An Automatic Blocking Method for Transformer Differential Protection Based on the Quartering Method

Abstract

The present invention belongs to the field of relay protection of power systems, and discloses an automatic locking method for transformer differential protection based on the quartering method. After the differential protection is triggered, the transformer differential current signal is sampled; the direction coefficient B is calculated according to the sampled data L , and compared with the threshold B set ; if B L >B set , it is determined as single-sided inrush current, otherwise the next determination is made; for the differential current data that does not meet the foregoing criterion, the interference data is filtered out, the remaining data after filtering is divided into four equal parts, the variable value is obtained by solving, and the value of the action coefficient QSI is calculated and compared with the threshold Q set ; if QSI>Q set , it is determined as inrush current and the differential protection is locked; otherwise, it is determined as fault current and the differential protection is opened. The present invention accurately identifies single-sided inrush current according to the direction coefficient and the action coefficient based on the quartering method, and solves the problem of misoperation of differential protection under multiple interferences compared with the traditional identification method.

Description

Technical Field

[0001] The present invention belongs to the technical field of relay protection in power systems, and particularly relates to a method for automatically locking a transformer differential protection based on the quadrant method. Background Art

[0002] As an essential electrical device in the power system, a power transformer undertakes the important task of power transformation and distribution. In the field of relay protection technology, pilot differential protection is used as the main protection of the transformer to reduce the damage caused by internal faults and ensure the reliability of the power system and the continuity of power transmission. Different from other types of protection, the correct action rate of transformer protection is relatively low, and the essence is that the differential protection cannot be correctly locked. Long-term production practice shows that transformer inrush current is one of the reasons for the incorrect action or locking of pilot differential protection. Similarly, it is worthy of great attention that affected by different interference conditions, the waveform distortion of differential current is extremely likely to cause the refusal and incorrect action of differential protection. It can be seen that the accurate identification of inrush current and fault current under multiple interferences is of great significance for improving the correct action rate of the main protection of the transformer.

[0003] Inrush current is generated when the transformer is put into operation without load or when the voltage recovers after an external fault is removed. It is an abnormal operating state rather than a fault, and the transformer protection should be locked when inrush current occurs. When the CT is saturated, the waveform of the secondary differential current obtained through transformation is distorted, or a fault arc is generated within the protected area of the transformer, which seriously affects the correctness of differential protection. Therefore, accurately identifying inrush current and fault current under multiple interferences is an urgent problem to be solved.

[0004] In engineering, a relatively mature method is to use the second harmonic criterion or the discontinuous angle criterion to identify inrush current. However, when facing the interference of CT saturation and fault arc, their accuracy drops severely, and they are no longer suitable for today's complex power system and it is difficult to meet the high requirements for the correct action rate of protection. Regarding the identification problems of inrush current and fault current, researchers at home and abroad have proposed many new methods in recent years. For example, the methods based on the kurtosis and skewness coefficients in statistics and the addition of artificial intelligence algorithms have continuously promoted the development of inrush current identification. However, when facing the influence of multiple interferences, there are still various problems to be solved. At present, it is still necessary to continuously innovate technologies to effectively identify inrush current and fault current under multiple interferences and improve the correct action rate of transformer protection. Summary of the Invention

[0005] The object of the present invention is to overcome the above deficiencies and provide a method for automatically locking a transformer differential protection based on the quadrant method.

[0006] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0007] An automatic locking method for transformer differential protection based on the quartering method, comprising the following steps:

[0008] Step 1: Sample the differential current signal in the differential circuit of the power transformer;

[0009] Step 2: If the differential current signal satisfies the differential protection criterion, go to Step 3; otherwise, return to Step 1;

[0010] Step 3: Calculate the direction coefficient B through Equation (1) L :

[0011]

[0012] In the formula, N is the number of samples in one power frequency cycle; {i1, i2,......, i N} is the data sequence composed of the differential current signal data of a newly collected cycle, where r = 1, 2,......, N;

[0013] Step 4: Compare B L with the threshold B set ; if B L is greater than the threshold B set , directly determine it as single-sided inrush current and lock the transformer differential protection; otherwise, go to Step 5;

[0014] Step 5: Find the data with the largest absolute value in the data sequence {i1, i2,......, i N}, and assign the absolute value of this data to M after taking the absolute value; compare the absolute values of the data in {i1, i2,......, i N} with 0.15M in turn, and store the data with absolute values greater than 0.15M in the data sequence {e1, e2,......, e L} in turn, and store the sequence numbers of these data in {i1, i2,......, i N} in the data sequence {t1, t2,......, t L} in turn; where L is the total number of data in {i1, i2,......, i N} with absolute values greater than 0.15M;

[0015] Step 6: Let the variable h take values of 1, 2, 3,......, L in turn, retrieve all the data in the data sequence {t1, t2,......, t L}, if h satisfies Equation (2), then {e h-0.05N , e h-0.05N+1 ,......, e h} is removed from the data sequence {e1, e2,......, eL} is deleted from it, and at the same time, {t h-0.05N , t h-0.05N+1 ,......, t h} is deleted from the data sequence {t1, t2,......, t L};

[0016]

[0017] If h satisfies equation (3), then {e h+1 , e h+2 ,......, e h+0.05N} is deleted from the data sequence {e1, e2,......, e L}, and at the same time, {t h+1 , t h+2 ,......, t h+0.05N} is deleted from the data sequence {t1, t2,......, t L};

[0018]

[0019] Step 7: Assign the remaining data of the data sequence {e1, e2,......, e L} after step 6 to the data sequence {x1, x2,......, x s}; Assign the remaining data of the data sequence {t1, t2,......, t L} after step 6 to the data sequence {d1, d2,......, d s}; where s is the total number of remaining data after step 6; Determine the quarter-step length q, and divide the data sequence {x1, x2,......, x s} into four parts: {x1, x2, …, x q}, {x q+1 , x q+2 , …, x 2q}, {x 2q+1 , x 2q+2 , …, x 3q}, {x 3q+1 , x 3q+2 , …, x s}; At the same time, divide the data sequence {d1, d2,......, d s} into four parts: {d1, d2, …, d q}, {d q+1 , d q+2 , …, d 2q}, {d 2q+1 , d 2q+2 , …, d3q [[ID=1}},{d 3q+1 ,d 3q+2 ,…,d s};

[0020] Step 8: Obtain the values of variables α1, β1, and θ1 when Equation (4) reaches the minimum value; and obtain the values of variables α2, β2, and θ2 when Equation (5) reaches the minimum value;

[0021]

[0022]

[0023] In Equation (4) and Equation (5), f is the objective function value of the first part and the fourth part; g is the objective function value of the second part and the third part; α1, β1, θ1, α2, β2, and θ2 are variables to be solved;

[0024] Step 9: Calculate the action coefficient QSI through Equation (6):

[0025]

[0026] Step 10: Compare QSI with the threshold Q set ; if QSI is greater than the threshold Q set , it is determined that the differential current signal is symmetric inrush current or saturated inrush current, and the transformer differential protection is blocked; otherwise, it is determined that the differential current signal is one of traditional fault current, saturated fault current, and arc fault current, the transformer differential protection is opened, and the transformer differential protection is allowed to operate.

[0027] Preferably, the four-step length q in Step 7 is obtained through Equation (7):

[0028]

[0029] In the formula, is the floor function of towards negative infinity.

[0030] Preferably, the value of the threshold B set in Step 4 is 0.8.

[0031] Preferably, the value range of the threshold Q set in Step 10 is 0.25 to 0.35.

[0032] All of the above steps include two determination processes. Steps 1-4 are criterion 1; Steps 5-10 are criterion 2; Step 4 is to screen out the data that needs to be filtered; the process from Step 5 to Step 6 is the process of filtering out interference data; Steps 7 and 8 are the process of designing a four-equal-part method to obtain variable values.

[0033] The principle of the present invention is as follows:

[0034] The inventive concept of the present invention is as follows. Starting from the difference in differential current waveforms, certain technical means are used to solve the problem of misoperation of differential protection under multiple interferences (CT saturation interference, arc interference).

[0035] Analyze the reasons for the misoperation of differential protection: In the past, relay protection devices mostly adopted the second-harmonic braking principle and the interrupted angle principle when identifying magnetizing inrush current and fault current. However, in the face of the increasingly complex power system environment, it is necessary not only to identify conventional magnetizing inrush current and internal fault current, but also to have the ability to solve the problem of identifying fault current under multiple interferences. Although the above two identification principles can meet the former requirements, they cannot be perfect and comprehensively ensure the correct non-misoperation of differential protection when facing multiple interferences. Specifically, due to the sharp increase in the harmonic content of the saturated fault current transmitted through the secondary side when the CT is saturated, the second-harmonic principle misjudges the fault current as magnetizing inrush current; the interrupted angle principle will be challenged by both arc fault current and saturated magnetizing inrush current. The arc fault current is misjudged as magnetizing inrush current due to the waveform distortion resulting in an interrupted angle, and the saturated magnetizing inrush current is misjudged as fault current due to the waveform distortion causing the disappearance of the interrupted angle.

[0036] The present invention comprehensively considers the influence of interference on the differential current waveform and designs a two-step discrimination scheme to accurately identify abnormal states and faults such as single-sided magnetizing inrush current, symmetrical magnetizing inrush current, internal short-circuit fault current, saturated magnetizing inrush current, saturated fault current, and arc fault current.

[0037] In the first step, sample the differential current for multiple cycles. The number of samples in one power frequency cycle is N. Utilize the unidirectionality unique to magnetizing inrush current only to calculate the direction coefficient B through formula (1) L and compare it with the threshold B set If the differential current data greater than the threshold is determined to be single-sided magnetizing inrush current distributed on one side of the time axis; otherwise, proceed to the second step for further identification of the differential current.

[0038] Step 2: Considering the sine characteristic of the fault current waveform, a design algorithm is selected to delete the interference data to improve the similarity between the differential current and the ordinary sine signal. In the present invention, the interference data is divided into two parts and deleted one by one. The first part is the discontinuous angle of the inrush current and the discontinuous angle generated by the waveform distortion of the CT saturation fault current and the arc fault current. Since the value of the discontinuous angle is very small and generally approximated to zero or a small part is equal to zero, the discontinuous angle is filtered by the magnitude of the absolute value of the differential current within a period. Specifically, find the maximum value after taking the absolute value, and compare all the data with the absolute value within the period with 0.15 times the maximum value, and delete all the data smaller than it; The second part is a small number of distorted data signals of the CT saturation fault current and the arc fault current around the discontinuous angle. These few distorted signals may appear on the left or right side of the discontinuous angle. In the present invention, the deletion position is determined by comparing the slope magnitudes of N / 20 data on the left and right sides of the discontinuous angle, and N / 20 data is deleted. In particular, considering that the distorted waveform of the saturated inrush current is similar to the fault current but it belongs to an abnormal operating state rather than a fault, the present invention cleverly uses the method of quadruple equal division and then recombination of data to fully display the part different from the sine signal, making its non-sine characteristics more prominent. On the contrary, for the differential current signals in other fault states, after deleting the interference data, the differential current waveform is identical to the sine waveform, and the quadruple equal division and recombination have little impact on it, and there is a high recognition rate from the inrush current. It should be noted that the values of the variables α1, β1, θ1 and the variables α2, β2, θ2 are obtained when the formulas (4) and (5) are satisfied, where the formula E=(R T R) -1 R T Y can be selected to solve the variable values, where E is the variable to be solved, R is the matrix composed of the coefficients of the variables α1, β1, θ1 or the variables α2, β2, θ2 in the formulas (4) and (5), and Y is the column vector composed of the differential current values corresponding to the first part and the fourth part or the column vector composed of the differential current values corresponding to the second part and the third part. The variable values are obtained through calculation, but not limited to this calculation method. Finally, the excellent effect of the above algorithm is demonstrated by calculating the QSI value, and the calculation result is compared with the threshold Q set . If it is greater than the threshold Q set , it is determined as inrush current, otherwise it is determined as fault current.

[0039] The beneficial effects of the present invention are: 1) The present invention accurately identifies the single-sided inrush current according to the direction coefficient and the action coefficient based on the quadrant method. Compared with the traditional identification method, the present invention solves the problem of misoperation of differential protection under multiple interferences; 2) Compared with the existing artificial intelligence and other algorithms, the present invention has a small calculation amount, high accuracy and is easy to implement. Brief Description of the Drawings

[0040] Figure 1 is the flow chart of the present invention;

[0041] Figure 2 is the simulation model of the embodiment of the present invention;

[0042] Figure 3 is the single - sided inrush current waveform diagram obtained from the embodiment of the present invention;

[0043] Figure 4 is the single - sided inrush current direction coefficient B obtained from the embodiment of the present invention L Figure

[0044] Figure 5 is the symmetric inrush current waveform diagram obtained from the embodiment of the present invention;

[0045] Figure 6 is the symmetric inrush current direction coefficient B obtained from the embodiment of the present invention L Figure;

[0046] Figure 7 is the symmetric inrush current operating coefficient QSI diagram obtained from the embodiment of the present invention;

[0047] Figure 8 is the internal fault current waveform diagram obtained from the embodiment of the present invention;

[0048] Figure 9 is the internal fault current direction coefficient B obtained from the embodiment of the present invention L Figure;

[0049] Figure 10 is the internal fault current operating coefficient QSI diagram obtained from the embodiment of the present invention;

[0050] Figure 11 is the saturated inrush current waveform diagram obtained from the embodiment of the present invention;

[0051] Figure 12 is the saturated inrush current direction coefficient B obtained from the embodiment of the present invention L Figure;

[0052] Figure 13 is the saturated inrush current operating coefficient QSI diagram obtained from the embodiment of the present invention;

[0053] Figure 14 is the saturated fault current waveform diagram obtained from the embodiment of the present invention;

[0054] Figure 15 is the saturated fault current direction coefficient B obtained from the embodiment of the present invention L Figure;

[0055] Figure 16The QSI diagram of the saturation fault current obtained in the embodiment of the present invention;

[0056] Figure 17 The waveform diagram of the arc fault current obtained in the embodiment of the present invention;

[0057] Figure 18 The B diagram of the arc fault current direction coefficient obtained in the embodiment of the present invention L diagram;

[0058] Figure 19 The QSI diagram of the arc fault current action coefficient obtained in the embodiment of the present invention. Detailed implementation manners

[0059] The present invention will be clearly described below in conjunction with the accompanying drawings in the embodiments of the present invention and specific embodiments. The description here is only used to explain the present invention, but not to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts, any modifications, equivalent replacements, improvements, etc., shall be included in the protection scope of the present invention.

[0060] Embodiment

[0061] The present invention provides a method for automatically locking the transformer differential protection based on the quadrilateral method,

[0062] As Figure 2 shown, it is the simulation model of this embodiment. In the figure, the power supply is connected in series with the primary side of the transformer through a circuit breaker and a current transformer. The connection method of the secondary side of the transformer is the same as that of the primary side except for the parameter settings. Build a simulation system according to the Figure 2 shown simulation model, and the three-phase transformer module is set as follows: the rated voltage is 220kV / 110kV, the rated capacity is 100MVA, the connection group is Ynd11, and a saturated iron core is selected. The three-phase power supply line voltage is set to 220kV, and the initial phase angle of phase A is set to 0°. The current transformer ratio is 300A / 5A or 600A / 5A. The system frequency is 50Hz, and the sampling frequency is set to 4000Hz. The simulation results of the arc fault current are obtained by adding a Mayr model module to the secondary side of the transformer. The simulation obtains single-sided inrush current, symmetrical inrush current, internal fault current, saturated inrush current, saturated fault current, and arc fault current, as shown in the accompanying drawings Figure 3 , Figure 5 , Figure 8 , Figure 11 , Figure 14 , Figure 17 shown. According to the Figure 1 shown process, the specific identification steps are as follows:

[0063] Step 1: Simulate the inrush current, internal three-phase short-circuit fault current, saturated inrush current, saturated fault current, and arc fault current respectively, collect the differential current signals for multiple cycles, and form a sampling data sequence {i1, i2, …, i w}, where N is the number of sampling data in one cycle, then w = 9N;

[0064] Step 2: Calculate the direction coefficient B L respectively according to the foregoing formula (1), and the results are as shown in Figure 4 , Figure 6 , Figure 9 , Figure 12 , Figure 15 , Figure 18 ;

[0065] Step 3: Compare B L with the threshold B set . If it is greater than the threshold B set , it is determined as single-sided inrush current; otherwise, go to Step 4. According to this criterion, Figure 3 is determined as single-sided inrush current, and the rest go to Step 4;

[0066] Step 4: Find the data with the largest absolute value in a cycle of differential current data sequence {i1, i2,......, i N}, and assign the absolute value of this data to M after taking the absolute value; compare the absolute values of the data in {i1, i2,......, i N} with 0.15M in turn, and store the data with absolute values greater than 0.15M in the data sequence {e1, e2,......, e L} in turn, and store the sequence numbers of these data in {i1, i2,......, i N} in the data sequence {t1, t2,......, t L} in turn; where L is the total number of data with absolute values greater than 0.15M in {i1, i2,......, i N};

[0067] Step 5: Let the variable h take values of 1, 2, 3,......, L in turn, retrieve all the data in the data sequence {t1, t2,......, t L}. If h satisfies formula (2), then delete {e h-0.05N , e h-0.05N+1 ,......, e h} from the data sequence {e1, e2,......, e L}, and at the same time delete {t h-0.05N , t h-0.05N+1 ,......, t h}Delete from the data sequence {t1, t2,......, t L}; If h satisfies Equation (3), then delete {e h+1 , e h+2 ,......, e h+0.05N} from the data sequence {e1, e2,......, e L}, and at the same time delete {t h+1 , t h+2 ,......, t h+0.05N} from the data sequence {t1, t2,......, t L};

[0068] Step 6: Assign the remaining data of the data sequence {e1, e2,......, e L} after Step 5 to the data sequence {x1, x2,......, x s}; Assign the remaining data of the data sequence {t1, t2,......, t L} after Step 5 to the data sequence {d1, d2,......, d s}; where s is the total number of remaining data after Step 5; Determine the quarter-step length q, and divide the data sequence {x1, x2,......, x s} into four parts: {x1, x2,…, x q}, {x q+1 , x q+2 ,…, x 2q}, {x 2q+1 , x 2q+2 ,…, x 3q}, {x 3q+1 , x 3q+2 ,…, x s}; At the same time, divide the data sequence {d1, d2,......, d s} into four parts: {d1, d2,…, d q}, {d q+1 , d q+2 ,…, d 2q}, {d 2q+1 , d 2q+2 ,…, d 3q}, {d 3q+1 , d 3q+2 ,…, d s};

[0069] Step 7: Obtain the values of the variables α1, β1, θ1 when Equation (4) reaches the minimum value; and obtain the values of the variables α2, β2, θ2 when Equation (5) reaches the minimum value;

[0070] Step 8: Calculate the action coefficient QSI through Equation (6), and the calculation results are as follows Figure 7 , Figure 10 , Figure 13 , Figure 16 , Figure 19 as shown;

[0071] Step 9: Compare QSI with the threshold Q set . If QSI is greater than the threshold Q set , it is determined that the differential current signal is symmetrical inrush current or saturated inrush current, and the transformer differential protection is blocked; otherwise, it is determined that the differential current signal is one of traditional fault current, saturated fault current or arc fault current, and the transformer differential protection is opened to allow the transformer differential protection to operate.

[0072] Figure 7 , Figure 10 , Figure 13 , Figure 16 , Figure 19 The determination results of are symmetrical inrush current, internal fault current, saturated inrush current, saturated fault current, and arc fault current respectively;

[0073] According to the calculated B based on the differential current data L , it shows that the present invention can correctly identify single-sided inrush current through the direction coefficient B L and block the differential protection; according to the calculated QSI, it shows that the present invention can correctly identify symmetrical inrush current, saturated inrush current and fault current by comparing QSI with the threshold.

[0074] The above is only the preferred embodiment of the present invention, but the protection scope of the present invention is not limited to the above embodiments. It should be noted that within the scope of the technical field of the present invention, without departing from the technical idea of the present invention, simple retouches, improvements, substitutions, simplifications and combinations made to the present invention should all be included in the protection scope of the present invention.

Claims

1. An automatic blocking method for transformer differential protection based on the quartering method, characterized in that, It includes the following steps: Step 1: Sample the differential current signal in the differential circuit of the power transformer; Step 2: If the differential current signal meets the differential protection criterion, go to Step 3; otherwise, return to Step 1; Step 3: Calculate the direction coefficient B according to formula (1) L :[[-END]] Wherein, N is the number of sampling times in a power frequency period; {i1, i2,......, i N} is a data sequence composed of differential current signal data in a newly acquired period, where r = 1, 2,......, N; Step 4: Compare B L with the threshold value B set ; if B L is greater than the threshold value B set , directly determine it as single-sided inrush current and block the transformer differential protection; otherwise, proceed to Step 5; Step 5: Find the data with the largest absolute value in the data sequence {i1, i2,......, i N}, take the absolute value of this data and assign it to M; compare the absolute values of the data in {i1, i2,......, i N} with 0.15M in sequence, and store the data whose absolute value is greater than 0.15M in the data sequence {e1, e2,......, e L} in sequence, and store the sequence numbers of these data in {i1, i2,......, i N} in the data sequence {t1, t2,......, t L} in sequence; where L is the total number of data in {i1, i2,......, i N} whose absolute value is greater than 0.15M; Step 6: Let the variable h take values of 1, 2, 3,......, L in turn, and retrieve all the data in the data sequence {t1, t2,......, t L}. If h satisfies equation (2), then delete {e h-0.05N , e h-0.05N+1 ,......, e h} from the data sequence {e1, e2,......, e L}, and at the same time delete {t h-0.05N , t h-0.05N+1 ,......, t h} from the data sequence {t1, t2,......, t L}; If h satisfies Equation (3), then {e h+1 , e h+2 ,......, e h+0.05N} is deleted from the data sequence {e1, e2,......, e L}, and at the same time, {t h+1 , t h+2 ,......, t h+0.05N} is deleted from the data sequence {t1, t2,......, t L}; Step 7: Assign the remaining data of the data sequence {e1, e2,......, e L} to the data sequence {x1, x2,......, x s}; Assign the remaining data of the data sequence {t1, t2,......, t L} to the data sequence {d1, d2,......, d s}; where s is the total number of remaining data after Step 6; Determine the quarter step size q, and divide the data sequence {x1, x2,......, x s} into four parts: {x1, x2, …, x q}, {x q+1 , x q+2 , …, x 2q}, {x 2q+1 , x 2q+2 , …, x 3q}, {x 3q+1 , x 3q+2 , …, x s}; At the same time, divide the data sequence {d1, d2,......, d s} into four parts: {d1, d2, …, d q}, {d q+1 , d q+2 , …, d 2q}, {d 2q+1 , d 2q+2 , …, d 3q}, {d 3q+1 , d 3q+2 , …, d s}; Step 8: Obtain the values of variables α1, β1, θ1 when Equation (4) reaches the minimum value; and obtain the values of variables α2, β2, θ2 when Equation (5) reaches the minimum value; In Equation (4) and Equation (5), f is the objective function value of the first part and the fourth part; g is the objective function value of the second part and the third part; α1, β1, θ1, α2, β2, θ2 are variables to be solved; Step 9: Calculate the operating coefficient QSI through Equation (6): Step 10: Compare QSI with the threshold Q set ; if QSI is greater than the threshold Q set , then determine that the differential current signal is symmetric inrush current or saturated inrush current, and block the transformer differential protection; otherwise, determine that the differential current signal is one of traditional fault current, saturated fault current, and arc fault current, open the transformer differential protection, and allow the transformer differential protection to operate.

2. The automatic blocking method for transformer differential protection based on the quartering method according to claim 1, characterized in that, In Step 7, the quarter-step length q is obtained through Equation (7): In the formula, is to round down towards negative infinity.

3. The automatic blocking method for transformer differential protection based on the quartering method according to claim 1, characterized in that, The threshold B in step 4 set is set to 0.

8.

4. The automatic blocking method for transformer differential protection based on the quartering method according to claim 1, characterized in that, The threshold Q in step 10 set has a value range of 0.25 to 0.35.

Images