CC BY
21
УДК 621.313.322-843.6
DOI: 10.25559/SITITO.14.201803.603-608
Cognitive information technologies in control systems
SYSTEM OF LOAD DISTRIBUTION OF SHIP DIESEL GENERATORS BASED ON NEURAL NETWORKS
Sergey E. Kuznetsov, Nikolay A. Alekseev, Andrey A. Vinogradov
Admiral Makarov State University of Maritime and Inland Shipping, Saint-Petersburg, Russia
СИСТЕМА РАСПРЕДЕЛЕНИЯ НАГРУЗКИ СУДОВЫХ ДИЗЕЛЬ-ГЕНЕРАТОРОВ НА ОСНОВЕ НЕЙРОННЫХ СЕТЕЙ
С.Е. Кузнецов, Н.А. Алексеев, А.А. Виноградов
Государственный университет морского и речного флота им. адмирала С.О. Макарова, г. Санкт-Петербург, Россия
I© Kuznetsov S.E., Alekseev N.A., Vinogradov A.A., 2018|
Keywords
Abstract
Ship diesel-generator set; ship power station; parallel work; load distribution; electric power; specific fuel consumption; fuel consumption; minimization of fuel consumption; steady mode; neural network; neural network control.
The dependence of the specific fuel consumption of the ship diesel-generator set on the generated electric power and the power factor is given. An algorithm for load distribution between parallel marine diesel generator sets is given, taking into account the dependencies of the specific fuel consumption on the generated power and the power factor of each marine diesel generator set. The possibility of using neural networks to solve the problem of minimizing the specific fuel consumption for parallel operation of ship diesel-generator sets in the steady state mode is considered. A neural network model for reducing fuel consumption is proposed by distributing the total electric load of a ship power plant between diesel generators operating in parallel. Such a model, based on the database obtained in the training process, which is implemented by microprocessor controllers with the algorithms of searching for a combination of capacities inherent in them for each of the parallel running generator, at which the minimum flow is reached, independently determines how it is necessary to load the units operating in parallel, to achieve a minimization of fuel consumption; and periodically checks the correspondence between the estimated (calculated) and the actual fuel costs, which ensures the control of the technical condition of the mechanisms.
Ключевые слова
Аннотация
Судовой дизель-генераторный агрегат; судовая электростанция; параллельная работа; распределение нагрузки; электрическая мощность; удельный расход топлива; расход топлива; минимизация расхода топлива; установившейся режим; нейронная сеть; нейросетевое управление.
Приведена зависимость удельного расхода топлива судового дизель-генераторного агрегата от вырабатываемой электрической мощности и коэффициента мощности. Приведен алгоритм распределения нагрузки между параллельно работающими судовыми дизель-генераторными агрегатами с учетом зависимостей удельного расхода топлива от вырабатываемой мощности и коэффициента мощности каждого судового дизель-генераторного агрегата. Рассмотрена возможность использования нейронных сетей для решения задачи минимизации удельного расхода топлива при параллельной работе судовых дизель-генераторных агрегатов в установившемся режиме. Предложена нейросетевая модель снижения расхода топлива путем распределения общей электрической нагрузки судовой электростанции между дизель-генераторами, работающими параллельно. Такая модель на основании базы данных, полученных в процессе обучения, которое реализуется микропроцессорными контроллерами с заложенными в них алгоритмами поиска комбинации мощностей, приходящихся на каждый из параллельно работающий генератор, при которых достигнут минимальный расход, самостоятельно определяет, как необходимо нагружать агрегаты, работающие в параллель, чтобы достичь минимизации расхода топлива; а также периодически проверяет соответствие между предполагаемым (рассчитанным) и фактическим расходами топлива, что обеспечивает контроль технического состояния механизмов.
Abouttheauthor^
Sergey E. Kuznetsov, Doctor of Technical Sciences, Professor, Admiral Makarov State University of Maritime and Inland Shipping (5/7 Dvinskay Str., Saint-Petersburg 198035, Russia), ORCID: http://orcid.org/0000-0002-8605-192X, KuznecovSE@gumrf.ru
Nikolay A. Alekseev, Candidate of Technical Sciences, Professor, Head of the Department, Admiral Makarov State University of Maritime and Inland Shipping (5/7 Dvinskay Str., Saint-Petersburg 198035, Russia), ORCID: http://orcid.org/0000-0001-6132-260X, kaf_saees@gumrf.ru Andrey A. Vinogradov, Associate Professor, Admiral Makarov State University of Maritime and Inland Shipping (5/7 Dvinskay Str., Saint-Petersburg 198035, Russia), ORCID: http://orcid.org/0000-0003-4192-4034, eric_esseker@mail.ru
Vol. 14, no 3. 2018 ISSN 2411-1473 sitito.cs.msu.ru
Modern Information Technologies and IT-Education
604
Когнитивно-информационные технологии С.Е. Кузнецов,
в системах управления Н.А. Алексеев, А.А. Виноградов
Introduction
To date, in the field of transport and operation of energy systems, the question of minimizing fuel consumption remains topical. The study of issues related to the reduction of fuel consumption during the operation of diesel power plants is given in [1], [2], [3], [4]. For example, in the field of water transport, ship power plants, equipped with two or more marine diesel generator sets (SDG) units operating in parallel, are set forth by the classification societies (Russian Maritime Register of Shipping (RMRS)) classification and construction of ships [Part XI «Electrical Equipment» ND №2-020101104, 2018. - 138 p.]. According to these requirements, in parallel operation of alternators in the range from 20 to 100% of the total load, its distribution to each generator should occur in proportion to their capacities and should not differ by more than 15% from the calculated load of the larger generator or 25% from the calculated load of the generator under consideration, whichever is less [5]. Currently, the total load is distributed between parallel-working SDG-units according to these requirements, but without taking into account the dependence of the specific fuel consumption (URT) on the generated electric power of each SDG-unit.
Main part
The minimum fuel consumption of a single diesel generator set when the load changes is determined by its dependence of specific fuel consumption q on the generator electric power output Pg, originally obtained from factory tests of a serviceable diesel generator using standard fuel and given in the data sheet of each SDG-unit. The minimum specific fuel consumption usually corresponds to the load range of 75-85% of the nominal electric power of the unit, and going beyond this range results in excessive fuel consumption. In the passport data is usually given the dependence of the specific fuel consumption only from the active electric load diesel generator unit, that is, when the power factor is equal to unit cos^ = 1. With a decrease in cos^ of the load, the specific fuel consumption increases.
In the process of long-term operation of SDG-units, there is a change in the dependence of the specific fuel consumption on the output power of the diesel (generator), that is, an increase in the specific fuel consumption and a shift in the range of loads corresponding to the minimum specific fuel consumption (SFC). In addition, even for single-type diesel generator sets of the same rated power and assembly before commissioning (or after overhaul), the dependence of the SFC on the relative power of the generator
P = P
P
and the power factor cos^, as a rule, do not coincide (PGN is the rated power generator power), and for SDG-units of unequal power, the difference in the dependences may be more significant. In fig. 1 shows the URT dependencies on power for two ship SDG units with brush-less synchronous generators of the same assembly and the same power (PGN = 750 kW, Un = 440 V) with a power factor of cos^ = 1, built according to factory test data before commissioning. Questions on the dependence of fuel consumption on the load, the nature of the load, period of operation, quality of electricity, etc., are displayed in [6], [7], [8], [9], [10].
For generators operating in parallel, the distribution of the total load of the shipboard power plant in strict proportion to their nominal capacities may not be effective in terms of minimizing specific fuel
consumption due to the uneven nature of the SFC change from electric power and power factor; and during long-term operation of SDG-units, these dependences change, and the difference between them can increase significantly.
Rules for the classification and construction of ships RMRS deviations in load distribution between parallel generators are allowed in proportion to their capacities, therefore, taking into account the difference in SFC dependencies on power, the load between generators can be divided taking into account these dependencies so that the total specific fuel consumption is less than if the load was shared strictly proportionally.
qj«
q2m
0.5
Fig. 1. The dependence of the specific fuel consumption of SDG-unit from the power load Рис. 1 Зависимость удельного расхода топлива ДГА от мощности нагрузки
To do this, it is necessary to know the regression dependence of the specific fuel consumption q (PG) on the value of the electric power PG and power factor cosф for each SDG-unit operating in parallel, and taking into account their technical condition at the considered time point.
In accordance with the methodology described in [11], [12], the dependence of the specific fuel consumption of the SDG unit on the generator power and cosф can be represented as follows:
q( Pr) =
cos4 ф
'Pr3 +-
here q (PG) =
cos ф
10 3Q
■ Pr2 + (k5 +
cos2 ф
k
) ■ Pr+ кз + P
Pr
(gr/kWh) - specific fuel
consumption of SDG-unit; Pg (kW) - synchronous generator power; Q (kg/h) - fuel consumption; cos ip - power factor of load; ki + fo - constant coefficients depending on the technical
5
Современные информационные технологии и ИТ-образование
Том 14 № 3 (2018) ISSN 2411-1473 sitito.cs.msu.ru
Cognitive information technologies in control systems
state of SDG-unit: k{ =
10
15
9 U
К г ■ КД2 (gr/kW4 h],
2-Ю9 ,
К = ■ Кг ■ КД2 (gr/kw3 h], Âr3 =10 -кт,
(gr/kW h), k, = 103-Qconst, (gr/ h], k5 = 103 ■ КД2,
10 *-Кг-КД1
(gr/kW2h], k6=--[gr/kW2h], Kr
(kW/A2), Kfli Ckg/kW h), Kfl2 (kg/kW2 h); Q fkg/h) - fuel consumption; U (V) - generator voltage; Qconst (kg/h) - the constant component of fuel consumption corresponding to the fuel consumption at idle SDG-unit.
To search for such a combination of electrical power, falling on each of the parallel working SDG-units, in which the total specific fuel consumption would be minimal than with a strictly proportional load distribution between the SDG-units, the problem of choosing the power of parallel operating SDG-units ( PGt and PG2) satisfying the conditions:
P = P + P
l l n t- Jr2
(2)
q(P) = [q(Pri ) + q(Pr2 )] ^ min
(3)
corresponding to the minimum total URT, with known electrical power power plants Pj, which can be performed by modern microprocessor tools developed by Russian and foreign companies for maritime fleet. Such a database can be formed by a neural network for a certain period of time in use to solve the problem of minimizing the fuel consumption of the microprocessor controller, in which the above-mentioned algorithm is laid (Fig. 2). After a set of sufficient numbers (combinations - Pj PG1; PG2; qmin), that is, after the training stage is completed, the neural network itself determines how it is necessary to load the SDG-units operating in parallel to achieve minimization of SFC. As noted above, the coefficients that determine the dependence q (P) for SDG-units depend on the technical state of the mechanisms and change with time.
The neural network forms a database for fixed values of coefficients (KG, KD1, KD2, Qconst) for each SDG-unit, received at some moment of time, and does not take into account their change over time, which is its disadvantage. To eliminate this drawback, it is proposed to compare the results calculated by the neural network (minimum total SFC) and the actual values of the SFC with a given periodicity, and if these values diverge more than the set limit, generate a signal for the correction of coefficients (KG, KD1, KD2, Qconst), conduct retraining of the neural network with adequately new coefficient values (KG, KD1, KD2, Qconst). Thus, with the help of such a model of a cognitive-information complex based on a neural network, the technical state of SDG aggregates can also be monitored.
The enlarged scheme of the algorithm for solving this problem is shown in Fig. 2 — a continuous measurement of the power and cos^ of the shipboard power plant is made, the measurement results and the regression dependencies (1) for each SDG unit are entered into a computer model in which, based on the analysis, load distributions between parallel SDG-units for a given power of a marine power plant and cos^ is determined by the load distribution with the minimum total specific fuel consumption.
According to the studies carried out in [11], [12], the coefficients KG, KD1, KD2, Qconst in expression (1) are determined on the basis of periodic tests of the SDG-unit at different loads. When the voltage U, the frequency f and the power factor cos^ is constant, the coefficients KG, KD1, KD2, Qconst determine the dependence q (PG). It is important to note that the values of the coefficients KG, KD1, KD2, Qconst during operation will change due to changes in the technical state of the SDG-unit, and therefore periodic correction of these coefficients is necessary, for example, during maintenance or after repair. According to known values KG, KD1, KD2, Qconst for each parallel SDG-unit and the value of the total electric power P in the steady state, the task of finding a combination of power (PG1 and PG2) for parallel SDG-units, corresponding to the minimum total specific consumption fuel can be solved using a convergent cognitive-informational complex based on neural networks for the ship's power station, equipped with two or more SDG-units operating in parallel. In addition, using such a complex, it is possible to predict the above factors using the database of SDG aggregates obtained during operation, which will allow to correct the dependence q (P) for each SDG-unit The reliability of the q (P) dependence is important for increasing the accuracy of minimizing the total SFC. The use of neural network modeling seems to be the most promising for modeling input-output ratios of parameters in ship power systems. Implementing a neural network providing the minimum total fuel consumption at a given value of the total power of the ship power station in the steady state is possible with a database of previous power measurements of the parallel operating SDG-units PG1 and PG2
Math modeling: []G1=f[PG1, со5ф); qG2=F[PG2, соэф); P=PG1+PG2 q=qG1+qG2
Fig. 2. A block diagram of the load distribution algorithm between parallel operating ship diesel generator sets Рис.2. Блок-схема алгоритма распределения нагрузки между параллельно работающими судовыми дизель-генераторными агрегатами
Vol. 14, no 3. 2018 ISSN 2411-1473 sitito.cs.msu.ru
Modern Information Technologies and IT-Education
Когнитивно-информационные технологии С.Е. Кузнецов,
в системах управления Н.А. Алексеев, А.А. Виноградов
In fig. 3 shows a neural network model for minimizing fuel consumption.
n inner hidden layers
Fig. 3. Neural network model to minimizefuel consumption Рис. 3. Нейросетевая модель минимизации расхода топлива
The model consists of the input layer, the n-th numb er of inner hidden layers, the output layer, the comparison element EC. In the steady state mode, the signals from the sensors come to the model input: - the total electrical power of the shipboard power station P and the value of the power factor coscp. Based on the generated database, the trained neural network model sets the values of P and PG2 at which the total specific fuel consumption q^ is minimal and generates signals CS1 and CS2 - control signals for setting the fuel consumption corresponding to the minimum fuel consumption for a given load.Control signals CS1 and CS2 change the position of the fuel rail SDG-units. The minimum SFC qa calculated by the model is compared with the actually measused SFC, qface value and in the EC comparison unie: when the difference between the calculated and measured SFC values exceeds the set limit, a significant change is made in the technical state of the SDG-units "S" and the need to retrain the neural network.
In the works [13], [14] it is shown how the reduction of fuel consumption of diesel generator sets in systems with variable speed is carried out.
Conclusion
Automatic control using neural networks of one process or another is a promising direction in the development of modern technology; and is currently being actively tested and used in engineering [15]. Most of the difficulties encountered in the operation of marine diesel generators are associated with a large number of controlled parameters and the inability to control their change in full. In such conditions, neural network control is effective for solving certain technical problems. So in this work, the neural network model, previously trained, provides minimal fuel consumption and monitors changes in the technical state of SDG-units even in the presence of interference in the input data or in conditions of lack of information. Also, issues related to optimization, including the use of intelligent systems, the operation of autonomous diesel power plants for fuel economy, are discussed in [16] - [20].
References
[1] Meyer R.T. et al. Hybrid Optimal Power Management of a Ship. ASME 2014 International Mechanical Engineering Congress and Exposition. Vol. 4B: Dynamics, Vibration, and Control, pp. V04BT04A007, 2014. 10 p. DOI: 10.1115/IMECE2014-38181
[2] Giannoutsos S.V., Manias S.N. Energy management and D/G fuel consumption optimization in the power system of marine vessels through VFD-based process flow control. 2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC). Rome, pp. 842-850, 2015. DOI: 10.1109/EEEIC.2015.7165274
[3] Lundh M. et al. Estimation and optimization of vessel fuel consumption. IFAC-PapersOnLine. 2016; 49(23):394-399. DOI: 10.1016/j.ifacol.2016.10.436
[4] Skovorodnikov E.I., Lifanov M.V., Anisimov A.S. The comparative analysis of the overall performance diesel engines PD1M and 1-PD4D under operating conditions. Izvestia Transsiba = Journal of Transsib Railway Studies. 2010; 3(3):32-44. (In Russian)
[5] Rules for the classification and construction of ships [Pravi-la klassifikacii i postrojki morskih sudov]. SPb.: RMRS, 2015. 807 p. (In Russian)
[6] Kuznetsov S.Ye., Kudryavtsev Yu.V., Vinogradov A.A. Estimation of electricity quality at the vessel in the operation process. Vestnik gosudarstvennogo universiteta morskogo i rech-nogo flota imeni admirala S.O. Makarova. 2017; 9(6):1320-1329. (In Russian) DOI: 10.21821/2309-5180-2017-9-61320-1329
[7] Kuznetsov S.Ye. Losses and efficiency of ship synchronous generator. Ekspluatatsiya morskogo transporta. 2009; 3:6771. Available at: https://elibrary.ru/item.asp?id=12947687 (accessed 02.07.2018). (In Russian)
[8] Vinogradov A.A. et al. Specific fuel consumption of ship diesel generator. Marine technology and engineering. Safety of the marine industry. Kaliningrad: KGTU, pp. 141-143, 2017. (In Russian)
[9] Kuznetsov S.Ye., Kudryavtsev Yu.V. Effect of load and power factor on the efficiency of the ship's synchronous generator. Sudostroyeniye. 2011; 5:29-32. Available at: https://elibrary. ru/item.asp?id=17035265 (accessed 02.07.2018). (In Russian)
[10] Kuznetsov S.Ye., Bashkirev O.O., Vinogradov A.A. Specific fuel consumption and ship diesel generator losses. Transport Business of Russia. 2017; 5:176-179. Available at: https://eli-brary.ru/item.asp?id=30734732 (accessed 02.07.2018). (In Russian)
[11] Kuznetsov S.Ye., Kudryavtsev Yu.V. Effect of load and power factor on fuel consumption of marine diesel generator set Sudostroyeniye. 2011; 6:30-32. Available at: https://elibrary.ru/ item.asp?id=17255608 (accessed 02.07.2018). (In Russian)
[12] Alekseev N.A., Vinogradov A.A., Kuznetsov S.Ye. Sposob raspredeleniya nagruzki mezhdu parallel'no rabotayushchimi sudovymi dizel'-generatornymi agregatami [The method of load distribution between parallel working ship di-generator sets]. Patent RF, no. 2645387, 2018.
[13] Dar'yenkov A.B., et al. Research of control system of frequency of rotation of the shaft diesel generator hybrid wind farm. Trudy NGTUim R.E. Alekseyeva. 2015; 3:231-237. (In Russian)
[14] Dar'yenkov A.B., Khvatov O.S. Control system of autonomous
Современные информационные технологии и ИТ-образование
Том 14 № 3 (2018) ISSN 2411-1473 sitito.cs.msu.ru
Cognitive information technologies in control systems
diesel generator characterized by alternating frequency rotation of shaft. Trudy NGTUim R.E. Alekseyeva. 2013; 5:303-3-8. (In Russian)
[15] Khvatov O.S., Dar'yenkov A.B., Polyakov I.S. Neural Network Algorithm of Control System of Fuel Supply of Frequency Rotation Diesel Generator Set. Vestnik of Ivanovo State Power Enginiring University. 2013; 3:50-53. Available at: https:// elibrary.ru/item.asp?id=19125155 (accessed 02.07.2018). (In Russian)
[16] Rao K.S., Chauhan P.J., Panda S.K., Wilson G., Liu X., Gupta A.K. Optimal scheduling of diesel generators in offshore support vessels to minimize fuel consumption. IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society, Yokohama, 2015. Pp. 004726-004731. DOI: 10.1109/ IEC0N.2015.7392838
[17] Chauhan P.J., Rao K.S., Panda S.K., Wilson G., Liu X., Gupta A.K. Fuel efficiency improvement by optimal scheduling of diesel generators using PSO in offshore support vessel with DC power system architecture. 2015 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC). Brisbane, QLD, pp. 1-6, 2015. DOI: 10.1109/APPEEC.2015.7380963
[18] Rao S.S. Engineering optimization: theory and practice. 4th ed. John Wiley & Sons, 2009. 813 p.
[19] Wu W., Wang D., Arapostathis A., Davey K. Optimal power generation scheduling of a shipboard power system. IEEE Electric Ship Technologies Symposium (ESTS 2007). Arlington, Virginia, U.S.A., May 21-23, pp. 519-522, 2007. DOI: 10.1109/ ESTS.2007.372135
[20] Khan M.M.S., Faruque M.O. Energy storage management for MVDC power system of all electric ship under different load conditions. 2017 IEEE Electric Ship Technologies Symposium (ESTS). Arlington, VA, pp. 192-199, 2017. DOI: 10.1109/ ESTS.2017.8069280
Submitted 02.07.2018; revised 10.09.2018; published online 30.09.2018.
Список использованных источников
[1] Meyer R.T. et al. Hybrid Optimal Power Management of a Ship // ASME 2014 International Mechanical Engineering Congress and Exposition. Vol. 4B: Dynamics, Vibration, and Control. 2014. Pp. V04BT04A007. 10 p. DOI: 10.1115/ IMECE2014-38181
[2] Giannoutsos S.V., Manias S.N. Energy management and D/G fuel consumption optimization in the power system of marine vessels through VFD-based process flow control // 2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC). Rome, 2015. Pp. 842-850. DOI: 10.1109/EEEIC.2015.7165274
[3] Lundh M. et al. Estimation and optimization of vessel fuel consumption // IFAC-PapersOnLine. 2016. Vol. 49, issue 23. Pp. 394-399. DOI: 10.1016/j.ifacol.2016.10.436
[4] Сковородников Е.И., Лифанов М.В., Анисимов А.С. Сравнительный анализ эффективности работы дизелей ПД1М и 1-ПД4Д в условиях эксплуатации // Известия Транссиба. 2010. № 3(3). С. 32-44.
[5] Правила классификации и постройки морских судов. СПб.: РМРС, 2015. 807 с.
[6] Кузнецов С.Е., Кудрявцев Ю.В., Виноградов А.А. Оценка качества электроэнергии на судне в процессе эксплуатации // Вестник Государственного университета морского и речного флота имени адмирала С.О. Макарова. 2017. Т. 9, № 6(46). С. 1320-1329. DOI: 10.21821/2309-5180-2017-96-1320-1329
[7] Кузнецов С.Е. Потери и коэффициент полезного действия судового синхронного генератора // Эксплуатация морского транспорта. 2009. № 3. С. 67-71. URL: https://elibrary. ru/item.asp?id=12947687 (дата обращения: 02.07.2018).
[8] Виноградов А.А. Удельный расход топлива судового дизель-генератора / С.Е. Кузнецов, А.А. Виноградов // Тезисы докладов V МНК «Морская техника и технологии. Безопасность морской индустрии», 21-27 мая 2017. Калининград: КГТУ 2017. C. 141-143.
[9] Кузнецов С.Е., Кудрявцев Ю.В. Влияние нагрузки и коэффициента мощности на коэффициент полезного действия судового синхронного генератора // Судостроение. 2011. № 5. С. 29-32. URL: https://elibrary.ru/item. asp?id=17035265 (дата обращения: 02.07.2018).
[10] Кузнецов С.Е., Башкирев О.О., Виноградов А.А. Удельный расход топлива и потери мощности судового дизель-генератора // Транспортное дело России. 2017. №. 5. C. 176179. URL: https://elibrary.ru/item.asp?id=30734732 (дата обращения: 02.07.2018).
[11] Кузнецов С.Е., Кудрявцев Ю.В. Влияние нагрузки и коэффициента мощности на расход топлива судового дизель генераторного агрегата // Судостроение. 2011. № 6. С. 3032. URL: https://elibrary.ru/item.asp?id=17255608 (дата обращения: 02.07.2018).
[12] Способ распределения нагрузки между параллельно работающими судовыми дизель-генераторными агрегатами: пат. 2645387 Российская Федерация / А.Н. Алексеев, А.А. Виноградов, С.Е. Кузнецов, заявитель и патентообладатель Государственный университет морского и речного флота имени адмирала С.О. Макарова. № 2017118845; заявл. 30.05.2017.
[13] Дарьенков А.Б. и др. Исследования системы управления частотой вращения вала дизель-генератора гибридной ветроэлектростанции // Труды НГТУ им. РЕ. Алексеева. 2015. № 3. С. 231-237.
[14] Дарьенков А.Б., Хватов О.С. Система управления автономным дизель-генератором переменной частоты вращения // Труды НГТУ им. РЕ. Алексеева. 2013. № 5. С. 303-308.
[15] Хватов О.С., Дарьенков А.Б., Поляков И.С. Нейросетевой алгоритм системы управления топливоподачей дизель-генератора переменной скорости вращения // Вестник Ивановского государственного энергетического университета. 2013. № 3. С. 50-53. URL: https://elibrary.ru/item. asp?id=19125155 (дата обращения: 02.07.2018).
[16] Rao K.S., Chauhan P.J., Panda S.K., Wilson G, Liu X, Gupta A.K. Optimal scheduling of diesel generators in offshore support vessels to minimize fuel consumption // IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society, Yokohama, 2015. Pp. 004726-004731. DOI: 10.1109/ IEC0N.2015.7392838
[17] Chauhan P.J, Rao K.S., Panda S.K., Wilson G, Liu X, Gupta A.K. Fuel efficiency improvement by optimal scheduling of diesel generators using PSO in offshore support vessel with DC power system architecture // 2015
Vol. 14, no 3. 2018 ISSN 2411-1473 sitito.cs.msu.ru
Modern Information Technologies and IT-Education
Когнитивно-информационные технологии С.Е. Кузнецов,
в системах управления Н.А. Алексеев, А.А. Виноградов
IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC). Brisbane, QLD, 2015. Pp. 1-6. DOI: 10.1109/APPEEC.2015.7380963
[18] Rao S.S. Engineering optimization: theory and practice. 4th ed. John Wiley & Sons, 2009. 813 p.
[19] Wu W., Wang D., Arapostathis A., Davey K. Optimal power generation scheduling of a shipboard power system // IEEE Electric Ship Technologies Symposium (ESTS 2007). Arlington, Virginia, U.S.A., May 21-23, 2007. Pp. 519-522. DOI: 10.1109/ ESTS.2007.372135
[20] Khan M.M.S., Faruque M.O. Energy storage management for MVDC power system of all electric ship under different load conditions. 2017 IEEE Electric Ship Technologies Symposium (ESTS). Arlington, VA, 2017. Pp. 192-199. DOI: 10.1109/ ESTS.2017.8069280
Поступила 02.07.2018; принята в печать 10.09.2018; опубликована онлайн 30.09.2018.
|об авторах:|
Кузнецов Сергей Емельянович, доктор технических наук, профессор, Государственный университет морского и речного флота им. адмирала С.О. Макарова (198035, Россия, г. Санкт-Петербург, ул. Двинская, д. 5/7), ORCID: http://orcid.org/0000-0002-8605-192X, KuznecovSE@gumrf.ru Алексеев Николай Андреевич, кандидат технических наук, профессор, Государственный университет морского и речного флота им. адмирала С.О. Макарова (198035, Россия, г. Санкт-Петербург, ул. Двинская, д. 5/7), ORCID: http://orcid.org/0000-0001-6132-260X, kaf_saees@gumrf.ru Виноградов Андрей Александрович, доцент кафедры, Государственный университет морского и речного флота им. адмирала С.О. Макарова (198035, Россия, г. Санкт-Петербург, ул. Двинская, д. 5/7), ORCID: http://orcid.org/0000-0003-4192-4034, eric_esseker@mail.ru
This is an Open Access article distributed under the terms ofthe Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted reuse, distribution, and reproduction in any medium provided the original work is properly cited.
Современные информационные технологии и ИТ-образование
Том 14 № 3 (2018) ISSN 2411-1473 sitito.cs.msu.ru
