Introduction and Architecture of Numerical Relay

Most of us are aware of Electromagnetic Relays and Static Relays but most us may not be well acquainted with Numerical Relay. If I define a Numerical Relay, honestly speaking it will seem to be quite tough but in reality they are very user friendly and easy to implement different types of protection scheme. However I am going to define a Numerical Relay.

Numerical Relays are basically Digital Relays for which manufacturers have developed specified hardware which can be used in conjunction with suitable Software o meet different protection needs. 

A Digital Relay comprises both Hardware and Software. The Hardware part is briefly described below.

CPU: CPU stand for Central Processing Unit which is responsible for the processing of protection algorithms and digital filtering.

Memory:Memory is of two types. One is RAM (Random Access Memory) and ROM (Read Only Memory). RAM serves for the purpose of retaining the input data to the Relay and processing the data during the compilation of algorithm.

ROM is used to store Software needed for the working of Relay. ROM is also needed for storing Event and Disturbance data. Event and Disturbance Recording is a must feature for a digital relay because these data are used for troubleshooting any event. A typical Numerical Relay can store as much as 520 Events and 50 Disturbances. The most attractive feature of such relay is that it works on FIFO (First In, First Out). Suppose if it happens to be the number of disturbances exceeds 50 then the Relay will delete the last Disturbance and will register new disturbance.

Input Module: The analog single from the Power System is stepped down using Current Transformer and Potential Transformer and then fed to the Numerical Relay using low pass filter. Low pass filter is incorporated in the input module to eliminate any noise single induced in the line due to corona or induction effect of nearby high voltage line. The output from the Filter is then fed to Sample and Hold (S/H)circuit.

A Sample and Hold (S/H) circuit is used to keep the rapidly changing instantaneous value constant during the period of conversion for processing.

In addition to the analog input, Numerical Relays are designed to accept digital input too. Separate terminals are provided for the analog and digital inputs.

Multiplexer and Analog to Digital Converter:

The CPU accepts the input in digital form but the input from Current Transformer CT and Potential Transformer PT are analog in nature. Therefore and A/D converter is used to convert the analog signal to digital signal. In case more than one analog quantity is to be converted into digital form, Multiplexer is used for selecting any analog input at a time to convert into digital form.

 

Output Module:

Output module provided in Numerical Relay is digital contacts which are actuated when a trip decision is taken by the CPU. These output digital contacts are a pulse which is generated as a response signal. The timing of pulse can be changed by the user.

Digital Input / Communication Module: Numerical Relays are provided with serial and parallel ports for the interconnection with control and communication system of the substation. Digital output contacts of Numerical Relays are used for wiring with the Auxiliary Relays to extend tripping command to the Circuit Breaker.

Software: Numerical Relays are equipped with software to communicate with external device to program to Relay or one can program by navigating through the Relay Menu.

Hardware for Metering: In principle, the hardware setup discussed above can be used for both measurement and protection function. However, considering the order of difference between current magnitudes in case of fault and load, there can be loss of accuracy during metering applications. Consider a hypothetical case where in maximum load current is 100 A and maximum fault current is 20 times this load current i.e. 2000 A. Let a 12 bit unipolar A/D converter be used for sampling current signal. This implies that resolution of A/D converter is 2000/(2¹²-1)=0.488 A. This resolution may be inadequate for metering purposes.

One solution is to increase resolution i.e. the number of bits in A/D converter. For example, one may use 16 bit A/D converter in place of 12 bit A/D converter.

However, increasing the number of bits of A/D converter also affects the selection of processor. A good design guideline is to choose a processor with double the number of bits of A/D converter. This ensures that truncation and numerical precision problems associated with finite precision arithmetic do not cause significant loss of accuracy. For example, with 16 bit A/D converter, 32 bit processor is the natural choice. Alternatively, a variable gain amplifier can be used along with the A/D converter. At low currents, high gain setting is used and at high currents low gain setting is preferred. However, during the change from one setting to another, loss of information can take place. Therefore, a simple solution would be to keep metering and protection functionality separate.