DonNTU Master's portal
Dmitrii Bogdanov


Dmitrii Bogdanov


Faculty of Computer Information Technology and Automatics
Department of Electronic Technics
Speciality: Electronic Systems



Theme of master’s work:
Design and justification of electronic broken phase detection system for three-phase industrial power nets with isolated neutral.


Scientific adviser: Alexander Kochin.


MASTER'S WORK ABSTRACT

1. Urgency topic

Requirements to electrical protection growing up nowadays, in particular, requipments to reducing of permissible touch voltage [3]. That requires development of new technical solutions for the protection of labor. One such solution is bypassing damaged phase by its automatic coupling with a grounding device [4]. This solution require a reliable and accuracy operation from damaged phase examination system (DPES). Most preferable DPES, Among the various circuit design which implementing DPES, is high speed digital devices,  based on the definition of initial phase relative to the neutral ground.

2. Purpose and objectives

The main objective of master's work is development damaged phase examination system for industrial three-phase power networks with isolated neutral, which main purpose is accurate and reliable detection phase with ground fault in mining three-phase networks with voltage up to 1 kV and issuing tripping commands to the protective bypass system.

Thus the main tasks of master's work are:

  • Analysis  of industrial three-phase network’s characteristics;
  • Analysis of the processes which taking place during the phase ground fault through a Hi-Z load (human);
  • Modeling of these processes;
  • Analysis of existing methods of  damaged phase, with the help of the constructed model;
  • Construction on the basis of data modeling working algorithm for DPES;
  • Full-scale testing of the system.

3. Scientific novelty

The method, described in this work, includes two basic approaches to the damaged phase examination methods based on magnitude and phase of three-phase net parameters. The main advantage of this method is speed improvement as compared with existing methods. That improvement can be achieved when fact that the damage is exist determined by the amplitude parameters of the network, and actually damaged phase exanimate by phase parameters.

4. The choice of solutions

4.1 Existing solutions

Phase and amplitude parameters of power network is used in the known methods of damage phase examination. The use of phase parameters ensures higher accuracy signal processing as well as useful information extracted with respect to their zero-point crossing that prevents recognition errors with small changes in the amplitude parameters for the limited the time of the conditions of electrical safety.

A method of damaged phase examination realized in [6], in which zero sequence voltage and  one of the line voltage is measured (using sensors) and get phase difference between them - primary phase voltage zero-sequence control line (with a threshold signal splitter ) values of the initial phase of zero sequence voltage reference phase intervals as large 90o, given for each phase of the network and involving a value of the initial phase of the supply phase voltage, phase and determine the damaged network - fixed phase of the network as corrupted by a match.

This method has low sensitivity to leaks because of the supporting the phase spans limitation which has the size of 90o. The actual change initial phase magnitude of the zero sequence, when one-phase danger leakage is exist, may exceed 90o because of influence produced by the natural asymmetry of the network insulation resistance. This is especially manifested in the high-leakage - leakage resistance, close to the asymmetric insulation resistance. As a result, output value of the specified initial phase beyond the phase of the reference period of the damaged phase of the network is not fixed. Increasing of the leaks determination sensitivity will require an expansion of support phase intervals.

4.2 Proposed solution

Timely and accurate determination of the mining power network phase with ground leakage, is a major factor in the protection of human, which connected to a phase voltage. Fast way to solution of this problem described in [1]. Its essence is based on the hypothesis of sinusoidal variation of the neutral potential u N (t) lies in the definition of its argument (initial phase) of two adjacent points of the instantaneous value of u N (t 1) and u N (t an + Δt). If the difference in phase between these points Δωt fixed and varies from 10 to 20 degrees, the problem reduces to the solution of the system.

Уравнение (1);                                                                   (1)
Уравнение (2)                                                                    (2)

where U Nmax, ψ, - respectively, the amplitude and initial phase (argument) of neutral relative to earth;

Eliminating from (1 and 2) the amplitude U Nmah, we obtain the function y (ψ) and its derivative y '(ψ):

Уравнение (3);                                                           (3)
Уравнение (4).                                                                      (4)

From the analysis of (3) and (4). First, during the two functions ψ of π and, consequently, on the interval 0 ... 2π dependence (3), in contrast to the system (1, 2) has two roots, one of which "extra". Its easy to set up a test in (1) or (2).

Secondly, the derivative y '(ψ) is always positive. Consequently, the function of y (ψ) is always increasing and within a period (0 ... π) has only one root, which can easily be distinguished by using the property function (3) that it has discontinuities of the second kind for the values of the argument ψ, meet the condition:

ω∙t1 + Δωt + ψ = 0,  π,  2π  и т. д.                                                            (5)

Lack of first option is a low accuracy is determined by the argument of ψ because of the influence, first of all, dynamic components of the transitional regime, which is accompanied by a process of diversion, the presence of higher harmonics, the transverse asymmetry of the network, and a number of other factors. Its main advantages are: relatively easy separation of roots depending (4), their identification and verification.

The second version of implementing digital DPES considers the availability exponential part of the transition process in the event of leakage and is based on the assumption that the instantaneous value of neutral respect to the law:

Закон (6),                         (6)

where the α - the time (angle) of leakage;

τ - time constant of exponential component.

Unlike the first version of the solution (6) requires four values of the neutral potential, taken at regular intervals Δωt. As a result, we arrive to a system of four equations (the number four unknowns ψ, U Nmax, τ, α.):

Закон (7),  (7)

where t 1 - time of first reference;

k = 0 ... 3 - point defining the next three frames.

From a formal point of view of four unknowns of the system (7) can be determined by Newton's method using the Jacobian matrix. [5] However, Newton's method "works" only in space, where the vector of roots (ψ, U Nmax, τ, α) is already allocated. The system (7) unknown quantities can be set to a wide range. For instance, the value of U Nmax depends on the leakage resistance can be in the range of 0.1 to 1 on the amplitude of the nominal phase voltage. The values of ψ, and α can be any, within the period, moreover, the moment of switching on the α adopted reference system is random. The time constant τ, which depends largely on the leakage resistance and total capacitance of cable communications, is in the range from 0.01 to 40 milliseconds. In these circumstances, the selection of roots is a task no less difficult than solving the basic system (7). In fact, it means that implement the second option with a simple and fast algorithm is technically difficult. Moreover, even if its implementation are not addressed his shortcomings related to the effect of higher harmonics and unbalance the various types of networks.
Animation, 18 frames, 19 sec, 126 kb.
Fig. 2. Process of injured human identification. Animation, 18 frames, 19 sec, 126 kb.

REFERENCE
  1. Чорноус Е.В., Об альтернативном варианте защиты шахтной участковой сети от однофазной утечки // Электричество. ­ 2009. ­ №8. ­ C.33-36.
  2.  Ягудаев Б.М., Шишкин Н.Ф., Назаров В.В. Защита от электропоражения в горной промышленности. – М., Недра, 1982, 152 с.
  3. Научные основы (теория) электробезопасности. /Малиновский А.А., Никонец Л.А., Голубов С.В., Шелех Ю.Л., Радченко В.Н., Никонец А.Л. / Под редакцией Никонца Л.А. – Львов: НВФ „Українські технології”, 2008 р. – 224 с
  4. Шуцкий В.И., Жидков В.О., Ильин Ю.Н. Защитное шунтирование однофазных повреждений электроустановок. – М.: Энергоатомиздат, 1986. – 152 с.:ил.
  5. Демидович Б.П., Марон И.А. Основы вычислительной математики. – М.: “Наука“, – 664 с.:илл.
  6. А.с. СССР N 1379857, H 02 H 3/16, опубл. в 1988, БИ N 9.
  7. А.с. СССР N 943959, H 02 H 3/16, опубл. в 1982, БИ N 26.
  8. Патент на винахід № 10320, УКРПАТЕНТ, H02H3/16, G01R31/08, 12.04.2005.
  9. Патент на изобретение № 2150167, Российская федерация МПК7  H02H3/16, G01R31/08, 10.06.1999. [Электронный ресурс] http://ru-patent.info
  10. Журахівський А.В., Кенс Ю.А., Мединський Р.В., Засідкович Н.Р. Особливості розрахунку напруги зміщення нейтралі трифазних електромереж. \\ Збірник наукових праць Донецького Національного Технічного університету. Серія: “Електротехніка і енергетика”, випуск 50: Донецьк: ДонНТУ, 2002. с.(102-106).
  11. В.С. Перхач. Математичні задачі електроенергетики. – Львів.: Видавництво при Львівському Державному Університеті видавничого об`єднання «Вища Школа», 1982. – 378с.
  12. Sri Kolla, Logan Varatharasa. Identifying three-phase induction motor faults using artifical neural networks. \\ISA Transactions, 39, Elsevir.p. 433-439. [Электронный ресурс] http://hinari-gw.who.int