A Fault Location Method for a Distribution Network with Distributed Generation Based on 5G Communication

Abstract

The present invention proposes a fault location method for a distribution network with distributed power sources based on 5G communication. At each node in the distribution network, a traveling wave detection device and a 5G communication terminal are installed to accurately capture the arrival time of the fault traveling wave generated at the fault point and the traveling wave generated by the reflection / refraction at the line boundary. The arrival time information of the traveling wave detected at each node is transmitted to the opposite node through 5G communication, and the distance between the fault point and the node is calculated to achieve accurate location of the fault in the distribution network with distributed power sources. The present invention solves the problems of the traditional double-ended traveling wave fault location being affected by the wave velocity, incorrect selection of traveling waves, and inaccurate capture time in the distribution network with distributed power sources, and greatly improves the accuracy of fault location.

Description

Technical Field

[0001] The present invention belongs to the field of relay protection for distribution networks, and particularly relates to a fault location method for a distribution network with distributed power sources based on 5G communication. Background Art

[0002] After a distribution network fails, rapid and accurate fault location is beneficial to the maintenance and fault isolation of the distribution network, and is of extremely important significance for the safe and stable operation of the distribution network. Traditional fault location methods include those based on various algorithms such as neural network algorithms, genetic algorithms, harmony algorithms, and particle swarm algorithms. The fault location methods based on such algorithms highly depend on the establishment of switching functions, and their adaptability and fault tolerance need to be improved. In addition, there are many fault location methods based on line electrical characteristic quantities, such as methods using line current, voltage, sequence components, and calculated impedance to achieve fault location. Most of these methods are affected by the operating state, topological structure, and transition resistance of the distribution network, and have certain engineering practical limitations.

[0003] In addition, to solve the problems of large losses and poor peak shaving in centralized power supply, new energy technologies have developed rapidly, making great contributions to energy conservation, loss reduction, and peak shaving. The way of new energy accessing the power grid is mainly through distributed power sources. The distributed power sources have small capacities and are scattered, causing the traditional distribution network to gradually evolve from a single-source radial network to a complex network with multiple power sources. A distribution network with distributed power sources has a relatively complex topological structure. When a fault occurs, the power flow changes from unidirectional flow to bidirectional flow, and multiple distributed power sources will provide fault current simultaneously, greatly increasing the branches through which the fault current flows, which will inevitably have a great impact on the traditional fault location methods based on electrical characteristic quantities.

[0004] To solve the above problems, many experts and scholars have extended the fault location methods in transmission networks to distribution networks. Among them, the traveling wave method has a relatively broad application prospect due to its advantages such as being unaffected by transition resistance and fault type. However, if the value of the traveling wave propagation speed deviates from the actual value, or the traveling wave detection device has a truncation error in the arrival time of the traveling wave due to a low sampling frequency, or the topological structure of the distribution network becomes complicated due to the access of distributed power sources, resulting in the traveling wave detection device detecting non-designated fault traveling waves, all of the above factors will cause the fault location result to deviate far from the actual value. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a fault location method for a distribution network with distributed power sources based on 5G communication to improve the accuracy of fault location.

[0006] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A fault location method for a distribution network with distributed power sources based on 5G communication, including:

[0008] S1: The startup of the fault location algorithm is started simultaneously with the startup criterion of the main protection. The traveling wave detection device installed at each node in the distribution network detects and records the current signal on the line in real time after the startup of the fault location algorithm;

[0009] S2: After performing the Karenbauer transform on the detected current signal, the α-mode component, β-mode component, and 0-mode component are obtained. The derivative waveform of the α-mode component is obtained by taking the derivative of the α-mode component;

[0010] S3: Based on the derivative waveform of the α-mode component, the four-point fitting method is used to capture the time t when the first fault traveling wave reaches the local line node M M , and the time t when the first fault traveling wave detected by the opposite line node N is obtained through 5G communication N ;

[0011] S4: Based on the time when the first fault traveling wave reaches the nodes on both sides of the line respectively and the total length of this section of the line, calculate the time range of the second fault traveling wave reflected from the fault point to the line node; within the time range, based on the derivative waveform of the α-mode component, the four-point fitting method is used to capture the second fault traveling wave reaching t fM ; The time t when the second fault traveling wave detected by the opposite line node is obtained through 5G communication fN ;

[0012] S5: Based on the time when the first and second fault traveling waves reach the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this section of the line, determine whether the fault occurs within this section. If so, enter step S6 to calculate the fault distance; otherwise, end the process;

[0013] S6: According to the time when the first and second fault traveling waves reach the nodes on both sides of the line and the total length of this section of the line, calculate the distances of the fault point from the nodes on both sides.

[0014] Furthermore, the time when the first fault traveling wave reaches is the moment when the fault traveling wave generated at the fault point directly propagates to the local node traveling wave detection device, and the time when the second fault traveling wave reaches is the moment when the fault traveling wave generated at the fault point is sequentially transmitted from the local node to the fault point and then reflected back to the local node by the fault point.

[0015] Furthermore, in S2, the Karenbauer transform is performed on the detected current signal, and the transformation formula is:

[0016] I a 、I b 、I c are the three-phase current signals on the line; Iα , I β , I0 are respectively the α-mode component, β-mode component and 0-mode component obtained by the Karen-Bell transform, and their discrete arrays are represented as I α = [i α1 , i α2 , …, i αN , I β = [i β1 , i β2 , …, i βN , I0 = [i 01 , i 02 , …, i 0N ;

[0017] Then, take the derivative of the α-mode component. The derivative method is: the derivative of the k-th data point is i′ αk = f s (i αk - i α(k-1) ), where

[0018] In the formula, i′ αk is the derivative of the k-th data point corresponding to the α-mode component, f s is the sampling frequency, i αk is the k-th data of the α-mode component, i α(k-1) is the (k - 1)-th data of the α-mode component. Then, the final derivative result of the α-mode component is represented as I′ α = [i′ α1 , i′ α2 , …, i′ αN .

[0019] Furthermore, based on the derivative waveform of the α-mode component, the specific method for capturing the arrival time of the fault traveling wave using the four-point fitting method is as follows:

[0020] According to the derivative waveform of the α-mode component, screen out all the maximum value points of the derivative, and select the four sampling points with the largest derivative values in the neighborhood of the maximum value of the derivative. The four sampling points with the largest derivative values in the neighborhood include the peak value, the sampling point one before the peak value, and the two sampling points after the peak value, or the peak value, the two sampling points before the peak value, and the sampling point after the peak value;

[0021] Assume that the current change rate at each sampling point changes linearly with time, and the rising and falling rates of the change rate are the same. The time corresponding to the true peak value point can be fitted from the four maximum value points near the peak value of the change rate, that is, the arrival time of the fault traveling wave.

[0022] Furthermore, in S4, based on the time when the fault traveling wave first reaches the nodes on both sides of the line and the total length of this section of the line, calculate the time range of the α-mode component traveling wave reflected from the fault point to the line node. The specific formula is:

[0023]

[0024]

[0025] Among them, the subscripts M and N respectively represent the nodes M and N at both ends of the branch, L is the total length of the line in this section, and v min is the minimum value of the traveling wave velocity, which is 0.936 times the speed of light, and v max is the maximum value of the traveling wave velocity, which is 0.987 times the speed of light.

[0026] Furthermore, in the step S5, based on the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this line section, it is determined whether the fault occurs within this section. The specific criterion is:

[0027] When is satisfied, it is determined that the fault point is located within this line section. When is satisfied, it is determined that the fault point is not within this line section, where Δt MN is the time required for the fault traveling wave to propagate from node M to node N in this line section.

[0028] Furthermore, the Δt MN is calculated according to the formula , where L is the total length of the line in this section and v is the traveling wave propagation velocity; or a traveling wave similar to the frequency component of the fault traveling wave is injected into the line during normal operation of the line, and the traveling wave detection device captures the arrival time of the traveling wave to obtain the actual value of Δt MN .

[0029] Furthermore, in the step S6, according to the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the total length of the line in this section, the distances from the fault point to the nodes on both sides are calculated. The specific formula is as follows:

[0030]

[0031] Among them, L is the total length of the line in this section, L fM is the distance from the fault point to node M, and L fN is the distance from the fault point to node N.

[0032] The beneficial effects of the present invention are as follows: The fault location method for a distribution network with distributed power sources based on 5G communication solves, on the one hand, the problem of high construction and maintenance costs of optical fiber communication, and on the other hand, solves the problems of the influence of wave velocity, incorrect selection of traveling waves, and inaccurate capture time on the traditional traveling wave fault location results in a distribution network with distributed power sources, greatly improving the accuracy of fault location. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of a distribution network structure with distributed power sources.

[0034] Figure 2 It is a schematic diagram of the fault line wave folding / reflection process provided by an embodiment of the present invention.

[0035] Figure 3 It is a schematic diagram of the first case of the four-point fitting method provided by the present invention.

[0036] Figure 4 It is a schematic diagram of the second case of the four-point fitting method provided by the present invention.

[0037] Figure 5 It is a flowchart of a fault location method for a distribution network with distributed power sources based on 5G communication provided by an embodiment of the present invention.

[0038] Figure 6 It is a schematic diagram of the arrival time of the first fault line wave transmitted from the fault point to node E captured by the four-point fitting method provided by an embodiment of the present invention.

[0039] Figure 7 It is a schematic diagram of the α modulus of the current detected at node E.

[0040] Figure 8 It is a schematic diagram of the α modulus of the current detected at node F. Detailed implementation manners

[0041] The present invention will be further described in detail below in combination with the above methods and the accompanying drawings of the specification.

[0042] As Figure 5 described, it is a flowchart of a fault location method for a distribution network with distributed power sources based on 5G communication provided by an embodiment of the present application, including:

[0043] S1: The start of the fault location algorithm is started simultaneously with the start criterion of the main protection. The traveling wave detection devices installed at each node in the distribution network detect and record the current signals on the line in real time after the start of the fault location algorithm;

[0044] S2: After performing the Karen-Bell transform on the detected current signals, α-mode components, β-mode components, and 0-mode components are obtained, and the derivative waveform of the α-mode components is obtained by taking the derivative of the α-mode components;

[0045] S3: Based on the derivative waveform of the α-mode components, the four-point fitting method is used to capture the time t M when the first fault line wave reaches the node M on the local line, and the arrival time t N of the first fault line wave detected at the node N on the opposite line is obtained through 5G communication;

[0046] S4: Calculate the time range of the second fault traveling wave reflected from the fault point to the line node based on the time when the first fault traveling wave reaches the nodes on both sides of the line and the total length of this section of the line; within this time range, based on the derivative waveform of the α-mode component, use the four-point fitting method to capture the second fault traveling wave reaching time t fM ; Obtain the time t when the second fault traveling wave detected at the opposite line node arrives through 5G communication fN ;

[0047] S5: Based on the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this section of the line, determine whether the fault occurs within this section. If so, enter step S6 to calculate the fault distance; otherwise, end the process;

[0048] S6: Calculate the distances of the fault point from the nodes on both sides according to the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the total length of this section of the line.

[0049] In a preferred embodiment, the arrival time of the first fault traveling wave is the moment when the fault traveling wave generated at the fault point directly propagates to the traveling wave detection device at the local node, and the arrival time of the second fault traveling wave is the moment when the fault traveling wave generated at the fault point is sequentially transmitted from the local node to the fault point and then reflected back to the local node by the fault point.

[0050] In a preferred embodiment, in S2, perform the Karenbauer transform on the detected current signal, and the transformation formula is:

[0051] I a 、I b 、I c are the three-phase current signals on the line; I α 、I β 、I0 are the α-mode component, β-mode component, and 0-mode component obtained through the Karenbauer transform respectively, and their discrete arrays are expressed as I α =[i α1 ,i α2 ,…,i αN , I β =[i β1 ,i β2 ,…,i βN , I0=[i 01 ,i 02 ,…,i 0N ;

[0052] Then take the derivative of the α-mode component, and the derivative method is: the derivative of the kth data point is i′ αk =f s (i αk -iα(k-1) ), where

[0053] in the formula, i′ αk is the derivative of the k-th data point corresponding to the α-mode component, and f s is the sampling frequency, and i αk is the k-th data of the α-mode component, and i α(k-1) is the (k - 1)-th data of the α-mode component. Then the final derivative result of the α-mode component is expressed as I′ α = [i′ α1 , i′ α2 , …, i′ αN .

[0054] In a preferred embodiment, based on the derivative waveform of the α-mode component, the specific method for capturing the arrival time of the fault traveling wave using the four-point fitting method is as follows:

[0055] According to the derivative waveform of the α-mode component, all the maximum points of the derivative are screened out, and four sampling points with the largest derivative values in the neighborhood of the maximum value of the derivative are selected. The four sampling points with the largest derivative values in the neighborhood include the peak value, the sampling point one before the peak value, and the two sampling points after the peak value, or the peak value, the two sampling points before the peak value, and the sampling point after the peak value;

[0056] Assume that the current change rate at each sampling point changes linearly with time, and the rising and falling rates of the change rate are the same. The time corresponding to the true peak point can be fitted from the four maximum points near the peak value of the change rate, that is, the arrival time of the fault traveling wave.

[0057] In a preferred embodiment, in S4, based on the time when the fault traveling wave first arrives at the nodes on both sides of the line and the total length of this section of the line, the time range of the α-mode component traveling wave reflected from the fault point to the line node is calculated. The specific formula is:

[0058]

[0059]

[0060] where the subscripts M and N respectively represent the nodes M and N at both ends of the branch, L is the total length of this section of the line, and v min is the minimum value of the traveling wave speed, which is 0.936 times the speed of light, and v max is the maximum value of the traveling wave speed, which is 0.987 times the speed of light.

[0061] In a preferred embodiment, in S5, based on the first and second arrival times of the fault traveling wave at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this section of the line, it is determined whether the fault occurs within this section. The specific criterion is:

[0062] When it satisfies When it is determined that the fault point is within this line section, when is satisfied, it is determined that the fault point is not within this line section, where Δt MN is the time required for the fault traveling wave to propagate from node M to node N of this line section.

[0063] In a preferred embodiment, the Δt MN is calculated according to the formula where L is the total length of the line in this section, and v is the traveling wave propagation speed; or a traveling wave similar to the frequency component of the fault traveling wave is injected into the line when the line is operating normally, and the traveling wave detection device captures the arrival time of the traveling wave to obtain the actual value of Δt MN .

[0064] In a preferred embodiment, in S6, according to the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the total length of the line in this section, the distances from the fault point to the nodes on both sides are calculated. The specific formula is as follows:

[0065]

[0066] where L is the total length of the line in this section, L fM is the distance from the fault point to node M, and L fN is the distance from the fault point to node N.

[0067] Next, in combination with Figure 1 the shown radial distribution network with distributed power sources, the method provided by this application will be specifically derived. As Figure 1 shown, A is the main power source of the distribution network, and B, C, and D are distributed power sources connected to the distribution network. Among them, B and C are connected to the same node E in the main line. It is stipulated that the line where the power flow direction of the main power source is located is the main line, Figure 1 in which the main line is A-E-G-H, and the lines where the power flow directions of the distributed power sources connected to the main line are located are branch lines, Figure 1 in which B-F, C-F, F-E, and D-G are branch lines. 5G communication terminals and traveling wave detection devices are installed at all power sources and nodes.

[0068] Taking Figure 1 the line F-E-G-D composed of the main line E-G and the branch lines F-E and D-G in as an example for illustration:

[0069] As Figure 2 shown, when a fault point occurs on the main line EG, the fault traveling wave generated by the fault point will propagate along the lines on both sides of the fault point, and will reach the nodes at both ends of the line at times t E and t G respectively. Subsequently, reflections / refractions will occur at nodes E and G, and the fault traveling wave reflected to the fault point f will be reflected / refracted again. The reflected fault traveling waves will reach at times tfE , t fG reaches nodes E and G for the second time at the moment. For convenience of analysis, only the fault traveling waves reflected / refracted at the fault point f of nodes E and G are drawn in the figure. The solid lines are the traveling waves reflected from the boundary point, and the dashed lines are the traveling waves refracted at the boundary point. The specific steps for accurate fault location are as follows:

[0070] Step1: The fault location algorithm starts simultaneously with the starting criterion of the main protection. The traveling wave detection devices installed at each node in the distribution network detect and record the three-phase current signals I a , I b , I c on the line in real time after the fault location algorithm starts, where I a , I b , I c are all discrete signals. If the total data length is N, then there are I a = [I a1 , I a2 , …, I aN , I b = [I b1 , I b2 , …, I bN , I c = [I c1 , I c2 , …, I cN .

[0071] Step2: Perform the Karenbauer transform on the collected three-phase current to obtain the line-mode current component and the zero-mode current component. The specific transformation method is as follows:

[0072]

[0073] In Equation (1), I α , I β , I0 are the α-mode component, β-mode component, and 0-mode component obtained by the Karenbauer transform respectively. Their discrete arrays can also be expressed as I α = [i α1 , i α2 , …, i αN , I β = [i β1 , i β2 , …, i βN , I0 = [i 01 , i 02 , …, i 0N .

[0074] Derive the α-mode component of the sampled current to obtain the derivative value of the α-mode component of the current. The specific implementation method is: for the discrete current α-mode component signal I obtained by the Karenbauer transformα = [i α1 , i α2 , …, i αN , the derivative method is as follows:

[0075] i′ αk = f s (i αk - i α(k-1) ) (2)

[0076] In formula (2), i′ αk is the derivative of the k-th data point corresponding to the α-mode component, f s is the sampling frequency, i αk is the k-th data of the α-mode component, i α(k-1) is the (k - 1)-th data of the α-mode component. Then the final derivative result of the α-mode component can be expressed as I′ α = [i′ α1 , i′ α2 , …, i′ αN .

[0077] Step3: Capture the first α-mode component traveling wave reflected from the fault point to the node. Since the fault traveling wave can directly propagate from the fault point f to the node, the first fault traveling wave α-mode component detected by the traveling wave detection device at the node is the α-mode component traveling wave reflected from the fault point to the node, without being interfered by the reflections of other nodes. Therefore, based on the α-mode component derivative waveform, the four-point fitting method can be used to directly capture the time when the first fault traveling wave reaches the local line node, and the time when the first fault traveling wave detected by the opposite line node reaches can be obtained through 5G communication. The specific implementation method for correcting the true arrival time of the fault traveling wave by the four-point fitting method is as follows:

[0078] Step31: Screen out all the maximum value points of the derivative of the α-mode component, and select the four largest derivative values within the neighborhood of the maximum value point, including the selected maximum value point and the two data points before and the one data point after the maximum value point or the one data point before and the two data points after the maximum value point. That is, if the maximum value is i′ αx , then the four maximum values selected are [i′ α(x-1) , i′ αx , i′ α(x+1) , i′ α(x+2) or [i′ α(x-2) , i′ α(x-1) , i′ αx , i′ α(x+1) .

[0079] Step32: Perform four-point fitting correction according to the selected four maximum value points to calculate the arrival time of the fault traveling wave. The specific method is as follows:

[0080] As shown in the accompanying drawings of the specification Figure 3 The abscissa represents time and the ordinate represents the derivative value. M0 is the maximum value of the derivative of the α-mode component. M1, M2, M3, and M4 are the four largest derivative values within the neighborhood of the selected maximum derivative value. t, t1, t2, t3, and t4 are the corresponding times of the derivative values M0, M1, M2, M3, and M4 respectively, and the corresponding coordinate points are O, O1, O2, O3, and O4. At this time, there are two derivative values M2 in front of the maximum derivative value and two derivative values M3 and M4 behind it, corresponding to the four points [i′ α(x-1) ,i′ αx ,i′ α(x+1) ,i′ α(x+2) selected in the first case. At this time, according to Figure 3 it can be known that the following is satisfied:

[0081] M0 = M2 + (t - t2)·K = M3 + (T s -(t - t2))·K (6)

[0082] In formula (6), T s is the sampling period, and t - t2 is the time interval between the sampling point O2 and the arrival time t of the fault traveling wave. The value of K is the slope of the waveform change rate, that is, the absolute value of the slope of the waist of the isosceles triangle in the figure. To reduce the error caused by noise, the slope K is taken as the arithmetic mean of the slopes calculated from two points on the left and right sides of O, that is:

[0083]

[0084] According to formulas (6) and (7), the time interval t - t2 between the sampling point O2 and the arrival time point O of the fault traveling wave can be obtained:

[0085]

[0086] Then the arrival time of the fault traveling wave obtained by the four-point fitting method is:

[0087]

[0088] In formula (9), t2 is the time corresponding to the derivative point O2, and t is the arrival time of the fault traveling wave obtained by the four-point fitting method, that is, the time corresponding to the extreme point O.

[0089] When the four selected maximum derivative value points satisfy the second condition, that is, when the maximum value point is i′ αx , the four selected maximum values are [i′ α(x-2) ,i′ α(x-1) ,i′ αx ,i′ α(x+1) , corresponding to the accompanying drawings of the specification Figure 4, at this time, the derivative point M3 is a maximum point. According to the above derivation, similarly, the arrival time of the fault traveling wave is the same as Equation (9).

[0090] Step4: Capture the α-mode component traveling wave reflected from the fault point to the node for the second time. As shown in the attached figure Figure 1 When the fault point is very close to node G, as shown, the second fault traveling wave captured by node E may be the fault traveling wave reflected from node G to node E, rather than the fault traveling wave reflected from the fault point f to node E. To prevent the interference of reflected traveling waves from other nodes, it is necessary to screen out the fault traveling wave reflected from the fault point f to node E for the second time. The specific scheme is as follows:

[0091] Let the fault occurrence time be t f , according to the traveling wave propagation theory, there is

[0092]

[0093] In Equation (3), L fE is the distance from the fault point f to node E, L fG is the distance from the fault point f to node G, L is the total length of the branch EG, t E is the time when the traveling wave first arrives at node E, t G is the time when the traveling wave first arrives at node G. From Equation (3), the times when the traveling wave detection devices at nodes E and G detect the second fault traveling wave reflected from the fault point can be obtained as

[0094]

[0095] Since it is difficult to accurately obtain the engineering actual value of the traveling wave propagation speed v in Equation (4), the time determined by Equation (4) is only an estimated value. According to engineering experience, the traveling wave propagation speed of the α-mode component is generally taken as 0.936 to 0.987 times the speed of light. When v takes the minimum value v min , that is, 0.936 times the speed of light, it is the latest arrival time of the fault traveling wave. When v takes the maximum value v max , that is, 0.987 times the speed of light, it is the earliest arrival time of the fault traveling wave. There is

[0096]

[0097] Capturing the fault traveling wave within the time range of Equation (5) can narrow the traveling wave capture range, prevent the interference caused by refraction / reflection of other nodes, and greatly improve the accuracy of capturing the traveling wave reflected from the fault point f to the node.

[0098] Based on the derivative of the sampled current α-mode component and the time range of the fault traveling wave reflected from the fault point f for the second time to the node, within the time range, the true arrival time of the fault traveling wave is corrected by the four-point fitting method, and the arrival time of the second fault traveling wave detected at the opposite line node is obtained through 5G communication. The specific implementation method is as follows:

[0099] Step41: Screen out all the maximum value points of the derivative of the α-mode component, and select the four largest derivative values within the neighborhood of the maximum value point, including the selected maximum value point and the two data points before and the one data point after the maximum value point or the one data point before and the two data points after the maximum value point. That is, if the maximum value is i′ αx , then the four maximum values selected are [i′ α(x-1) , i′ αx , i′ α(x+1) , i′ α(x+2) or [i′ α(x-2) , i′ α(x-1) , i′ αx , i′ α(x+1) .

[0100] Step42: Perform four-point fitting correction according to the selected four maximum value points to calculate the arrival time of the fault traveling wave. The specific method is as follows:

[0101] As shown in the accompanying drawings of the specification Figure 3 , the abscissa is time and the ordinate is the derivative value. M0 is the maximum value of the derivative of the α-mode component, M1, M2, M3, and M4 are the four largest derivative values within the neighborhood of the selected derivative maximum value, t, t1, t2, t3, and t4 are the corresponding times of the derivative values M0, M1, M2, M3, and M4, and the corresponding coordinate points are O, O1, O2, O3, and O4. At this time, there are two derivative values M2 on the front side of the derivative maximum value and two derivative values M3 and M4 on the back side, corresponding to the four points [i′ α(x-1) , i′ αx , i′ α(x+1) , i′ α(x+2) selected in the first case. At this time, according to Figure 3 , it can be known that:

[0102] M0 = M2 + (t - t2)·K = M3 + (T s - (t - t2))·K (6)

[0103] In formula (6), T s is the sampling period, and t - t2 is the time interval between the sampling point O2 and the arrival time t of the fault traveling wave. The value of K is the slope of the waveform change rate, that is, the absolute value of the slope of the waist of the isosceles triangle in the figure. To reduce the error caused by noise, the slope K is taken as the arithmetic mean of the slopes calculated from the two points on the left and right sides of O, that is:

[0104]

[0105] According to Equations (6) and (7), the time t - t2 between the sampling point O2 and the arrival time point O of the fault traveling wave can be obtained:

[0106]

[0107] Then, the arrival time of the fault traveling wave obtained by the four-point fitting method is:

[0108]

[0109] In Equation (9), t2 is the time corresponding to the derivative point O2, and t is the arrival time of the fault traveling wave obtained by the four-point fitting method, that is, the time corresponding to the extreme point O.

[0110] When the four selected derivative maximum points satisfy the second condition, that is, when the maximum point is i′ αx , and the four selected maximum values are [i′ α(x-2) , i′ α(x-1) , i′ αx , i′ α(x+1) , corresponding to the accompanying drawings of the specification Figure 4 , at this time, the derivative point M3 is the maximum point. According to the above derivation, the arrival time of the fault traveling wave is obtained in the same way as the same as Equation (9).

[0111] Step5: Based on the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this line section, locate the fault point in the branch section. The specific implementation plan is as follows:

[0112] When the traveling wave device detects that the fault traveling wave is reflected twice from the fault point f to the two end nodes E and G at times t E , t fE and t G , t fG respectively, according to the traveling wave propagation theory, the time interval between the traveling wave detection device continuously capturing the specified fault traveling wave twice is:

[0113]

[0114] Define the time required for the fault traveling wave to travel from end E to end G of the line as Δt EG , then there is

[0115]

[0116] When the fault occurs inside the branch EG, satisfying L fE + L fG = L, then from Equations (10) and (11), it can be known that

[0117]

[0118] When the fault occurs outside the branch EG, such as Figure 2 At f1 or f2, there is L fE +L fG >L, then from equations (10) and (11) we can know

[0119]

[0120] In formula (12) and formula (13), Δt E With Δt G It can be calculated from the time detected by the traveling wave detection device:

[0121]

[0122] The Δt given by equation (11) EG The expression is still affected by the wave velocity v and can only be used to determine Δt EG In engineering practice, to obtain Δt EG To accurately calculate the value, a traveling wave with a frequency component similar to that of the fault traveling wave can be injected into the line during normal operation, and the arrival time of the traveling wave can be captured by the traveling wave detection device to obtain Δt EG The actual value of .

[0123] Therefore, the complete criterion for locating the fault point branch section is

[0124]

[0125] Step 6: If Step 5 determines whether the fault occurs in this section, further calculate the precise location of the branch where the fault point is located. The specific implementation plan is as follows:

[0126] By combining equation (3) and equation (10), we can get the distances between the fault point and nodes E and G respectively:

[0127]

[0128] Since L is a known quantity, Δt E , Δt G They are all calculated by the traveling wave detection device in conjunction with 5G communication. It can be seen from formula (16) that the fault location result obtained is independent of the wave velocity, that is, on the same line, it is only necessary to satisfy the uniform distribution of line parameters, that is, the traveling wave velocity is a constant, which can eliminate the error introduced by the wave velocity change caused by different parameters of different lines.

[0129] The following is attached Figure 1The system data of the distribution network topology structure with distributed power sources shown is described in more detail. Taking the attached figure Figure 1 The specific implementation plan is described by taking the distribution network topology structure with distributed power sources shown as an example. Among them, node A is the main power source, B, C, and D are the connected distributed power sources, and node H is the load node. A traveling wave detection device and a 5G communication terminal are installed at each node, which are used for detecting the wavefront of the fault traveling wave and communication respectively. The line lengths are AE = 20.1 km, EG = 3.9 km, GH = 2.6 km, BF = 2.4 km, CF = 3.2 km, FE = 4.3 km, DG = 2.7 km, and the sampling frequency is 10 MHz.

[0130] Example: When a fault occurs on the branch EF where the distributed power source is located, 1 km away from node F. The fault location process is as follows:

[0131] After the fault occurs, the fault traveling wave propagates to both sides from the fault point. Since the fault point is closer to node F on the branch EF, nodes F and E successively sense the fault traveling wave. Subsequently, nodes F and E refract the fault traveling wave to the other nodes and reflect the fault traveling wave to the fault point at the same time. The existence of the fault point will also cause the traveling wave to refract and reflect at the fault point. Therefore, the propagation process of the fault traveling wave is very complex. When the fault traveling wave generated by the fault point reaches nodes F and E, nodes F and E use the aforementioned four-point fitting method, and the arrival time of the first fault traveling wave at nodes F and E can be accurately captured by Equation (8). In order to capture the second fault traveling wave reflected from the fault point to both ends of the branch EF, rather than the fault traveling wave reflected from other nodes, Equation (5) is used to estimate the time period of the arrival time of the second fault traveling wave, and the current waveform in this time period is intercepted, and then the arrival time of the fault traveling wave in this time period is captured by Equation (8). For nodes other than nodes F and E, the arrival time of the two fault traveling waves reflected from the fault point to the node is captured in the same way as above. When the two arrival times of the fault traveling wave are captured at both ends of each branch, the upstream node transmits the two fault times to the downstream node through the 5G communication terminal, that is, it is transmitted according to the pre-fault power flow direction. For the fault branches F and E, the 5G communication terminal of node F is transmitted to node E through 5G communication. Subsequently, the section of the fault point is discriminated by Equation (15). In this case, only the FE branch satisfies the first formula in Equation (15), and the other branches satisfy the second formula. Thus, it is discriminated that the fault point appears on the FE branch. Subsequently, according to the arrival time of the fault traveling wave captured by nodes FE, the specific position of the fault point can be calculated by Equation (16). The corresponding flow chart is as shown in the attached figure Figure 5 shown. The specific method is as follows:

[0132] If the fault occurs at 0 s, the α-mode current components at nodes E and F are respectively as shown in the attached figure Figure 7 、 Figure 8As shown below, all the following moments are represented by the number of sampling points. First, the four-point fitting method is used to capture the first fault traveling wave reflected from the fault point to nodes E and F. Taking the first fault traveling wave captured by node E as an example for illustration: As shown in the attached drawing Figure 6 shown, they are the four maximum points within the neighborhood of the maximum point of the fault traveling wave. The coordinates corresponding to the four derivative maximum points are O1(113, 6.09), O2(114, 6.59), O3(115, 6.95), and O4(117, 6.41) respectively. According to Equation (9), the arrival time of the wavefront obtained by the four-point fitting method can be calculated as t E = 115.81. Similarly, t F = 35.12 is calculated. Further, the time range of the fault traveling wave reflected from the fault point f to the node for the second time is judged by Equation (5): 227.64 ≤ t fE ≤ 239.45. The time range corresponds to the dashed box in the attached drawing Figure 7 . Thus, within this time range, t fE = 231.62 is further obtained by the four-point fitting method. Similarly, the range of t fF can be obtained, and t fF = 70.09 is further obtained by the four-point fitting method. The remaining nodes are calculated in the same way as nodes E and F and are successively substituted into Equation (15) for discrimination. The moment information t F , t fF is transmitted from node F to node E through 5G communication. Substituting the data of nodes E and F, it is obtained that is approximately equal to 1, then it can be judged that the fault occurs in the EF branch. Further, according to Equation (16), it can be calculated that the distance between the fault point f and the node is 0.9929 km, and the error is only 0.0071 km. The fault location result is very accurate.

[0133] The above has introduced the embodiments of the present application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application. At the same time, those skilled in the art, based on the idea of the present application, the changes or deformations made in the specific implementation manner and application scope of the present application all belong to the protection scope of the present application. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A fault location method for a distribution network with distributed power sources based on 5G communication, characterized in that, Including: S1: The startup of the fault location algorithm is started simultaneously with the startup criterion of the main protection. The traveling wave detection device installed at each node in the distribution network detects and records the current signal on the line in real time after the startup of the fault location algorithm. S2: After performing the Karen-Bell transform on the detected current signal, the α-mode component, β-mode component, and 0-mode component are obtained. The derivative of the α-mode component is calculated to obtain the derivative waveform of the α-mode component. S3: Based on the derivative waveform of the α-mode component, use the four-point fitting method to capture the time t when the first fault traveling wave reaches the line node M on this side M , and obtain the arrival time t of the first fault traveling wave detected at the line node N on the opposite side through 5G communication N ; S4: Calculate the time range of the second fault traveling wave reflected from the fault point to the line node based on the time when the first fault traveling wave reaches the nodes on both sides of the line and the total length of the line section in this area; within the time range, capture the second fault traveling wave reaching time \(t\) based on the derivative waveform of the α-mode component using the four-point fitting method fM ; Obtain the arrival time \(t\) of the second fault traveling wave detected at the opposite line node through 5G communication fN ; S5: Based on the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this line section, it is judged whether the fault occurs in this section. If so, go to step S6 to calculate the fault distance; otherwise, end the process. S6: According to the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the total length of this line section, calculate the distances of the fault point from the nodes on both sides.

2. The fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, wherein, The arrival time of the first fault traveling wave is the moment when the fault traveling wave generated at the fault point directly propagates to the traveling wave detection device of the local node. The arrival time of the second fault traveling wave is the moment when the fault traveling wave generated at the fault point is sequentially transmitted from the local node to the fault point and then reflected back to the local node after reflection at the fault point.

3. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, characterized in that, In S2, the Karen-Bell transform is performed on the detected current signal, and the transformation formula is: I a , I b , I c are three-phase current signals on the line; I α , I β , I0 are the α-mode component, β-mode component, and 0-mode component obtained by the Karenbauer transform respectively, and their discrete arrays are represented as I α = [i α1 , i α2 , …, i αN , I β = [i β1 , i β2 , …, i βN , I0 = [i 01 , i 02 , …, i 0N ; Then, the derivative of the α-mode component is calculated. The calculation method is as follows: the derivative of the k-th data point is i' αk = f s (i αk - i α(k-1) ), where where i' αk is the derivative of the k-th data point corresponding to the α-mode component, f s is the sampling frequency, i αk is the k-th data of the α-mode component, i α(k-1) is the (k - 1)-th data of the α-mode component, then the final derivative result of the α-mode component is expressed as I' α = [i' α1 , i' α2 , …, i' αN .

4. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, characterized in that, The specific method for capturing the arrival time of the fault traveling wave using the four-point fitting method based on the derivative waveform of the α-mode component is: According to the derivative waveform of the α-mode component, all the maximum value points of the derivative are selected, and four sampling points with the largest derivative values in the neighborhood of the maximum value of the derivative are selected. The four sampling points with the largest derivative values in the neighborhood include the peak value, the sampling point before the peak value, and the two sampling points after it, or the peak value, the two sampling points before the peak value, and the sampling point after it. Assume that the current change rate at each sampling point changes linearly with time, and the rising and falling rates of the change rate are the same. The moment corresponding to the true peak value point, that is, the arrival time of the fault traveling wave, can be fitted by the four maximum value points near the peak value of the change rate.

5. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, characterized in that, In S4, based on the arrival times of the first fault traveling wave at the nodes on both sides of the line and the total length of this line section, the time range of the second α-mode component traveling wave reflected from the fault point to the line node is calculated. The specific formula is: Among them, the subscripts M and N respectively represent the nodes M and N at both ends of the branch, L is the total length of the line in this section, v min is the minimum value of the traveling wave velocity, which is 0.936 times the speed of light, v max is the maximum value of the traveling wave velocity, which is 0.987 times the speed of light.

6. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, characterized in that, In S5, based on the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the time for the fault traveling wave to travel back and forth in this line section, it is judged whether the fault occurs in this section. The specific criterion is: When is satisfied, it is determined that the fault point is within this line section. When is satisfied, it is determined that the fault point is not within this line section, where Δt MN is the time required for the fault traveling wave to propagate from node M to node N of this line section.

7. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 6, characterized in that, The said Δt MN is obtained according to the formula where L is the total length of the line section and v is the traveling wave propagation speed; or a traveling wave similar to the frequency components of the fault traveling wave is injected into the line during normal line operation, and the arrival time of the traveling wave is captured by the traveling wave detection device to obtain the actual value of Δt MN .

8. A fault location method for a distribution network with distributed power sources based on 5G communication according to claim 1, characterized in that In S6, according to the arrival times of the first and second fault traveling waves at the nodes on both sides of the line and the total length of this line section, the distances of the fault point from the nodes on both sides are calculated. The specific formula is as follows: Among them, L is the total length of the line in this section, L fM is the distance from the fault point to node M, L fN is the distance from the fault point to node N.

Images