HVDC System Fault Analysis Through Wavelet Analysis Technique

HVDC System Fault Analysis Through Wavelet Analysis Technique

Rahil Abrol

Post Graduate Student

School of Electronics & Electrical Engineering, Reg. No- 11300098, Lovely Professional University, Jalandhar, Punjab, India

Mr. Anshul Mahajan Asst. Professor

Department of Electrical Engineering, Lovely Professional University, Punjab, India

Abstract: Nowadays, power system has become more complex, interconnected and vary in sizes and configurations. Some of the amount of electricity is generated through non- conventional sources of energy. Transmission networks are commonly classified into four parts: transmission system, sub transmission system, primary distribution system and secondary distribution system. It is possible to go for high voltage (HVDC) transmission for long-distance power transfer through high voltage semiconductor devices. As, we know that the high voltage direct current (HVDC) is most important for large power transfer and high power demands. Maintenance of the power quality has become very difficult because of the large scale demands. Faults that occur on the system are classified as symmetrical and non- symmetrical faults. These faults may lead damage to HVDC system and other equipments of system due to high power transfer. Presently, the fast identification of fault is the one of the primary concern for the stability of any power system. For the fast identification of faults, WAVELET ANALYSIS technique is one of the best methods which is used to identify the different types of faults in HVDC transmission system. The faults that occur in HVDC system due to some disturbances can be classified by monitoring the signals both on AC and DC sides of the HVDC system. In this report, the modelling and the simulation of a typical network of HVDC system is consider with the help of WAVELET ANALYSIS. This WAVELET technique provides a proper and reliable solution for fault identification of fault and is used to improve the performance of the HVDC system using MATLAB/SIMULINK in power system block set.

Keywords: HVDC, Faults, Wavelet, Multi Resolution Analysis.

1. INTRODUCTION

   The use of application of electricity is getting started with the use of direct current. The very first power station was established in 1882 in New York (USA). That station was a first dc station and the power supplied by this station is 110V dc to an area of the 1.5 mile radius. In few years of the development of this power station, many dc stations were built. In the last few years the demand of power system increases day by day. There are some problems of long transmission system using ac are: voltage regulation, dynamic stability, steady state and transient state with different load conditions. To overcome with these problems of ac transmission system is replaced with the high voltage transmission system. With the use of high

   voltage dc (HVDC) transmission long distance power is possible. The control of the ac power over the line is possible through high voltage devices. HVDC link requires converter stations at each end of the line. Main equipment which is used in a converter station are transformers and thyristor valves. At the sending end of the converter station the thyristor valves act as rectifiers to convert the ac to dc which is transmitted over the line and at the receiving end of the converter stations the thyristor valves act as a inverters to convert dc to ac which is utilized at the receiving end of the line. In this HVDC system each converter can function as rectifier or inverter. For the fast identification of faults, WAVELET ANALYSIS technique is one of the best methods which is used to identify the different types of faults in HVDC transmission system.

2. MATLAB/SIMULINK MODEL OF 12 PULSE HVDC SYSTEM

   In this paper work we have considered a 12- pulse HVDC system in MATALB/Simulink environment. A 1000 MW DC interconnection is used to transmit power from 500 KV, 50 Hz network to 345 KV, 1000MVA, 50 Hz network. In this model AC networks represent the L-R equivalents with an angle of 80 degree at fundamental frequency of 50 Hz or 60 Hz and at the third harmonic.

   Figure 1 Simulink Model of 12 Pulse HVDC System

3. RESULT AND DISCUSSION

To identify and classify the different faults in HVDC system (i.e. AC faults and DC faults) wavelet transform is used. From the system, voltage and current signals are monitored at ac inverter side and dc rectifier side.

The following fault cases were simulated

1. Normal operating case

2. Dc line fault

3. Ac fault(LG) at inverter end

For each case following four signals were discussed.

Dc voltage, DC current, inverter side phase voltage and inverter side phase current. In this two signals were monitored an AC side and two signals at DC side of the system. After that the wavelet based extraction technique was applied to these signals to identify the faults. The following steps are used to identify the type of fault

Figure 2 Dc Voltage and Current Under Normal Condition

Figure 3 DC voltage and DC current for DC Fault

Figure 4 DC Voltage and DC Current for AC Fault

Figure 5 Phase Voltage and Current Signals for Normal Case

Figure 6 Phase Voltage and Current Signals for DC Fault

Figure 7 Phase Voltage and Current Signals for AC Fault

IV COMPARISON OF WAVELETS AND THEIR COEFFICIENTS

• Normal voltage case

  Figure 8 db4 for DC Voltage for Normal Condition

  Figure 9 sym4 for DC Voltage for Normal Condition Table 1 Comparisons of Coefficients of db4 and sym4 for DC

  Voltage for Normal Condition

  Coefficients/wavelets	db4	Sym4
  Mean	-0.3058	-3.11365
  Standard deviation	6552.5	6600.2

• Normal current case

  Figure 10 db4 for DC Current for Normal Condition

  Figure 11 sym4 for DC Current for Normal Condition Table 2 Comparisons of Coefficients of db4 and sym4 for DC

  Current for Normal Condition

  Coefficients/wavelets	db4	Sym4
  Mean	-0.013	0.0
  Standard deviation	8.839	8.829

  • DC fault for voltage case

    Figure 12 db4 for DC Voltage for DC Fault Case

    Figure 13 sym4 for DC Voltage for DC Fault Case

    Table 3 Comparisons of Coefficients of db4 and sym4 for DC Voltage for DC Fault Case

    Coefficients/wavelets	db4	Sym4
    Mean	63.206	176.89
    Standard deviation	16707	16040.25

  • DC fault for current case

    Figure 14 db4 for DC Current for DC Fault Case

    Figure 15 sym4 for DC Current under DC Fault Case

    Table 4 Comparisons of Coefficients of db4 and sym4 for DC Current for DC Fault Case

    Coefficients/wavelets	db4	Sym4
    Mean	0.00464	0.0460
    Standard deviation	8.3765	10.636

  • AC fault for voltage case

    Figure 16 db4 for DC Voltage under AC Fault Case

    Figure 17 sym4 for DC Voltage under AC Fault Case Table 5 Comprisons of Coefficients of db4 and sym4 for DC

    Voltage for AC Fault Case

    Coefficients/wavelets	db4	Sym4
    Mean	3.4481	-4.053
    Standard deviation	9342.55	3202.81

  • AC fault for current case

Figure 18 db4 for DC Current for AC Fault Case

Figure 19 sym4 for DC Current for AC Fault Case Table 6 Comparisons of Coefficients of db4 and sym4 for DC

Current for AC fault Case

Idc

3000

2000

1000

Idc

0

Normal DC fault AC fault

vdc

1.5

1

0.5

vdc

0

Normal DC fault AC fault

Figure 20 Variations in DC Voltage and Current under Three Conditions

Figure 20 shows that when DC side is observed whenever the disturbance occurs correspondingly the mean and standard deviation values of DC current and DC voltage deonised signals are increased with respect to disturbance. If it is DC fault the mean values and standard deviation of current and voltage value is increased and if the disturbance is AC fault correspondingly the mean values and standard deviation of current and voltage is more increased compared to normal operating condition.

1. CONCLUSION

   In this paper, a new technique wavelet based multi resolution analysis is used to extract the features of the signals with and without faults. This technique is also used to identify the fault that occurs in the system. So, that after the identification of fault we can provide the high speed of protection to the system which make the system more accurate, reliable and improve the performance of the system. In this dissertation technique we use two wavelets namely Dabuchies and Symlet at level 4 and decompose the signals at different levels and calculate the mean and standard deviation of the system with and without fault and compare the wavelet which one is more accurate. Results show that Symlet wavelet is more accurate as compare to Dabuchies.

2. REFERENCES

1. A.Keswani Rashmi Identification of Fault in HVDC Converters using Wavelet Based Multi-Resolution Analysis 2008 IEEE

2. Anuradha.V, Anitha.S, Apoorva.D.C, Priyanka.N, Somashekar.B Harmonic Analysis in HVDC System International Journal of Emerging Technology and Advanced Engineering (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 4, Issue 5, May 2014.

3. ATTHAPOL NGAOPITAKKUL The combination of discrete wavelet transform and fuzzy logic algorithm for fault classification of fault on transmission system International Journal of Emerging Technology and Advanced Engineering, October 2012

4. B. NAGU, NAVANEETHA. A, P.V. R. RAO Improvement of Power System Stability using Fuzzy Logic based HVDC Controls ISBN: 978-960-474-274-5.

5. Gale P. F, UMIST, Ge Yaozhong B. J.Cory J. Barker R.G. FAULT LOCATION BASED ON TRAVELLING WAVES.

6. Gupta B.R, power system analysis and design S.CHAND and COMPANY LTD., NEW DELHI.

7. Gaouda A.M., El-Saadany E.F., El-Saadany M.M.A. Salama Monitoring HVDC Systems Using Wavelet Multi-resolution Analysis.

8. Huang Shyh-Jier, Hsieh Cheng-Tao, Huang Ching-Lien Application of Morlet Wavelets to Supervise Power System Disturbances IEEE Transactions on Power Delivery, Vol. 14, No. 1, January 1999.

9. J Nandana, Pannala Krishna Murthy, Satya Durga VSC – HVDC Transmission System Analysis Using Neural Networks International Journal of Engineering Science and Innovative Technology (IJESIT) Volume 2, Issue 3, May 2013.

10. Kandil N., Sood V.K., Patel R.V. (1992) Fault identification in AC- DC transmission systemusing Neural Networks.