 7 p.

[4] Kabalci, E. (2017). Reactive Power Compensation in AC

Power Systems. In: Mahdavi Tabatabaei, N., Jafari

Aghbolaghi, A., Bizon, N., Blaabjerg, F. (eds) Reactive

Power Control in AC Power Systems. Power Systems.

Springer, Cham. https://doi.org/10.1007/978-3-319-51118-

4_8

[5] Management of reactive power of generator installations.

Monograph /Mykola Mukha, Leonid Vyshnevskyi,

Sergiy Dudko. – LAP LAMBERT Academic Publishing:

Saarbrücken, Germany, 2014. – 145 p. ISBN: 978-3-659-

53978-7 (Rus)

[6] Naser Mahdavi Tabatabaei, Ali Jafari Aghbolaghi, Nicu

Bizon, Frede Blaabjerg. Reactive Power Control in AC

Power Systems. Fundamentals and Current Issues.

Springer Cham, Boston, MA - 2017. 634 p.

https://doi.org/10.1007/978-3-319-51118-4 ISBN: 978-3-

319-51117-7

[7] Hong Chen, "Practices of reactive power management

and compensation," 2008 IEEE Power and Energy Society

General Meeting - Conversion and Delivery of Electrical

Energy in the 21st Century, Pittsburgh, PA, 2008, pp. 1-4,

https://doi:10.1109/PES.2008.4596588

[8] Montoya, O.; Chamorro, H.; Alvarado-Barrios, L.; Gil-

González, W.; Orozco-Henao, C. Genetic-Convex Model

for Dynamic Reactive Power Compensation in

Distribution Networks Using D-STATCOMs. Appl. Sci.

2021, 11, 3353.

[9] Mukha M. Dynamic compensation of reactive power in

shipbased autonomous electric power systems: Dr. tech.

sci. diss: 05.09.03/ Lviv Polytechnic National University.

Lviv. 2018. 404 p. (Ukr)

[10] (2006). Continuation Power Flow. In: Ajjarapu, V. (eds)

Computational Techniques for Voltage Stability

Assessment and Control. Power Electronics and Power

Systems. Springer, Boston, MA.

https://doi.org/10.1007/978-0-387-32935-2_3

[11] Iorgulescu, M., Ursu, D. (2017). Reactive Power Control

and Voltage Stability in Power Systems. In: Mahdavi

Tabatabaei, N., Jafari Aghbolaghi, A., Bizon, N.,

Blaabjerg, F. (eds) Reactive Power Control in AC Power

Systems. Power Systems. Springer, Cham.

https://doi.org/10.1007/978-3-319-51118-4_6

[12] Hassan Shokouhandeh, Sohaib Latif, Sadaf Irshad,

Mehrdad Ahmadi Kamarposhti, Ilhami Colak and Kei

Eguchi. Optimal Management of Reactive Power

Considering Voltage and Location of Control Devices

Using Artificial Bee Algorithm. - Appl. Sci. 2022, 12, 27.

https://doi.org/10.3390/app12010027

[13] Vyshnevskyi L. V. Methods of implementing discrete

regulators on electronic elements / L. V. Vyshnevskyi, M.

Y. Mukha, A. M. Veretennik. // Shipping. - Odessa:

OGMA. – 2002. – №5. – P. 25-29 (Rus)

[14] Burbelo M.Y. Dynamic compensation of reactive power

in starting modes of electric drives / M.Y. Burbelo, A.V.

Gadai. Monograph. - Vinnytsia: VNTU, 2010. - 104 p.

(Ukr)

[15] Cong Zhang, Haoyong Chen, Zipeng Liang, Manlan

Guo, Dong Hua, Honwing Ngan, "Reactive Power

Optimization Under Interval Uncertainty by the Linear

Approximation Method and Its Modified Method", IEEE

Transactions on Smart Grid, vol.9, no.5, pp.4587-4600,

2018.

[16] Mircea Eremia, Mohammad Shahidehpour, "Voltage

and Reactive Power Control" in Handbook of Electrical

Power System Dynamics: Modeling, Stability, and

Control, pp.340-450, 2013.

[17] Bolong Shi, Chang Wu, Weizhen Sun, Wei Bao, Ruipeng

Guo, "A Practical Two-level Automatic Voltage Control

System: Design and Field Experience", 2018 International

Conference on Power System Technology

(POWERCON), pp.1397-1407, 2018.