Double-K Differential Protection Principle of Regional Power Grid Based on Voltage Vector Compensation

In the regional active distribution network where the Distributed Generation (DG) penetration rate is getting higher and higher. When a high-resistance ground fault occurs in a heavy-load line, the traditional current differential protection has very low sensitivity, the protection may refuse to act. And the power differential protection has the voltage dead zone. In addition, the T-type branch will further reduce the sensitivity of the traditional differential protection. In order to solve the problems, this paper proposes a new principle of double-K differential protection with voltage vector compensation. The principle can flexibly adjust the braking zone range by setting two parameters, and then introduce the voltage vector to compensate the operating point, which effectively improves the protection sensitivity in the case of high-resistance ground fault occurs in a heavy-load line, and there is no voltage dead zone. Simulation shows that the principle can greatly improve the reliability of protection.


Introduction
The current differential protection is not affected by the direction of the current, with simple principle and fast action, which can better adapt to the requirements of current power system operation and protection [1]. However, when a high-resistance ground fault occurs on the line with heavy load, there is a large through current, and the protection sensitivity is not high. Many literatures have been improved on the basis of traditional current differential protection to improve the sensitivity of protection [2][3][4][5][6][7]. Literature [8] proposed a calculation power differential protection based on a twoport network, but there is a voltage dead zone problem. Literature [9] proposed a virtual active power differential protection principle based on the fault component. Literature [10] proposed a three-stage active differential threshold protection principle based on the split-phase active power. Literature [11][12][13] proposed a new line protection principle suitable for new energy access by calculating the integrated impedance of the line. However, most of the above principles did not consider the T-type branch, and this situation needs to be considered in the context of an increasing proportion of new energy sources and flexible access to controllable sources.
According to the above analysis, this paper proposes a new principle of double-K differential protection with voltage compensation. The main work of this paper are: 1) Considering the T-type branch, a new principle of differential protection with dual setting values is proposed, which can flexibly adjust the range of the braking zone; 2) The voltage compensation value of the current differential protection is calculated by using the voltage signal and the line parameters, which effectively increases the protection sensitivity, The compensated action equation has the nature of Increasing K will increase the range of the braking zone, and the protection will not be easy to misoperation, but the sensitivity will decrease. Usually K is between 0.3 and 0.9 [14]. Under normal circumstances, the protection principle has very high reliability, but the protection sensitivity is low when the heavy load is grounded with high resistance. In particular, the Distributed Generation (DG) connected to the distribution network through T-type branch have a greater impact on traditional differential protection [15]. The differential current during normal operation or a fault outside the protection zone will increase; and will decrease during a fault within the protection zone. And the degree of effect will increase with the increase of T-type branch capacity, which seriously affects the reliability of protection.

Double-K Differential Protection Principle
In order to ensure the optimal protection reliability and sensitivity, it is necessary to introduce a parameter to change the traditional current differential protection into a double K value current differential protection. The operating conditions of the protection are: . After simplification, we get: The center of the braking zone of the new differential protection criterion is (-K 1 ,0) and the radius is K 2 . The range of the braking zone can be flexibly adjusted by changing the values of K 1 and K 2 .

The Principle Advantage of Double-K Differential Protection and the Selection of K Take
m I as the horizontal axis unit vector of the plane coordinate system to make the action characteristics of the current differential protection, as shown in figure 1.  Figure 1. Operating characteristics of current differential protection.
Define the length of the line segments AB and CD in figure 1 as the horizontal and vertical ranges of the braking zone, respectively.Operating point ρ is within the unit circle. Generally, the differential current when the line fails is relatively large, and ρ falls in the right semicircular area, and the protection operates reliably. However, when a high-resistance ground fault occurs in a heavy-load line, ρ will move to the left, approaching the braking zone, and the protection sensitivity is low.
For the active distribution network connected by DG through T-type branch, when the protection zone is normal, the differential current of the line is mainly reflected in the amplitude. This is because the line power factor is close to 1, and the inverter of DG usually sets the reactive current to 0. It can be considered that the proportion of T-type branch current is the same as that of T-type capacity. Different from the traditional differential protection with a single K, the double-K differential protection principle can realize the separate control of the horizontal and vertical ranges of the braking zone by changing the values of K 1 and K 2 . It can increase the horizontal range of the braking zone without changing the vertical range to improve the reliability of protection.
The selection of double-K needs to meet the requirements of protection reliability and sensitivity. In order to simplify the calculation, refer to the traditional current differential protection, here is a rough value method for K 1 and K 2 : 1) For lines without T-type branch, since the normal operating point is always near (-1,0), this point is set as the center of the braking zone, that is, K 1 takes 1. The selection of the radius of the braking zone needs to be determined according to the line operating conditions and the measurement error of the device. Generally, the reasonable value of K 2 is between 0.4 and 0.9; 2) For lines with T-type branch, assuming that the proportion of T-type branch capacity is  ( 01   ), it can be considered that the operating point of DG is near ( (1 )   , 0) at the rated output. Because of the uncertainty of new energy output, the actual operating point will fluctuate between ( (1 )   ,0) and (-1,0). Therefore, the center of the braking zone is set as the midpoint of the two points, that is, K 1 takes 2 2   , and the value of K 2 is adjusted to be between 2 0.6 2    and 2 0.1 2    accordingly.

Voltage Vector Compensation
In order to solve the problem of low protection sensitivity when the heavy-load line is grounded with high resistance, this paper introduces the split-phase bus voltage signal at the protection installation to compensate the operating point ρ and improve the protection sensitivity. This compensation is suitable for the protection of regional power grids with potential transformers at both ends.   In most cases, m δ has a good compensation effect. But when  is close to 1, point P is very close to point M, and there is almost no compensation effect, as shown in figure 5(a). And when the differential current at the time of failure is large and  is close to 0, the compensation angle 1  is large, it will shift ρ to the braking zone instead, which has the opposite effect, as shown in figure 5(b). Therefore, it is necessary to introduce an another compensation amount for the above two cases, and define the N-side compensation amount as:

DG is Connected to the Grid
The corrected operating point is   II , and replace the line resistance with 2 Z . The calculation method of the compensation amount in the two cases is the same as previous section.

Simulation
Build a 35kV system model in MATLAB/simulink, the system wiring diagram is shown in figure 8.

Simulation Results
Generally, the farther the operating point is from the braking zone in the event of a fault, the higher the sensitivity of the protection. Therefore, the distance from the fault operating point to the braking zone can be used to intuitively reflect the protection sensitivity.
By simulating the fault waveforms when different DG capacity proportion  is connected, the sensitivity comparison between different criteria is calculated and drawn.

Photovoltaic is Connected to the Grid Through the Bus.
The braking coefficient K of the traditional differential protection is 0.45. The two parameters of the double-K differential protection are K 1 =1 , K 2 =0.6 respectively. The sensitivity comparison is shown in figure 9. In this case, when the line is in normal operation and a phase-to-phase fault occurs at point D3, the measured operating points are shown in table 1. and , set the K of the traditional differential protection to 0.45, and the two parameters of the double-K differential are simply set to K 1 =0.95, K 2 =0.6. when 0.2 0.3

 
and , set the K of the traditional differential protection to 0.5, and the two parameters of the double-K differential are set to K 1 =0.9, K 2 =0.6. The sensitivity comparison is shown in figure 10.  In this case, when the line is in normal operation and a phase-to-phase fault occurs at point D3, the measured operating points are shown in table 2.  From table 1: When DG is connected to the line through the bus, the operating point is close to (-1,0) under normal operation and outside the protected area. And the differential current is very small, and the protection generally does not misoperation.

Result Analysis
From table 2: When DG is connected to the line through T-type branch, even no fault, there is still a large differential current, which is positively correlated with the DG capacity, and is mainly reflected in the current amplitude. This is consistent with the analysis in section 3.2. To prevent the protection from misoperation, the braking zone needs to be enlarged accordingly. For traditional differential protection, it can only be achieved by increasing the K. This will increase the horizontal and vertical range of the braking zone simultaneously, reducing the sensitivity of the protection. For double-K differential protection, it can move the center of the braking zone to the right (reduce K 1 ), and the radius remains unchanged (K 2 is unchanged) to achieve the effect of only increasing the horizontal range of the braking zone. It is more in line with the change of the operating point in the case of Ttype branch.
From figure 9 and figure 10 that due to the reasonable adjustment of the braking zone of the double-K differential protection in the case of T-type branch, the sensitivity of the protection is improved compared to the traditional differential protection.
Combining table 1 and table 2, after adding the voltage compensation amount, when the line is operating normally or there is a failure at point D3, m ρ , n ρ and ρ are basically the same, which verifies that the voltage compensation will not increase the risk of protection misoperation. Combined with figure 9 and figure 10, when a fault occurs in the protection zone, the addition of voltage compensation can effectively improve the sensitivity of the protection, and can distinguish the faults at D2 and D3.

Conclusion
This paper considers the different access methods of DG. Aiming at the problem that the traditional differential protection may have too low sensitivity and cause the protection to refuse to operate, through the analysis of the relationship between the electrical quantities of the line. A new principle of double-K differential protection with voltage vector compensation is proposed, and it is verified by simulation on the line with heavy load containing DG.
The simulation results show that the operating point ρ is close to the braking zone when the protection zone fails, which causes the traditional differential protection sensitivity to be too low. The new principle can flexibly control the range of the braking zone by setting two parameters, reasonably reduce the braking zone, and increase the protection sensitivity when a fault occurs. Then the operating point is compensated by a voltage signal. This compensation will not interfere with the normal operation of the system, and can effectively improve the sensitivity of the protection. There is no voltage dead zone in the compensation amount, and it can effectively identify faults inside and outside the protection zone to meet the selectivity of protection.