Time domain method fault positioning method for high-voltage direct-current transmission line
Through the time domain method, the uniform transmission line equation and Kendall tau correlation coefficient of the high-voltage DC transmission line are derived, and the rapid and accurate positioning of the fault of the high-voltage DC transmission line is achieved, which solves the problem of fault positioning in complex environments and improves the stable operation of the power system. and work efficiency.
Patent Information
- Application Number
- CN202410999693.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-08
AI Technical Summary
High-voltage DC transmission lines have difficulty in fault positioning in complex environments, resulting in long fault positioning time and low accuracy, which affects the stable operation and working efficiency of the power system.
By using the time domain method, the uniform transmission line equation of the high-voltage DC transmission line is derived, the voltage and current distribution matrix is calculated using the variational iteration method and the Kendall tau correlation coefficient, and the fault location is gradually accurately measured until the accuracy requirements are met.
It realizes fast and accurate fault positioning, improves the accuracy and efficiency of distance measurement results, reduces manpower and material consumption, and reduces economic losses.
Smart Images

Figure SMS_1 
Figure SMS_2 
Figure SMS_3
Abstract
Description
Technical Field
[0001] The invention relates to a time domain fault location method for a high-voltage direct current transmission line, and belongs to the technical field of power system relay protection. Background Art
[0002] High-voltage direct current transmission lines have many advantages, such as large transmission capacity, long transmission distance, low line cost, convenient grid interconnection, and easy power regulation.
[0003] In my country, the required cross-regional transmission distance is generally long, and the DC transmission lines often need to pass through areas with complex terrain and special environments, resulting in a harsh operating environment. Not only will it be affected by natural environments such as mountains, hills, and deserts, but it may also be affected by extreme weather such as heavy rain, lightning, hail, and typhoons, making faults more likely to occur. Moreover, after a fault occurs, the inspection is difficult due to the long inspection distance and complex inspection conditions. Therefore, to ensure the normal and stable operation of high-voltage DC transmission lines, finding a fast and accurate fault location technology can not only quickly clear faults and restore the stable operation of the power system in a timely manner, but also save manpower and material resources, improve the work efficiency of power system personnel, and greatly reduce the economic losses caused by faults. Summary of the invention
[0004] The purpose of the present invention is to provide a time domain method for locating faults in a high voltage direct current transmission line, so as to achieve rapid fault location and ensure the stable operation of the power system.
[0005] The present invention adopts the following technical scheme: a time domain method for fault location of a high-voltage direct current transmission line. Before a fault occurs, a derivation formula for solving a uniform transmission line equation of the high-voltage direct current transmission line is first derived; when a fault occurs in the transmission line, fault data at the beginning and end of the fault are measured and brought into the derivation formula to obtain the voltage and current distribution matrices at the beginning and end respectively, and the similarity of voltage waveforms at the same position is compared using the Kendall tau correlation coefficient, and the highest similarity point is the fault position obtained by the first calculation; the lengths of the two distance intervals before and after the first distance measurement result are taken as a new line, and the above steps are cycled after updating various data, and the distance measurement result is gradually refined until the difference between the two iterative results before and after meets the accuracy requirement.
[0006] The specific steps are as follows:
[0007] Step 1: Use the variational iteration method to solve the uniform transmission line equation of the high-voltage DC transmission line and derive the expressions of voltage and current;
[0008] Step 2: Substitute the measurement data at the beginning and end into the voltage and current expressions respectively, calculate the voltage and current distribution matrix, use the two-terminal method, list the distance measurement equation, and calculate the fault location;
[0009] Step 3: Use the binary search method to establish a new line with the first distance measurement result as the center, update its parameters, perform the above calculation again, and iterate the fault results until the accuracy requirements are met.
[0010] Step 1 is as follows:
[0011] Assume that the resistance R, inductance L, capacitance C, reactance G per unit length of the high-voltage DC transmission line, and the total length of the line is l. Therefore, the uniform transmission line equation of the high-voltage DC transmission line can be expressed as:
[0012]
[0013] Where: u(x,t) and i(x,t) are the instantaneous voltage and instantaneous current of the HVDC transmission line at any time t and any position x. The known initial conditions of the partial differential equation are: the instantaneous values of voltage and current at the beginning and end faults: u(0,t), u(l,t), i(0,t), i(l,t), and the target is the voltage and current distribution along the line at the fault time: u(x,t), i(x,t).
[0014] The variational iteration method (VIM) in the field of mathematics is selected to solve the above partial differential equations. First, the central finite difference method is applied to convert the time derivatives of voltage and current into the quotient of the difference between two adjacent moments and twice the time interval, and the above partial differential equations are transformed into the following differential equations:
[0015]
[0016] Using VIM to solve the above differential equations, we can get:
[0017]
[0018] In the formula, v j,k (x), i j,k (x) represents the instantaneous value of voltage and current at position x at time j after k iterations. From the above formula, it can be seen that the voltage and current values after any k iterations can be obtained one by one from the initial measurement values.
[0019] Step 2 is as follows:
[0020] Step 2.1: After solving the partial differential equation, the initial measurement data at the beginning and the end are substituted into the above formula to obtain the voltage and current distribution matrix over time and position derived at the beginning and the voltage and current distribution matrix over time and position derived at the end.
[0021] Step 2.2: From the relevant knowledge of power system analysis, it can be known that when a high-voltage direct current transmission line fails, the voltage to ground at the fault point is the smallest. Therefore, the voltage derived from the beginning and end of the line to the fault location is the lowest and closest. Therefore, the Kendall tau correlation coefficient is used to calculate the similarity of the voltage waveforms derived from the data at the beginning and end to the same point. The point with the highest similarity is the fault location.
[0022] Step 3 is as follows:
[0023] Step 3.1: After the first calculation, the interval with the first distance measurement result xf1 as the center, the two points before and after Δx as the endpoints, and 2Δx as the total length is considered as a new line. At this time, 0.5Δx and 0.5Δt are taken as Δx and Δt in the second calculation, and 2Δx is taken as the total length l of the line after the second iteration.
[0024] After repeated experiments, it is found that the initial data of the line head and end are selected according to the following scheme, and the distance measurement result is the most accurate. If the fault location calculated for the first time is located at The entire line is For the entire length of the line, the voltage of the point before the fault position derived from the head end is taken as the initial voltage of the head end after iteration, and the voltage of the point after the fault position derived from the tail end is taken as the initial voltage of the tail end after iteration;
[0025] If the fault location calculated for the first time is located at the front of the entire line or after Then take the voltage of the point before the fault position derived from the end as the initial voltage of the first end after iteration, and take the voltage of the point after the fault position derived from the first end as the initial voltage of the end after iteration. After updating all the data, follow the second step to perform the distance calculation again, and calculate successively until the distance measurement results xf before and after are satisfied. n With xf n+1 The error between them is less than the given value C set ,Right now:
[0026] The beneficial effects of the present invention are:
[0027] 1) The present invention uses the initial measurement data at the beginning and the end to respectively calculate the distribution matrix of the voltage and current along the time and distance after the fault, without requiring a high sampling frequency, and is suitable for the 10kHz sampling rate of the recording device that is now widely used;
[0028] 2) The present invention determines the fault location by comparing the similarity of the voltage waveforms at the same location, and then uses the binary method to iterate and accurately measure the results until the difference between the results of the two iterations before and after meets the accuracy requirements. The entire process does not require complex processing and transformation, which improves the accuracy of the distance measurement results, thereby improving the accuracy and efficiency of determining the fault location. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 A flow chart of a fault location method in one embodiment of the present invention;
[0030] Figure 2 A comparison chart of the fault distances calculated before and after the time array is deviated;
[0031] Figure 3 It is the fitting curve after weighted least squares fitting. DETAILED DESCRIPTION
[0032] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.
[0033] Example: Figure 1 As shown in FIG. 1 , a time domain fault location method for a high voltage direct current transmission line is used. Table 1 shows the simulation model parameters.
[0034] Table 1 Simulation model parameters
[0035] parameter Parameter Value Voltage(kV) 800 Power frequency (Hz) 50 Line length (km) 500 Sampling rate (kHz) 10 Resistance R(Ω / km) <![CDATA[8.4×10 -3 ]]> Inductance L(Ω / km) <![CDATA[8.28×10 -4 ]]> Capacitance C(Ω / km) <![CDATA[1.39×10 -8 ]]> Failure time(s) 0.505 Fault location (km) 100200300400
[0036] The time domain fault location technology of the present invention mainly uses the data of instantaneous measurement of high-voltage direct current transmission line faults, brings them into the simplified uniform transmission line equation, and obtains the distribution matrix of voltage and current over time and space. Then, the fault location is determined by comparing the similarity of voltage waveforms through the two-terminal method, and the distance measurement result is optimized by using the binary search method. The specific steps are as follows:
[0037] Step 1: In this embodiment, the initial voltage and current data of the first and last ends of the fault are measured and obtained, and the current data starting from the fault start time, t d The time window lengths are: v0s i0s v0r i0r, and the first time interval Δt = 1×10 4 s, Δx = 50km, the uniform transmission line equation of this high-voltage DC transmission line can be expressed as:
[0038]
[0039] Where: u(x,t) and i(x,t) are the instantaneous voltage and instantaneous current of the high-voltage DC transmission line at any time t and any position x.
[0040] The variational iteration method (VIM) in the field of mathematics is selected to solve the above partial differential equations. First, the central finite difference method is applied to convert the time derivatives of voltage and current into the quotient of the difference between two adjacent moments and twice the time interval, and the above partial differential equations are transformed into the following differential equations:
[0041]
[0042] Using VIM to solve the above differential equations, we can get:
[0043]
[0044] In the formula, v j,k (x), i j,k (x) represents the instantaneous value of voltage and current at position x at time j after k iterations. It can be seen from the above formula that the voltage and current values after any k iterations can be obtained one by one from the initial measurement values. Considering the calculation complexity and accuracy of the calculation results in this actual situation, this scheme selects the voltage and current values after 3 iterations as the target, and the iterative calculation formula is as follows:
[0045]
[0046]
[0047] When the initial data is input, V j ,I j Take v0s i0s, x0 is taken as 0; when the end data is brought in, V j ,I j Take v0r i0r, and x0 as 500. The voltage waveform is derived from the beginning and the end as follows Figure 2 As shown in the above formula, we can see that with each additional iteration, the final result will have 2 fewer points in the time dimension. Therefore, the final result of the first distance measurement of this scheme will have 6 fewer points in the time dimension.
[0048] Step 2: Based on the above expressions, we can obtain the voltage and current distribution matrices v3s, i3s, v3r, and i3r derived from the initial end over time and position. The total number of rows in these four matrices is 10, and the total number of columns is The data in the ith column and the ith row represent the waveform change curves of the current and voltage derived from the beginning and the end at the point at the (50i)km position of the line in the time window length of 0.1*(td-6)ms.
[0049] According to the relevant knowledge of power system analysis, when a high-voltage direct current transmission line fails, the voltage to ground at the fault point is the smallest. Therefore, the voltage derived from the beginning and end of the line to the fault location is the lowest and closest. Therefore, this paper uses the Kendall tau correlation coefficient to calculate the similarity of the voltage waveforms derived from the data at the beginning and end to the same point. The point with the highest similarity is the fault location. The specific calculation formula of the Kendall tau correlation coefficient is as follows:
[0050]
[0051] Among them, j represents the moment at which the voltage is obtained after calculation, that is, the number of columns of the voltage matrices v3s and v3r, and a represents the position corresponding to the voltage obtained after calculation, that is, the number of rows of the voltage matrices v3s and v3r.
[0052] Through the Kendall tau correlation coefficient, the similarity of the waveforms of v3s and v3r corresponding to the same position on the line can be calculated, so as to determine the point xf corresponding to the preliminary fault position. The first distance measurement results are shown in Table 1 below:
[0053] Table 1 Results of the first distance measurement
[0054] Actual fault location / km xf Ranging results error 100 6 125 5% 200 9 200 0 300 13 300 0 400 16 375 5%
[0055] It can be seen from the above table that during the first calculation, the selected distance interval was too large, so the measurement result error varied greatly and could not be taken as the final result.
[0056] Step 3: After the first calculation, the first distance measurement result xf is taken as the center, the two points 50km ahead and 50km behind are taken as the endpoints, and the interval of 100km is taken as the total length as a new line. The schematic diagram of the segmentation is as follows Figure 3 At this time, Δx = 25 km, Δt = 2 × 10 -4 As Δx and Δt for the second calculation, 100 km is taken as the total length l of the line after the second iteration. It can be seen from the above that the line is divided into five points during the second distance measurement, and the position of the point with the highest transformer waveform similarity is found.
[0057] If the fault location calculated for the first time is between 100 and 400 km, then the voltage of the point before the fault location derived from the head end is taken as the initial voltage of the head end after iteration, that is, v0s=v3s(xf-1), and the voltage of the point after the fault location derived from the tail end is taken as the initial voltage of the tail end after iteration, v0r=v3r(xf+1).
[0058] If the fault location calculated for the first time is between 0 and 100 km or between 400 and 500 km of the line, then the voltage of the point before the fault location derived from the end is taken as the initial voltage of the first end after iteration, that is, v0s=v3r(xf-1), and the voltage of the point after the fault location derived from the first end is taken as the initial voltage of the end after iteration v0r=v3s(xf+1). After updating all the data, according to the second step, the distance measurement calculation is performed again, and the calculation is performed successively until the target accuracy is reached. This time, five iterations are selected to finally obtain the target result. The results of multiple calculations are shown in Table 2 below:
[0059] Table 2 Multiple iterative calculation results
[0060] Actual fault location / km Successive xf Distance measurement result / km error / % 100 63221 112.5 2.5 200 93334 201.5625 0.3125 300 123243 296.875 0.625 400 163445 387.5 2.5
[0061] The specific implementation modes of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.
Claims
1. A time domain fault location method for a high voltage direct current transmission line, characterized in that: The steps include: Step 1: Use the variational iteration method to solve the uniform transmission line equation of the high-voltage DC transmission line and derive the expressions of voltage and current; Step 2: Substitute the measurement data at the beginning and end into the voltage and current expressions respectively, calculate the voltage and current distribution matrix, use the two-terminal method, list the distance measurement equation, and calculate the fault location; Step 3: Use the binary search method to establish a new line with the first distance measurement result as the center, update its parameters, perform the above calculation again, and iterate the fault results until the accuracy requirements are met.
2. A time domain fault location method for a high voltage direct current transmission line according to claim 1, characterized in that: The specific steps of Step 1 are as follows: Step 1: Given the resistance R, inductance L, capacitance C, reactance G per unit length of the HVDC transmission line and the total length l of the line, the uniform transmission line equation of the HVDC transmission line is expressed as: Where: u(x,t) and i(x,t) are the instantaneous voltage and instantaneous current of the HVDC transmission line at any time t and any position x, respectively. The known initial conditions of the partial differential equation are: the instantaneous values of voltage and current at the beginning and end fault times: u(0,t), u(l,t), i(0,t), i(l,t). The target is the voltage and current distribution along the line at the time of the fault: u(x,t), i(x,t); Using the central finite difference method, the time derivatives of voltage and current are converted into the quotient of the difference between two adjacent moments and twice the time interval, and the above partial differential equations are transformed into the following differential equations: Using the variational iteration method to solve the above differential equations, we can obtain: In the formula, v j,k (x), i j,k (x) respectively represent the instantaneous values of voltage and current at position x at time j after k iterations. The voltage and current values after any k iterations can be calculated one by one from the initial measurement values.
3. A time domain fault location method for a high voltage direct current transmission line according to claim 1, characterized in that: The specific steps of Step 2 are as follows: Step 2.1: Substitute the initial measurement data of the start and end into Step 1 to obtain the voltage and current distribution matrix over time and position derived at the start and the voltage and current distribution matrix over time and position derived at the end; Step 2.2 Use the Kendall tau correlation coefficient to calculate the similarity of the voltage waveforms when the data at both ends are pushed to the same point. The point with the highest similarity is the rough fault location.
4. A time domain fault location method for a high voltage direct current transmission line according to claim 1, characterized in that: The specific steps of Step 3 are as follows: Iterate the fault result obtained above, take the last distance measurement result xf as the center, and take the distance between the two points at Δx as the new line length. After updating various data, perform the second calculation again until the difference between the two iterative results meets the accuracy requirement. The initial data for the first distance measurement calculation are Δx1, Δt1, and l1. After the first calculation, 0.5Δx1 and 0.5Δt1 are taken as Δx2 and Δt2 for the second calculation; 2Δx1 is taken as the total length of the line after the second iteration, l2. If the fault location calculated in the first calculation is located at The entire line is If the fault location is between the entire length of the line, then the voltage of the point before the fault location derived from the head end is taken as the initial voltage of the head end after iteration, and the voltage of the point after the fault location derived from the end end is taken as the initial voltage of the end end after iteration. or after Then take the voltage of the point before the fault position derived from the end as the initial voltage of the first end after iteration, and take the voltage of the point after the fault position derived from the first end as the initial voltage of the end after iteration, and then calculate according to the second step to get the second distance measurement result, and then calculate successively until the distance measurement results before and after xf are satisfied n With xf n+1 The error between them is less than the given value C set ,Right now:
