As a key object of power protection, the power high-voltage test transformer is not only due to its high price, but also because it plays an irreplaceable role in maintaining the stability of the entire power system. As a widely used protection device, whether the longitudinal differential protection can be reasonably applied and correctly operated is crucial to the safe operation of the power grid. The problems that have recently appeared in the operation of high-voltage test transformers in longitudinal differential protection still deserve attention. 

1. Basic principle of longitudinal differential protection for power high-voltage test transformers 

The full name of longitudinal differential protection is longitudinal connection differential protection. The longitudinal differential protection of high-voltage test transformers defends against faults in the bushings and lead wires outside the oil tank. Under normal conditions or when a fault occurs outside the protection zone, the secondary currents of the current transformers on both sides are equal in magnitude and opposite in phase, so the differential current flowing through the relay is zero. However, if a short-circuit fault occurs within the protection zone, the differential current flowing through the relay is no longer zero, so the relay will operate, causing the circuit breaker to trip and providing protection. 

2. Common problems in longitudinal differential protection 

(1) First of all, it is necessary to mention that the most common problem is the problem that occurs during installation; currently, common current transformers are clearly marked with P1, P2 on the shell at the factory; taps S1, S2; meaning that when the primary current of the CT flows from P1 to P2, the direction of the induced current on the secondary side is from S1 to S2. The differential device takes the secondary induced currents of the two CTs at both ends of the protection zone for calculation. At this time, it is necessary to pay attention to the inherent characteristics of the differential protection device itself: whether it is 180-degree wiring or 0-degree wiring. The so-called 180-degree wiring requirement means that the currents entering the protection device from the two CTs at both ends are summed, and when the sum is zero, it does not operate; the 0-degree wiring requirement means that the currents entering the protection device from the two CTs at both ends are subtracted, and when the difference is zero, it does not operate. Installation workers and even some designers often habitually set the differential protection for generators to 0 wiring and for transformers to 180 wiring due to the ambiguity of this principle; this is likely to be inconsistent with the calculation attributes required by the differential protection device itself, thereby causing misoperation of the differential protection. Although the current adaptive wiring differential protection device has solved this problem well, this device is not widely used in power plants and is prone to problems, which requires on-site personnel to strictly verify during the construction process. 

(2) The wiring problem of the current circuit of the differential relay. Now power transformers are mainly divided into two types: dry-type transformers and oil-immersed transformers. In the specification parameters of the transformer, there is an item called the connection group label. That is, the wiring method usually mentioned. Let's temporarily use the conventional Dyn11 to illustrate the wiring problem of the current circuit of the differential relay. According to basic circuit theory, the line voltage of the delta connection is 30 degrees ahead of the phase voltage of the star connection, so for the high-voltage test transformer itself, the current on the high-voltage side will be 30 degrees ahead of the low-voltage side. Then, if the CTs on both sides adopt the same wiring method, the secondary current generated at the high-voltage side CT will also differ in phase by 30 degrees from the secondary current generated by the low-voltage side CT, so during normal operation, it may exceed the protection setting value and cause misoperation. 

For this problem, the current common solution is to change the wiring method of the CT secondary winding. Taking Dyn11 as an example, if the high-voltage side adopts delta connection, then the corresponding CT secondary winding on the high-voltage side adopts star connection; if the low-voltage side adopts star connection, then the corresponding secondary winding on the low-voltage side adopts delta connection; in this way, although the induced line voltage on the high-voltage side of the primary side will be 30 degrees ahead of the induced phase voltage on the low-voltage side, due to the wiring method, the induced current of the star-connected CT will lag behind the induced current of the delta-connected CT by 30 degrees. In this way, the two sets of currents flowing into the differential protection device can just reach a state of phase consistency. 

(3) The problem of magnetizing inrush current. When the transformer is closed or a fault occurs in the external area, a large current may be generated, and then quickly return to the normal no-load current value. This impact current is called magnetizing inrush current. The magnetizing inrush current is not dangerous for the high-voltage test transformer because the duration of this impact current is very short. The main solution is how to identify the inrush current, use some characteristics of the inrush current to form the blocking condition of the differential protection, and find accurate and reliable blocking criteria. 

The principle of the discontinuous angle is to use the fact that the short-circuit current waveform changes continuously, while the magnetizing inrush current waveform has obvious discontinuous angle characteristics as the criterion for identifying the inrush current. This method is based on measuring the discontinuous angle. The measurement of the discontinuous angle must consider the influence of the current transformer transformation on the magnetizing inrush current, especially the influence on the secondary current waveform when the current transformer is saturated. It is also affected by the sampling rate, sampling accuracy, and hardware limitations, so the application effect of this principle in transformer differential protection was not very ideal. However, with the in-depth and detailed research and a large number of experimental work in this area, the algorithm for restoring the discontinuous angle has been proposed, and the improved application effect is relatively ideal. 

Regarding the second harmonic method, at present, the second harmonic is often used to form the blocking condition of the differential protection to prevent misoperation caused by inrush current. There are also several schemes for second harmonic braking. The most commonly used is the three-phase "OR" blocking method. As long as it is judged that the content of the second harmonic in the differential current of one phase meets the condition of inrush current braking, the protection command is blocked and cannot be issued. The application effect of this principle in the field is relatively ideal, and it can basically effectively distinguish the real fault of the high-voltage test transformer from the inrush current during no-load closing or voltage recovery after external fault removal. 

The third harmonic component in the inrush current is also relatively large, second only to the second harmonic, but the third harmonic cannot be used as the characteristic quantity of the inrush current to form the braking or blocking part of the differential protection. If the DC component is used to form the blocking condition of the differential protection, when a short circuit occurs inside the high-voltage test transformer, the action speed of the protection will inevitably be delayed, and often one phase in the three-phase inrush current is a periodic current, that is, it does not contain a DC component. At this time, the operating current of the differential protection must be increased to avoid this periodic inrush current, which also reduces the sensitivity of the protection. 

(4) The problem of differential protection setting values. The current differential protection device setting values include the differential speed section input/output setting value; differential speed-breaking current setting value; ratio differential input/output setting value; ratio differential threshold current setting value; inflection point 1/2 current setting value; broken line 1/2 slope coefficient setting value; CT disconnection blocking input/output setting value; differential balance coefficient setting value; start time setting value; differential delay time setting value. The following mainly explains the differential balance coefficient, differential speed-breaking setting value, and ratio differential protection setting value: 

① Differential balance coefficient: used to compensate for the current balance of the differential circuit, converted according to the selected reference value. 

Kphl=In1/In2 

Where In is the rated current, calculated as follows: 

In=Pn /√3Un*Kl 

Where: Pn - rated capacity. 

Un - rated voltage of each side. 

Kl - current transformer ratio. 

② Differential speed-breaking setting value: the setting principle is to avoid the maximum unbalanced current during external faults, that is: Isd=Kr*In1 

Where: In1 is the rated current of the high-voltage side; Kr is the magnetizing inrush current multiple relative to the rated current, which can be set according to the system impedance and CT characteristics, generally taking 6~10Ie. 

③ Ratio differential protection setting value: including differential current threshold setting value Icd, * inflection point setting value Ir1, second inflection point setting value Ir2, ratio braking * section broken line slope K1, second section broken line slope K2, harmonic braking coefficient Kxb. The Icd setting value should be able to avoid the maximum unbalanced current of the differential circuit at rated load, that is, Icd=Kk*（Ktx*Fi+）*Ie 

In--rated current Kk - reliability coefficient, taking 1.3~1.5 

Ktx - current transformer same-type coefficient, taking 1.0 

Fi - maximum relative error of current transformer, taking 0.1. In practice, Icd is generally taken as 0.25~0.5Ie 

For longitudinal differential protection of transformers of different models and capacities, each is equipped with its own setting value manual at the factory. Before the debugging work starts after installation, it is necessary to pay attention to checking the parameters of the differential protection device according to the setting value manual. Choosing the appropriate protection setting value parameters is the key to ensuring the reliable operation of the transformer protection device.