The use of PIC Microcontrollers for Fault Classification of a Series Compensated Transmission Line

International Journal of Computer Systems (ISSN: 2394-1065), Volume 02 Issue 12, December, 2015 Available at http://www.ijcsonline.com/ The use of PIC Microcontrollers for Fault Classification of a Series Compensated Transmission Line El Sayed M. Tag Eldin A, Nivin Ghamry B A Electric Power and Machines Department, Cairo University, Faculty of Engineering, Cairo, Egypt. B Department of Information Technology, Cairo University Faculty of Computers and Information, Cairo, Egypt Abstract Microcontrollers have recently become an essential embedded system in protective devices used for medical and industrial instrument or other costly instrument which need to be saved from dangerous faults. This paper presents a PIC-microcontroller-based control approach for the protection of a three-phase power transmission line with series compensation. The microcontroller is programmed for detecting faults in the line and isolating the system or instrument connected to it through a digital relay which is interfaced to the microcontroller. Fault diagnosis is performed using wavelet transform and multi resolution analysis on the line currents at both ends. The directions and magnitudes of spikes of the obtained MRA wavelet coefficients are employed for fault detection. These information are fed to the microcontroller which includes a program for fault classification. The microcontroller accordingly sends a signal to the digital relay to trip the circuit and a signal to an LCD to display the type of fault that has occurred. The proposed protection scheme is simulated using Matlab Simulink. Matlab simulation results show that the scheme is able to detect all types of internal faults at different locations. The microcontroller used in this work is the high performance enhanced flash PIC18F4520 of microchip which has an on chip analog to digital converter peripheral. C and assembly language programs are developed to control the function of the digital relay. MPLAB Integrated Development Environment (IDE) is applied for the development of the proposed embedded applications employing PIC microcontroller. Keywords: microcontroller, PIC, series compensated transmission lines, fault location, fault recognition discrete wavelet transforms. I. INTRODUCTION Programmable interface controller (PIC) or microcontrollers is a family of modified Harvard architecture microcontrollers made by Microchip Technology and derived from the PIC1650 originally developed by General Instrument s Microelectronics Division [1]. Microcontrollers are used as embedded devices in automatically controlled applications, such as automobile engine control systems, remote controls, appliances, electrical and mechanical power tools etc. They have the advantage of reducing the size, cost, and power consumption compared to designs applying separate microprocessor, memory, and input/output devices. These features offered by microprocessor technologies encouraged the evolution of microprocessor-based approaches in industrial applications like the protection of power systems [2]. Microprocessor based protections have the benefits of adaptive and programmable logic, selftesting and monitoring with a high degree of accuracy. These features make them meet important requirements like high reliability, flexibility and fast operation compared to conventional protection schemes based on artificial intelligence like neural networks or fuzzy logic [3]. The challenge of system protection is the innovation of products that probably detect faults and initiate fast disconnection to the faulty components. Undesirable faults can be short circuits, over current, overvoltage, etc. Electromechanical relays [4], which have been the core of the power system protection for years and still has no significant drawbacks on their protection functions. Nevertheless, they are now being replaced by microprocessor-based digital relays in more advanced protection schemes. The components of digital relays are hardware and software. The characteristic is defined in the software, and can be reconfigured to provide optimum protection. The hardware consist the microcontroller which operates according to the software programming and monitors the operating state of the power system on a digital display. In the following a sample of recent research work is demonstrated. In [5] full design of alerting system using PIC16F877A microcontroller for power distribution network is illustrated. In [6] the design and construction of overcurrent and directional overcurrent relays with ground fault protection for the protection of three-phase sub-transmission and distribution systems, using a 16-bit microprocessor Intel 8096BH is described. More recent work can be found in [7, 8]. This work focuses on the design of a PIC microcontroller- based relay and integrating it in the protection system for series compensated transmission lines developed in [9]. The protection scheme involves fault detection and classification. As most of the fault information is included in the transient components discrete wavelet transform (DWT) is applied for transient analysis. During the occurrence of faults, the grid current and voltages undergo transients. These transients are 552 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015

analyzed and the directions and magnitudes of the spikes of the wavelet coefficients are determined to detect faults and discriminate between internal and external faults. Those signals are given in the PIC microcontroller. The program fed into the microcontroller compares the directions of the spikes of the wavelet coefficients with a set of predefined thresholds. The faults are then classified and the faulty phase(s) are detected. Accordingly, the microcontroller delivers a signal to the interfaced relay to trip the circuit and also send a parallel signal to the LCD to display the type of fault that has occurred. This paper is organized as follows: Section II gives a brief overviews of the theory of DWT. In section III the proposed fault location algorithm applying DWT for series compensated transmission power system is explained. In Section IV simulations using MATLAB are illustrated. Section V presents the proposed microcontroller-based internal fault classification, which includes tripping the circuit through the interfaced relay and monitoring the status of the power system on the LCD. II. A. Discrete Wavelet Transform DISCRETE WAVELET TRANSFORM Discrete Wavelet transform (DWT) has been widely used as powerful time-frequency method for transient analysis in signal and image processing. The main advantage of DWT over Fourier Transform (FT) is that the size of analysis window varies in proportion to the corresponding frequency band which offers a better compromise in terms of localization. Wavelet transform of transient signal is expressed by multi-resolution decomposition fast algorithm which utilizes the orthogonal wavelet bases to decompose the signal to components under different scales. The series of wavelet coefficients obtained correspond to a time domain signal that covers a specific octave of frequency band containing more detailed information. This is accomplished by using the analyzing wavelet functions, called mother wavelets. Figure 1 shows some known mother wavelet functions. An appealing procedure for implementing DWT was presented by S. Mallat in 1989[10]. Mallat s algorithm is a recursive computation of wavelet coefficients. The formulation of the algorithm is related to the theory of filter banks advocated by Strang [11]. Fig. 2 Tree structure implementation of DWT using filter-banks. The wavelet transform is efficiently implemented by using a series of low pass filters (averaging) and high pass filters (differencing) at successive decomposition levels. The approximations are the high scale, low frequency components of the signal produced by the low pass filter which delivers a smoothed version of the input signal. The details are the low scale, high frequency components of the signal produced by the high pass filter. The band width of both filters is the same. The smoothed signal is decomposed over a set of bases functions (the scaling functions) while the detail signal is decomposed over the wavelet basis. The results of filter convolution are downsampled by a factor two (decimated) and the same filters are applied to the output of the low pass filter from the previous level. Fig. 2 shows the tree structure implementation of filter-banks for one-dimensional DWT, where h(n) stands for the high-pass filters, g(n) for the lowpass filters, and the arrows for the down sampling process. Many applications of DWT for analyzing power system transient signals and transmission line fault have been recently reported in the literature [12,13]. III. DWT FOR ANALYZING SERIES COMPENSATED TRANSMISSION POWER SYSTEM A. Model of Series compensated transmission power system Figure 3 shows the schematic diagram for the system under study[9]. It has been simulated using the MATLAB power system toolbox. The transmission system is composed of two sections of 500 kv lines connected in series. Each section is 300 km long. Both sections are modeled via the distributed parameter model. The values of resistances, inductances, and capacitances per unit length of the lines are given in Table 1. Fig. 1 Mother wavelets (a) Haar, (b) Coiflet, (c) Daubechies, (d) Morlet. Fig. 3 Transmission line system 553 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015

The total reactance in zero sequence of each line is 88.0 Ohms. To compensate 60% of the reactance, a 60 mf series capacitor is placed in the center of each section. A metal oxide varistor is placed in parallel with the capacitor to protect it against over-voltage. A 330 MVA load with lagging power factor of 0.8 is connected between both lines. Two 500 kv sources are connected at the far ends of both lines. The nominal system frequency is equal to 50 Hz. TABLE I. Components ARAMETERS OF TRANSMISSION LINE Positive sequence Zero sequence Resistance Ω/km 0.01273 0.3864 Inductance mh/km Capacitance nf/km 0.9337 4.1264 12.74 97.751 B. The Proposed DWT Algorithm for fault detection Mother wavelets are characterized by properties such as orthogonality, compact support, symmetry and vanishing moment. These properties as well as the similarity between signal and mother wavelet are considered in selecting a mother wavelet. In the study of analyzing power system transients, the properties of compact support and vanishing moment were used to select to most optimal mother wavelet for this analysis. It was concluded that Daubechies wavelets Db, Coiflet and B-spline were equally suitable in detecting power system transients. In this paper Db4 are used as the mother wavelets. A sampling rate of 12.5 khz khz (250 samples per cycle) is selected to capture transient components and extract its features. The three phase currents at both ends B1 and B2 are measured through current transformers. Based on this sampling time, the signal is decomposed into 12 levels. The frequency band for coefficients of the 6 th level d6 of the fault current signal is 97.65 195.31 Hz, which covers the frequency components corresponding to the second and third harmonics. Directional relaying concept for fault detection at fault instant is used. Spikes are detected in the sixth level detail wavelet coefficients d6. The magnitudes and directions of the spikes are determined, and accordingly one of the following actions is taken. 1) If there are no spikes, or the magnitude of the spikes is less than a predefined threshold, then the relay output is no fault. 2) If the spikes are of the same direction, then the relay output is external fault. 3) If the spikes are of different directions, then the relay output is internal fault. The summation of the d6 coefficients, namely S A, S B and S C of the original fault currents of the three phases I A, I B and I C is calculated. These signals are used for the purpose of classification of faults on the transmission line. IV. SIMULATION RESULTS In order to test the validity of the proposed protection algorithm, the model with the mentioned system parameters is designed using Matlab/Simulink. Different fault types, like single line to ground, double line to ground and double line are considered. Furthermore, three-phase symmetrical and three phase to ground faults at different locations in front of the series compensator, as well as in back, and different inception angles are extensively investigated. The algorithm is tested on hundred normal cases with different loadings and small disturbances. Six different fault inception angles are employed individually. Ten fault types with different faulty phases are employed. Fig. 4 and 5 show the sixth detail wavelet coefficients of the three phase currents at each busbar for two cases: an external and an internal fault. In Fig. 4 an external fault occurs between Busbar B2 and Busbar B3 at inception angle of 18 and fault resistance of 1000 Ω. Both figure shows that the spikes in each phase are of the same direction. On the other hand, in Fig. 5 a single line to ground fault occurs at the center of the transmission line at distance 149km from Busbar 1 at inception angle of 18 and fault resistance of 1000 Ω. The spikes in each phase at both ends are of different directions. Fig. 5 shows that the magnitudes of the spikes in phase A are larger than in phases B and C. Fig. 4 Wavelet coefficients of the three phase currents for external single phase A to ground. 554 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015

wide operating voltage range (2.0V to 5.5V), industrial and extended temperature ranges Fig. 5 Wavelet coefficients of the three phase currents for single phase A to ground. V. MICROCONTROLLER BASED ARCHITECTURE FOR DIFFERENTIAL PROTECTION A. Microcontroller unit PIC18F4520 8-bit microcontroller of microchip is used in this work. The instruction set of PIC 18 consists of 35 single word instructions with three general formats for the instructions. All instructions are single cycle except for program branches or jumps. Some of the features are mentioned in the following [1]. 1) CPU: CPU speed of 10 MIPS, C compiler, optimized RISC architecture. 2) Memory: 32 k bytes flash program memory, 1,536 byte RAM data memory, 256 byte EEPROM data memory. 3) System: Internal oscillator support-31 khz to 8MHz with 4xPLL Fail-Safe Clock Monitor, Watchdog timer with separate RC oscillator 4) Power Managed Modes: Run, Idle and SLEEP modes, Idle mode for currents of typical values down to 5.8uA, Sleep mode for currents of typical values to 0.1uA 5) Analog Features:10-bit ADC, 13 channels, 100K samples per second, two analog comparators multiplexing. 6) Peripherals: Master synchronous, serial port supports master and slave mode, EUSART module including LIN bus support, four timer modules, five PWM outputs, two Capture / Compare 7) Special Microcontroller Features:100,000 erase/write cycle, enhanced FLASH program memory, 1,000,000 erase/write cycle Data EEPROM memory FLASH/Data EEPROM retention, Power-on Reset (POR), Power-up Timer (PWRT) and Oscillator Start-up Timer (OST), Programmable code protection, Power saving SLEEP mode, In-Circuit Debug (ICD) via two pins. 8) CMOS Technology: Low power, high speed FLASH/EEPROM technology, fully static design, More about PIC microcontroller is found in [13]. According to [13] there are seven important non I/O pins of PIC. First is the MCLR (master clear) pin, which is active low. A switch is connected from that pin to ground to reset PIC when necessary. Biasing are Vdd (pin 11 and 32) and Vss (pin 12, 31). An oscillator is connected across pin 13 and 14 to provide external clock to provide the timing signals necessary for program execution. The remaining 33 pin are configured as I/O pin. The analog signal received from the current transformers and board amplifier is interfaced with the ADC peripheral of the PIC microcontroller where it is converted to a digital outputs. The ADC module on the PIC has four special function registers associated with it: 1) Result High Register (high byte) ADRESH, 2) Result Low Register (low byte) ADRESL to store the output from the converter 3) Control Register 0 (ADCON0) 4) Control Register 1 (ADCON1) The ADCONO register is used to set the conversion time and select the analog input channel. The ADON bit is used to turn on the ADC else the ADC is turned off when the microcontroller is powered up to reduce power consumption. ADCS1 and ADCS0 set the conversion time. The GO_DONE bit is used to check if the conversion is finished. Setting this bit initiates the start of conversion then the bit is cleared when the conversion is complete. CHS2, CHS1 and CHS0 are the channel select bits to determine which input pin is routed to the ADC. ADCON1 is split into two sections. The first section is a single bit, the result format selection bit ADFM which selects if the output is right justified (bit set) or left justified (bit cleared). The advantage is the possibility to use as an 8 bit converter (instead of ten bit) by clearing this bit, and reading just ADRESH and ignoring the two least significant bits in ADRESL. The second section includes the A/D port configuration control bits PCFG3-0. The default of PCFG = 0000 makes the 8 pins RA0-RA3 and RA5 as well as RE0-RE2 used for analog inputs. The internal RC oscillator is used for the conversion clock source. B. LCD Modules: The LCD used has 14 pins. Vcc and Vss provide +5v and ground respectively and Vee is used for controlling LCD contrast. There are two registers inside the LCD. The RS Register Select pin is used for third selection. If RS=0, the instruction command code register is selected. If RS=1 the data register is selected, allowing the user to send data to be displayed on the LCD. The working is dependent upon the interfacing done between the microcontroller and the LCD. 555 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015

The complete circuit diagram is shown in Fig 8. The features of the proposed design are reduced hardware area and circuit complexity compared to the design given in, which applies an external ADC. Fig. 6 LCD Module C. Microcontroller- based Internal Fault Classification The current signals are fed through summation current transformers (CTs). The output signals are rectified and fed to the ADC converter pin of PIC 18F4520 microcontroller. The microcontroller reads the instantaneous magnitude of the received signal via the ADC converter pin. To classify the type of the fault and identify the faulty phase(s). The stored program achieves the following: the magnitude of the spike is compared with the predefined threshold and then one of these action is performed: 1) If the magnitude of only one phase is greater than the threshold, then the fault is classified as single phase to ground fault. 2) If the magnitudes of the three phase currents are greater than the threshold, then the fault is classified as three phase to ground fault. 3) If the magnitudes of only two phase currents are greater than the threshold and the third has a significant value, then the fault is classified as double line to ground. 4) If the magnitudes of two phase currents are greater than the threshold and the third is very small, then the fault is classified as double line fault. The relay senses the type of fault and through the program loaded in the microcontroller the result is displayed on the LCD screen. LCD Microcontroller interfacing is being done via a program loaded in the microcontroller to get the desired output on the display screen. The block diagram of the interface for realization of the proposed PIC 18F4520 microcontroller based differential relay in a three terminal system is shown in Fig. 7. Fig 7. The block diagram for realization of the PIC 18F4520 microcontroller based differential relay Fig 8 The complete circuit diagram. VI. CONCLUSIONS In this work a microcontroller-based fault classification system of a series-compensated transmission line was implemented. The microcontroller used is the powerful Flash PIC18F675. The advantage of PIC 18 is having an internal comparator and ADC on chip which reduces complexity of design. The function of the complete system is controlled by C and assembly language programs developed and run on MPLAB Integrated Development Environment. The microcontroller sends a signal to the interfaced digital relay to trip the circuit and a signal to an LCD to display the type of fault that has occurred. Wavelet MRA is employed for fault location to overcome the difficulties associated with conventional voltage- and current-based measurements for transmission-line fault location algorithms caused by fault inception angle, fault impedance, and fault distance. The simulation results show that the proposed scheme provides a fast, accurate and reliable technique for fault classification and location. REFERENCES [1] www.microchip.com [2] I. Ahmad, F. Khan and Pr.Abdul Mutalib Microprocessor base protection of 132KV line using impedance relay,international Journal of Computer Applications, vol. 34, November 2011. [3] A. Ibrahim, W.R.; Morcos, M.M, Artificial intelligence and advanced mathematical tools for power quality applications: a survey,, IEEE Trans. of Power Delivery, vol. 17, issue 2, pp. 668-673, April 2002. [4] S. H. Horowitz and A. G. Phadke, Power System Relaying Research Studies Press Limited. ISBN: 978-0-470-05712-4, 2008. [5] B. Hamed, Alerting system design using PIC16f877a for power distribution system, Proc.of the 4th International Engineering Conference Towards engineering of 21st century, pp. 1-15, Gaza, 2012. 556 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015

[6] G., Z.; Elangovan, S.; Liew, Microcontroller based overcurrent relay and directional overcurrent relay with ground fault protection, Electric Power Systems Research, Elsevier, vol. 38, July 1996, pp. 11-17(7). [7] Design, Analysis And Fabrication Of A Transmission Fault Detector, A New Approach, M. K. Das1, K. Rout and J. K.,International Journal of Engineering Research & Technology (IJERT), vol. 2, pp. June 2013. [8] A. Rafa, S. Mahmoud, Protection of power transformer using microcontroller-based relay, IEEE Student conference on Research and Development, SCORED, pp. 224-227,2002. [9] E. S. Tag Eldin, Fault Location for a Series Compensated Transmission Line Based on Wavelet Transform and an Adaptive Neuro-fuzzy Inference System, Proc. of Electric Power Quality and Supply Reliability Conference (PQ), Kuressaare, pp. 229 236, 2010 [10] S. G. Mallat, A theory for multiresolution signal decomposition: The wavelet representation, IEEE Trans. of Pattern Analysis and Machine Intelligence (TPAMI., vol. 11, pp. 674 693, July 1989. [11] G.Strang, Wavelet Transforms vs. Fourier Transforms, Bulletin (New Series) of AMS, 28(2), pp. 288-305, 1993. [12] P. Jose, V.R Bindu, Wavelet-Based Transmission Line Fault Analysis,International Journal of Engineering and Innovative Technology (IJEIT) vol. 3, issue 8, 2014. [13] A.N. Pongchaisrikul, A.Kundakorn, Analysis of characteristics of simultaneous faults in electrical power system using wavelet transform in Proc. of IEEE International Conference on Sustainable Energy Technologies, pp.249-252,2008. 557 International Journal of Computer Systems, ISSN-(2394-1065), Vol. 02, Issue 12, December, 2015