Background
With the rapid development of direct current protection and direct current breaker technology, the multi-terminal interconnected flexible direct current power grid is widely popularized and applied in engineering. Compared with direct current cables, the overhead line power transmission has a greater economic advantage. However, since overhead lines are exposed to air, without an insulating housing, they are subjected to complex terrain and run in severe weather conditions with a much higher probability of failure than dc cables. After the fault occurs, the rapid and accurate fault location can greatly shorten the maintenance time, thereby ensuring the safe and reliable operation of the direct current power grid. The natural main frequency of the traveling wave after the fault of the flexible direct current power grid is far lower than that of the traveling wave after the fault of the conventional direct current power transmission system.
The flexible direct current power grid protection and fault isolation action speed is far faster than that of a conventional direct current power transmission system, and is generally only a few milliseconds, so that the available data window is shortened greatly. The measurement of lower frequencies is realized by using a shorter data window, which can lead to the great reduction of the frequency measurement accuracy of the frequency measurement algorithm. In addition, the MUSIC algorithm itself has false spectral peak phenomenon, and the ranging result is greatly deviated. Therefore, the traditional fault location method based on the natural frequency has limitation in application in the flexible direct current power grid. Therefore, the research on the fault location method suitable for the flexible direct current power grid has important theoretical significance and engineering value.
The existing direct current transmission line fault location method is mainly divided into a traveling wave method, a fault analysis method and a natural frequency method in principle. The traveling wave method mainly utilizes the propagation characteristics of fault traveling waves to realize fault distance measurement, and has the outstanding advantages of high speed, high precision and the like. However, there are also inherent disadvantages such as high sampling frequency (hundreds of kHz to MHz) and weak transition resistance. The fault analysis method is used for writing equations according to the relation between the line parameters and the electrical quantity, and fault positioning is realized through optimizing and solving. The fault analysis method is theoretically independent of accurate capture of the traveling wave head, and has low requirement on sampling frequency. However, since the method utilizes the fixed line parameters to realize the algorithm design, the ranging accuracy is greatly influenced by the frequency-dependent characteristic of the line. The natural frequency method utilizes the functional relation between the natural frequency of the fault traveling wave and the fault distance to realize the accurate fault distance measurement of the power transmission line, the distance measurement method based on the natural frequency main component of the traveling wave is not limited by the identification of the traveling wave head and the like, the sampling rate is low, and the frequency-varying characteristics of the line parameters can be fully considered. The literature 'high-voltage direct-current transmission line fault location based on traveling wave natural frequency' proposes a new method for accurately locating faults of a direct-current transmission line by only utilizing single-ended voltage transient state quantity frequency spectrum information, and a fault traveling wave natural frequency main component is extracted by utilizing the transient state voltage information of a later section (10 ms) of faults through a multiple signal classification algorithm (Multiple signal classification, MUSIC). The method has good application prospect in a conventional direct current transmission system. However, the natural main frequency of the traveling wave after the fault of the flexible direct current power grid is far lower than the natural frequency main component of the traveling wave after the fault of the conventional direct current power transmission system. Meanwhile, the flexible direct current power grid protection and fault isolation action speed is far faster than that of a conventional direct current power transmission system (only a few milliseconds), so that the available data window is shortened greatly. The measurement of lower frequencies is realized by using a shorter data window, which can lead to the great reduction of the frequency measurement accuracy of the frequency measurement algorithm.
Aiming at the current situation, it is necessary to design an accurate fault location method suitable for the flexible direct current power grid so as to realize rapid and accurate fault location and realize accurate fault detection.
Disclosure of Invention
Aiming at a multi-terminal flexible direct current system which takes an overhead line as an electric energy transmission medium and is provided with a direct current breaker, the invention provides a fault distance measurement method based on a coincident direct current breaker, and the accurate fault distance measurement is realized by fully utilizing the electric quantity data of a longer time window in the reclosing period through a main/secondary frequency iterative calculation/verification type distance measurement algorithm.
The invention discloses a fault location method based on a coincident direct current breaker, which comprises the following steps of.
Step 1, if the reclosing strategy judges that the fault is a permanent fault, acquiring a residual voltage u of 20ms after the RCB is overlapped p ;
Step 2, and for the residual voltage u p Decoupling to obtain line mode voltage u L ;
Step 3, solving an energy spectrum of the line mode voltage by using a MUSIC algorithm;
step 4, obtaining the natural frequency f corresponding to the energy spectrum first peak 1 According to the natural frequency f corresponding to the energy spectrum head peak value 1 Calculating the reflection angle θ at that frequency 1 、θ 2 Propagation velocity v of traveling wave;
step 5, estimating a fault distance estimated value d';
step 6, estimating the natural frequency estimated value f by using the fault distance estimated value d 2 ' the natural secondary frequency estimation value f is searched in the energy spectrum 2 ' the closest spectral peak frequency, denoted as the natural secondary frequency f 2 ;
Step 7, calculating the natural frequency f 2 Angle of reflection theta 1 、θ 2 And the propagation speed of the traveling wave, and calculating a fault distance d;
step 8, if 1/K is satisfied<d/d′<K, the result of the two calculations is close and the spectrum peak corresponding to the selected primary frequency and the secondary frequency is not considered to be false spectrum peak, then the natural secondary frequency f is selected 2 As a final ranging calculation value; if the above conditions are not satisfied, it indicates that the natural frequency f of the first peak is extracted 1 The main component is a false frequency solution, the natural frequency corresponding to the second spectrum peak is selected as the main frequency, and the steps 5-7 are repeatedly executed; until a final ranging calculation value is obtained;
wherein K is a verification parameter of the difference between the accurate ranging result and the estimation result.
When the method is applied to a flexible direct current power grid, compared with the traditional method for realizing fault location by utilizing the inherent main frequency of the initial stage of the fault, the method for locating the fault can remarkably improve the precision of fault location and eliminate the problem of location deviation caused by the false spectral peak phenomenon of the MUSIC algorithm.
Detailed Description
The following detailed description of specific embodiments of the invention will be given with reference to the accompanying drawings.
For a multi-terminal direct current transmission system, direct current breakers are installed on two sides of an overhead line to ensure that a fault line is selectively cut off when a fault occurs. The current direct current circuit breaker suitable for the high-voltage direct current power grid comprises a hybrid direct current circuit breaker and a mechanical direct current circuit breaker. Compared with a hybrid high-voltage direct-current circuit breaker, the mechanical high-voltage direct-current circuit breaker has the advantages of high reliability, small loss, low cost, small occupied area, direct outdoor arrangement and the like. Therefore, the mechanical direct current breaker has good application prospect in a high-voltage direct current power grid. The topology of a mechanical dc breaker can be divided into three parts, namely: a through-flow branch consisting of a fast vacuum switch; a transfer branch composed of a precharge capacitor, an inductance element, and a trigger unit; an energy absorption branch consisting of lightning arresters. In addition, one side of the breaker is also connected with a residual current switch for breaking the residual current. By means of the sufficiently long electrical quantity information during reclosing of the reclosing direct current circuit breaker (residual current breaker, RCB), the invention provides a fault locating method capable of accurately extracting the natural frequency of line voltage and based on the natural frequency.
The invention discloses a fault distance measurement method based on a coincident direct current breaker, which has the following specific implementation scenes:
taking a four-terminal true bipolar direct current transmission system as an example. Fig. 1 is a schematic diagram of an equivalent model of a four-terminal dc system and a key device. Line is an overhead Line of the direct current power grid; DCCB (DCCB) 1 、DCCB 2 The two ends of the circuit are respectively provided with a mechanical direct current breaker. When the line breaks down, the circuit breakers at the two sides of the fault line act rapidly, and the fault line is cut off. After a suitable time of waiting for the release, the dc breaker RCB then performs reclosing.
Fig. 2 is a schematic diagram of an equivalent circuit of the dc system under different fault properties. After the RCBs coincide, the frequency domain two-port model of the fault loop can be equivalent to the dc system equivalent circuit shown in fig. 2.
As can be seen from fig. 2, the circuit-breaker line-side voltage U p (s) fault point voltage U f The expression of(s) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,
wherein E is 1 (s) is the Thevenin equivalent voltage source of the DC system converter station, Z 1 、Z C R is the impedance of the converter station (comprising the equivalent impedance of a converter, a current limiting reactor and a direct current breaker), the wave impedance of the line and the transition resistance respectively, W 1 (s) and W 2 (s) is a controlled voltage source reflecting the electromagnetic wave propagation process of the direct current overhead line, Γ 1 (s)、Γ 2 (s) are the reflection coefficients of the converter station side and the fault side of the overhead line, respectively.
From equation (1), the spectrum of the line voltage is defined by W 1 (s) and E 1 Spectral composition of(s), E 1 (s) is an ideal DC power supply (the frequency characteristic of which can be ignored), W 1 Characterization of the Spectrum of(s)Natural frequency of fault traveling wave. Thus W is 1 (s) a distance-frequency characteristic that characterizes the transient component of the line voltage after coincidence RCB, i.e.,
wherein, -pi/2<θ 1 +θ 2 <π/2,θ 1 、θ 2 Respectively the reflection coefficient Γ 1 (s)、Γ 2 Phase angle of(s), f n Is the frequency of the nth component of the line voltage.
Equation (3) shows that the transient frequency of the line voltage after the RCB is overlapped has a one-to-one correspondence with the fault distance, so that the fault distance measurement can be performed by utilizing the characteristic.
Fig. 3 is a flow chart of a fault location method based on a coincident direct current breaker. The method comprises the following specific steps:
step 1, if the reclosing strategy judges that the fault is a permanent fault, acquiring a residual voltage u of 20ms after the RCB is overlapped p ;
Step 2, and for the residual voltage u p Decoupling to obtain line mode voltage u L The expression is as follows:
wherein u is p + For the outlet voltage of the positive electrode line, u p - Is the negative line outlet voltage;
step 3, solving an energy spectrum of the line mode voltage by using a MUSIC algorithm;
step 4, obtaining the natural frequency f corresponding to the energy spectrum first peak 1 Obtaining a natural frequency f corresponding to the first peak value according to the energy spectrum 1 Calculating the reflection angle θ at that frequency 1 、θ 2 And propagation velocity v of the traveling wave, the expression is as follows:
wherein R is 0 、L 0 、C 0 、G 0 The resistance, inductance, capacitance and conductance parameters of the circuit under unit length are respectively;
step 5, estimating a fault distance value d' (according to (3));
step 6, estimating the natural frequency f by using the fault distance estimation value d 2 ' the natural secondary frequency estimation value f is searched in the energy spectrum 2 ' nearest spectral peak frequency f 2 ;
Step 7, calculating the natural frequency f 2 Angle of reflection theta 1 、θ 2 And the propagation velocity of the traveling wave, and calculates a fault distance d (according to formula (3));
step 8, if 1/K is satisfied<d/d′<K (K is used for checking whether the difference between the accurate ranging result and the estimation result is within a certain range, and a value slightly larger than 1 is generally adopted, such as 1.2), which indicates that the two calculation results are close and that the spectrum peaks corresponding to the selected primary frequency and secondary frequency are not false spectrum peaks, and then the inherent secondary frequency f is selected 2 As a final ranging calculation value;
if the above conditions are not satisfied, it indicates that the natural frequency f of the first peak is extracted 1 The principal component is false frequency solution, and the natural frequency corresponding to the second spectral peak is selected as the principal frequency f 1 Repeatedly executing the steps 5-7; until a final ranging calculation is obtained.
The main advantages of the above method include:
(1) the inherent main frequency of the fault traveling wave of the flexible direct current power grid is too low, and the ranging accuracy is difficult to ensure. The method utilizes the main frequency to carry out fault distance estimation, thereby obtaining the secondary frequency with a larger frequency value to carry out fault distance measurement. According to the formula (1), the ranging error caused by the same-sized ranging error in the high frequency band is obviously smaller than that in the low frequency band, so that the ranging accuracy can be obviously improved by the method.
(2) And the primary frequency ranging result (estimation) and the secondary frequency ranging result are used for mutual verification, and the ranging result is considered to be correct only when the difference of the primary frequency ranging result and the secondary frequency ranging result is within a certain range. Obviously, if one of the primary frequency used for estimating the distance and the secondary frequency used for calculating the distance is the frequency corresponding to the false spectral peak (namely, the false solution), or both are frequencies corresponding to the false spectral peak, the probability that the obtained estimation result d' and the calculation result d are less different is very low. Therefore, the method can effectively avoid the problem of error ranging results caused by false spectral peaks.