69 CHAPTER-4 TRANSMISSION LINE DIFFERENTIAL PROTECTION USING WIRELESS TECHNOLOGY 4.1 INTRODUCTION The probability of fault occurrence on the overhead lines is much more due to their greater lengths and exposure to atmospheric conditions. There are several methods available for the protection of transmission and distribution lines [69]. The first group of non unit type of protection includes time graded over current protection; current graded over current protection and distance protection. Such non unit type protections don t have pilots. The other group of protection of line includes pilot wire differential protection, carrier current protection based on phase comparison methods, etc. They fall in to the category of unit protection. In these schemes some electrical quantities at the two ends of the transmission line are compared and hence they require some sort of interconnection channel over which information can be transmitted, from one end to another. Such an interconnection channel is called as a pilot. In pilot wire differential protection a differential relay is used which responds to vector difference between two or more similar electrical quantities. This vector difference is achieved by suitable connection of current or voltage transformer secondary. Most of the differential relays in which vector difference between the current entering the winding and current leaving the winding is used for the relay operation. This differential protection is directionally secure and is simple to implement and set. Current differential protection using pilot wire is mainly applied for short transmission lines protection. Vector differences between the currents at both
70 ends of transmission line are used for relay operation [70]. Another type of differential protection uses balanced voltage principle across two ends of transmission line [71]. Pilot wire schemes are usually economical only for short distances with the restriction of voltage drop due to impedance of the wire, inaccurate system impedance and loss of relay function due to line disconnection. These in turn cause energy loss. The reliability of communication obviously impacts the reliability of the protection system. Pilot protection schemes may send analog values between the relays at each end of the line or they may use simple ON /OFF, permissive or blocking signals between relays. The advancement of this protection relaying allows high speed communication among protective relays called peer to peer communication [72]. There are many methods to send these signals. The most common methods currently in use are twisted pair cable, coaxial cable, fibre optic cable, power line communication and wireless communication. Wireless communication is the challenging media for the upcoming deregulated power system and smart grid applications [41]. A fibre optic pair available for exclusive use by the relays provides optimal performance for digital communications. Dedicated fibre gives a fast and error-free point-to-point connection. The main drawback is that a fibre cut will cause a channel interruption for long period of time. Many utilities lack expertise and equipment for replacing and splicing a damaged fibre cable. Of course, the installation and material costs for a dedicated fibre compared to conventional communication channels limits its availability for relaying.
71 To overcome the drawbacks of all conventional protection communication schemes and to maintain high reliability and reduction of energy loss, it is essential to use accurate communication medium like wireless communication protocol between the relays [73-75]. Now a day, there are many advanced robust communication techniques that can be used to improve protection, control, speed outage restoration, operation analysis, maintenance and planning. A laboratory investigation of transmission line protection using wireless RF communication is proposed in this chapter. It collects correct data from both ends of transmission line through intelligent electronic devices to communicate with each other and for relay decision. This method provides high degree of accuracy, reliability, energy efficient and cost effective. 4.2 LABORATORY MODEL Single line diagram of the laboratory model is shown in Fig.4.1. It consists of two relays. The two relays make final decision based on signals sent in both directions through a wireless communication network [75]. Fig.4.1. Single line diagram of differential protection The structure of laboratory model for differential protection of short transmission line is shown in Fig.4.2. It mainly consists of the following six different components. Power system model Sensing transformers
72 Signal conditioning circuits Data conversion circuits Communication network Software required Fig.4.2. Laboratory model of the protection system 4.2.1. Power system model A single machine infinite bus system is physically modeled in the power system laboratory. A three phase, 50Hz, 11MVA, 110kV (720VA, 48V are the scale down experimental values) is considered as the generating station. The station is connected to infinite bus having a load of 20Ω/phase through a transmission line. The transmission line is considered as short line of 12km, 110kV having a line resistance of 0.25Ω/km. Three phase relay circuits are used at two ends of the transmission line. Fault is created by using DPST switch. Any type of fault can be applied to the power system model. Fault location and fault resistance can also be changed. However, the principles discussed in this thesis can be applied to any circuit at any voltage level, assuming the protection requirements like speed, security and dependability are met.
73 4.2.2. Sensing transformers Sensing transformers are either voltage or current sensing. These are essential to create ratio metric reduction and isolation purpose. This should meet exact standards like the secondary of PT should be 110v and the CT secondary must be 5A irrespective of their primary levels. These devices must be kept in the sending and receiving ends for data acquiring purposes. 4.2.3. Signal conditioning circuits The main objectives of these devices are to rectify, filter, setting up the calibration limits, protecting the high voltage hazards and protecting the inputs and outputs discussed clearly in section 3.3.2. 4.2.4. Data conversion circuits Generally A/D converters will be separately used with multichannel. But in the proposed design, a state of art embedded system technology is used to reduce lot of electronic hardware. These devices consist of packed hardware inside; any device can be brought down to the front end and can be used. Output from signal conditioning circuits is connected to this circuit for A/D application. Simultaneously all analog data are fed and digitized data are sent to computer as RS-232 signals. The digitized data will be decoded for real time values later. The expected speed of this device is 9600 baud rate. Middle end embedded microcontroller like PIC16F877A is proposed, which consists of 8 channels 10bit ADC with lots of additional features discussed in section 3.3.5. These devices require very minimum supporting hardware like clock and reset circuits externally. 4.2.5. Communication network Over the years many communication techniques have been used in power system applications starts from simple voice to trip switch gears. The trend of
74 SCADA introduction to power system control draws attention of several kinds of communication technologies to acquire, control and for decision making. Most of the devices are onsite equipments used to acquire the data at one end supports to communication equipments and passed through leased lines, often the quality of communication is kept under leased line owners. Decoding the necessary data is much complex because many servers were introduced in between and causes to slow down data transfer rates. For short distance power system applications optical fibre cables (OFC) were used. It has some advantages but very expensive and non transferable with huge installation cost but very rugged. Installation of OFC is limited due to several constraints, but obsolutely noise free at any level. Online wireless communication is very reliable and dedicated, doesn t rely on leased lines. There are many communication techniques like Unidirectional RF, carrier communication, ZigBee, GSM, Internet, CDMA, Bluetooth and Bi-directional RF. Many of the above network communication systems are leased lines and dependency goes to service providers, few of them are suitable only for short distance communication up to 400 500 meters. The another challenging method to provide digital communications for pilot protection, that is reliable and affordable is Digital radio. This is an inexpensive method to provide digital communications for pilot protection at the distribution level [76]. But RF could be a reliable solution, can travel over a long distance by adjusting its parameters like carrier frequency limitations. Unidirectional wireless technology is well suitable for tripping, but this research used linear data transfer of current by using RF transmitter and RF receiver. In RF transmitter, current is converted into voltage, voltage to frequency and super imposed into carrier and transmitted in air. This frequency is received by the receiving circuit and demodulated to get the modulated signal (this is equal to the current in the
75 sending end).the received modulated output will be 5kHz maximum fed to a transister coupler and connected to a Schmitt trigger to get perfect square wave and applied to F-to-V converter, will be received voltage on-behalf of current, compared with the sending current.the differential current will be displayed and appropriate control action will be taken for tripping using Embedded microcontroller. 4.2.5.1 RF Communication RF communication works by creating electromagnetic waves at the source and being able to pick up those electromagnetic waves at a particular destination. These electromagnetic waves travel through the air at near the speed of light. The wavelength of an electromagnetic signal is inversely proportional to the frequency; the higher the frequency, the shorter the wavelength. Working of RF communication is explained clearly in the following section. 4.2.5.1.1 Voltage to Frequency converter The LM331 voltage-to-frequency converters (V-to-F) are ideally suited for use in simple low-cost circuits for analog-to-digital conversion, precision frequency-tovoltage conversion, long-term integration, linear frequency modulation or demodulation and many other functions. The output when used as a voltage-tofrequency converter is a pulse train at a frequency precisely proportional to the applied input voltage. Thus, it provides all the inherent advantages of the voltage-tofrequency conversion techniques, and is easy to apply in all standard voltage-tofrequency converter applications. Further, the LM331 attain a new high level of accuracy versus temperature which could only be attained with expensive voltage-tofrequency modules. Additionally the LM331 is ideally suited for use in digital systems at low power supply voltages and can provide low-cost analog-to-digital conversion in
76 microprocessor-controlled systems. The frequency from a battery powered voltage-tofrequency converter can be easily channeled through a simple photo isolator to provide isolation against high common mode levels. The LM331 utilizes a new temperature-compensated band-gap reference circuit, to provide excellent accuracy over the full operating temperature range, at power supplies as low as 4.0V. Features Guaranteed linearity 0.01% max Improved performance in existing voltage-to-frequency conversion applications Split or single supply operation Operates on single 5V supply Pulse output compatible with all logic forms Excellent temperature stability, ±50 ppm / C max Low power dissipation, 15 MW typical at 5V Wide dynamic range, 100 db min at 10 khz full scale frequency Wide range of full scale frequency, 1 Hz to 100 khz Low cost Fig. 4.3 Voltage to Frequency Converter Circuit
77 In the circuit of Fig.4.3, integration is performed by using a conventional operational amplifier and feedback capacitor CF. When the integrator s output crosses the nominal threshold level at pin 6 of the LM331, the timing cycle is initiated. The average current fed into the Op - Amp s summing point (pin 2) is i x (1.1 R t C t ) x f which is perfectly balanced with -V IN /R IN. In this circuit, the voltage offset of the LM331 input comparator does not affect the offset or accuracy of the V-to-F converter as it does in the stand-alone V-to-F converter; nor does the LM331 bias current or offset current. Instead, the offset voltage and offset current of the operational amplifier are the only limits on how small the signal can be accurately converted. Since op-amps with voltage offset well below 1 mv and offset currents well below 2nA are available at low cost, this circuit is recommended for best accuracy for small signals. This circuit also responds immediately to any change of input signal (which a stand-alone circuit does not) so that the output frequency will be an accurate representation of V IN, as quickly as 2 output pulses spacing can be measured. In the precision mode, excellent linearity is obtained because the current source (pin 1) is always at ground potential and that voltage does not vary with V IN or F OUT. (In the stand-alone V-to-F converter, a major cause of non-linearity is the output impedance at pin 1 which causes i to change as a function of V IN ). RF Transmitter The block description of the transmitter module and pin diagram is shown in Fig.4.4 and Fig.4.5 respectively. This module is implemented in order to transmit the frequency output of the LM331 to the remote computer which actually make the system wireless.
78 Fig. 4.4 Transmitter Module Fig. 4.5 Pin Diagram of Transmitter Module Receiver / Data acquisition section Receiver / Data Acquisition Section consist of three sub-sections. RF Receiver. Frequency to Voltage converter. Microcontroller interface Data Acquisition (Computer Program).
79 RF Receiver Fig.4.6 Receiver Module Fig.4.7 Pin diagram of Receiver Module Pin diagram which is shown in Fig.4.7 is used as a receiver module in order to implement wireless communication. The block diagram shown in Fig.4.6 will receive the data from the transmitter through the antenna and pass it to a Frequency to Voltage converter which is then converted to an analog data and that data will be passed to the microcontroller.
80 4.2.5.1.2 Frequency to Voltage converter Fig.4.8 Frequency to Voltage Converter Circuit In these applications, a pulse input at FIN is differentiated by a C-R network and the negative-going edge at pin 6 causes the input comparator to trigger the timer circuit. Just as with a V-to-F converter, the average current flowing out of pin 1 is I average = i x (1.1 Rt Ct) x f. In the simple circuit of Fig.4.8, this current is filtered in the network RL = 100 kw and 1 µf. The ripple will be less than 10 mv peak, but the response will be slow, with a 0.1 second time constant, and settling of 0.7 second to 0.1% accuracy. In the precision circuit, an operational amplifier provides a buffered output and also acts as a 2-pole filter. The ripple will be less than 5 mv peak for all frequencies above 1 khz, and the response time will be much quicker than in Fig.4.8. However, for input frequencies below 200 Hz, this circuit will have worse ripple than Fig.4.8. The engineering of the filter time-constants to get adequate response and small enough ripple simply requires a study of the compromises to be made.
81 Inherently, V-to-F converter response can be fast, but (Frequency to Voltage Converter) F-to-V response cannot. Here the frequency which is obtained from the receiver is given as input where the frequency is again converted into analog voltage value. Then the analog value is passed on to the microcontroller for analog to digital conversion. The voltage obtained from F-to-V converter will be fed to EMC to process and to convert the analog value into digital and further to serial communication and applied to computer. On the computer well designed software is employed to compare the current delivered and current value received from wireless. When the values are within the predefined limits tripping will not be executed, the difference between currents is greater than the limit, it is experiencing overload so tripping will be executed to save power. 4.2.6. Software required Front end software used is Microsoft visual studios VB6.0 enterprise version to visualize effective on screen management. The variables are processed to obtain real values. The calculations could be simple multiplication to complex algorithms. The digitized data will be decoded to get real values. The real values will be displayed on the computer screen for further processing. The real values will be put into database also. This can be viewed by enabling the text box. The hardware of transmitter and receiver circuits is shown in Fig.4.9 and Fig.4.10 respectively. Transmission line is connected between these circuits shown in Fig.4.11.
82 Fig.4.9.Hardware of the transmitter circuit Fig. 4.10. Hardware of the receiver circuit
83 Fig.4.11. Three phase transmission line 4.3. PROTECTION SCHEME Protection is based on measuring the current signals at both ends of the transmission line and transmitted via communication network. The protection scheme applied on the power system network is shown in Fig.4.12. Proposed technique can be explained through analysis of three key components. Synchronization element Differential element Decision element Fig.4.12 Power system network
84 4.3.1. Synchronization element In order to evaluate the differential protection based on current signals measured at both ends of the transmission line, the current samples have to be taken at the same time intervals at both ends. This requires relay operation to be synchronized, any time difference between the relay signals will translate into differential current that may cause panic operation of the relay. There are many methods used for synchronizing the data measured at both ends of the transmission line [77-78]. In most of the techniques there is a need for additional equipment or connecting equipment to the satellites which increases the cost of manufacturing the relays. In the proposed technique wireless communication network is used for the synchronization of the relays. 4.3.2. Differential element The differential element calculates the current deviation signals at fastest rate at a speed of 8/9600 sec using the following equation (4.1). p R, Y, B R, Y, B R, Y, B 1 2 q= 1 i ( p) = [ i ( q) i ( q)] 4.1 Where, i is the current deviation signal for phases R, Y, B; p = Number of samples; q = Index; i 1 = Sending end current; i 2 = Receiving end current;
85 4.3.3. Decision element The new current signals will be compared with the corresponding current signals of the pre-specified set value. The comparison can be mathematically explained as below. For Normal operation and external faults: R, Y, B R, Y, B i ( New) < i (Pr e) 4. 2 For Internal shunt faults: R, Y, B R, Y, B i ( New) > i (Pr e) 4. 3 For Internal series faults: R, Y, B i ( New) = 0 4.4 ` Where the subscript New refers to the present value and Pre refers to the pre-specified set point. From the equation 4.2, it is observed that the current which flows through the line may be equal both under normal condition and in the external fault conditions. From the equation 4.3 it is observed that, the differential protection scheme will receive the difference of sending and receiving end currents and difference of currents is greater than the set point value during the internal shunt faults. The difference of currents is equal to zero during the internal series faults as shown in equation 4.4. Flow chart in Fig.4.13 shows the process of protection scheme using wireless communication for the transmission line.
86 Fig.4.13 Flowchart for the relays at both ends of transmission line. 4.4. RESULTS & DISCUSSIONS Based on the flow chart given in Fig.4.13, program is developed using VB6.0 software and different faults are analyzed as given below. Tripping of power line is unavoidable to save grid collapse. There by save huge amount of power by considering shortest fault duration time (less than 2 cycles). Long duration faults may consume huge power in terms of MVA in transmission lines. That leads to failure of cables, fuses, switch gears and relevant transmission lines. These faults are not commonly notified using existing system. The existing system overload setting is for common distribution. If only one user works and other is absent, the concept of overload fails, fault duration will be panic in this situation.
87 Tripping will not be executed till the load reaches beyond the set point and will not be possible in practical case. But the proposed system senses perfectly and trips instantly using wireless communication and saves power. Fig.4.14(a) shows at normal condition (without fault). At this condition all phase currents at sending end I R1, I Y1, I B1 and at receiving end I R2, I Y2, I B2 are equal. There fore the difference of currents are with in the specified limit. Fig. 4.14(a) Normal condition The sending and receiving end currents are measured during single line to ground fault through fault resistance 10Ω, 50% of length from sending end. Differential current is calculated by using measured signals. This differential current is greater than the specified limit in phase R, while in phases Y and B it is less than the threshold value. This means that the fault is single line to ground and internal in phase R and is shown in Fig.4.14(b).
88 Fig.4.14 (b) Single line to ground The relays at both sending and receiving ends are using again the wireless communication network to exchange the decision. After exchanging the relays data at both ends of the line, it will be concluded that the line is faulted and the trip signals are applied to circuit breakers to disconnect. Fig.4.14(c) shows the currents and voltages immediately after tripping. Fig.4.14 (c) R-G fault after tripping The current signals at the sending and receiving ends on a double line to ground fault between the phases R and Y through fault resistance of 10Ω, at the mid point of the line are measured. The measured current signals are transferred through
89 wireless communication network. Fig.4.14(d) shows the deviation signals in phases R,Y&B. It is observed that the difference of current between sending and receiving ends is greater than 1000mA. Therefore trip signals are applied to the circuit breakers. Fig.4.14 (d) Double line to ground fault The current signals at the sending and receiving ends are measured during three phase to ground fault at 75% of length from the sending end. The measured signals are transferred through wireless communication network. Fig.4.14(e) shows the deviation signals in phases R,Y& B and the difference of currents between sending and receiving ends are greater than the specified value, the relays giving the trip signals to circuit breakers. Fig. 4.14 (e) Three phase fault
90 The measured signals at both ends of the line are also exchanged through the wireless communication network. The deviation signals for all the three phases are shown in Fig.4.14 (f). The deviation signals for phases R,Y & B are less than the specified threshold limit. This shows that fault is external. The two relays exchange the decision to assure that the fault is external and no tripping signals are communicated. Fig.4.14 (f) External fault Fig.4.15(a) shows data base results at normal (without fault) condition. This data shows the voltages, currents for the phases R, Y & B of the sending end, feedback current from the receiving end and difference between the sending and receiving end currents for each millisecond. These results will be useful for future verification purpose. Fig.4.15 (b) gives the data base results of voltages, currents for the phases R,Y & B of the sending end, feed back current from the receiving end and difference between the sending and receiving ends currents for each millisecond for the single line to ground fault at R phase.
91 Fig.4.15 (a) Data base results during normal condition Fig 4.15(b) Data base results during faulted condition The relays at both sending and receiving ends are using again the wireless communication network to exchange the decision. After exchanging the decision relays at both ends of the line are decided that there is an open circuit fault in R-phase and they give trip signals to circuit breakers. Fig.4.16(a) shows that R-phase current is zero, voltage is high and differential current is zero.
92 Fig.4.16 (a) R-phase open condition The current signals at the sending and receiving ends are measured during two line open circuit fault. The measured signals are transferred through wireless communication network. Fig. 4.16 (b) shows the deviation signals in phases R,Y& B and the difference of currents between sending and receiving ends are zero value, the fault is identified as a R-Y open circuit fault and giving trip signals to circuit breakers. Fig.4.16 (b) R-Y phases open condition
93 Fig.4.16(c) shows the deviation signals in phases R,Y& B and the difference of currents between sending and receiving ends are zero value, the fault is identified as a Y-B open circuit fault and giving trip signals to circuit breakers. Fig.4.16 (c) Y-B phases open condition. 4.5. CONCLUSIONS The laboratory model for differential protection of short transmission line using wireless RF communication is demonstrated. Depending on protection algorithm applied in relays at each end of the transmission line, current signals are measured and transferred through the wireless communication network. The loss due to various factors resulting differential currents and are exactly acquired using RF communication to the sending end. The differential current is transmitted to the sending end and comparison is made for tripping. Implementation of this will enhance the life time of conductors and can support for reduction of losses with economic viability. The wireless communication networks offers advantages over the conventional pilot wire systems.