Extra-high voltage hybrid line fault recovery method based on direct current controllability

Abstract

The invention discloses an extra-high voltage hybrid line fault recovery method based on direct current controllability, and the method comprises the steps: firstly, carrying out the signal injection through taking a current as a controlled variable after the line is deionization, and fundamentally avoiding an overcurrent problem possibly caused by voltage injection; and secondly, analyzing the control response difference of the hybrid MMC to the direct current based on system equivalent circuits under different fault properties, thereby forming a fault recovery strategy based on the controllability of the direct current. And finally, the validity of the provided strategy is verified through simulation. The result shows that the provided strategy can accurately and reliably identify the fault property under each fault working condition, communication is not needed, and the power supply reliability of the system is improved.

Description

Technical Field

[0001] The invention relates to the technical field of power lines, and in particular to a fault recovery method for an ultra-high voltage hybrid line based on direct current controllability. Background Art

[0002] High-voltage direct current transmission is an important means to solve the problem of large-capacity and long-distance power transmission, and plays an important role in the optimal allocation of energy. Among them, the hybrid direct current transmission system that combines the advantages of conventional direct current transmission (line commutated converter-high voltage direct current, LCC-HVDC) and flexible direct current transmission (modular multilevel converter-high voltage direct current, MMC-HVDC) has broad development prospects in the field of direct current transmission. From the current technical level, considering the manufacturing cost of DC circuit breakers and valve group insulation and other factors, it is advisable to adopt a single-pole double-valve group wiring method for ultra-high voltage hybrid direct current transmission systems (ultra-high voltage direct current, UHVDC), and use a sub-module with DC fault self-clearing capability on the inverter side to realize DC fault ride-through and valve group online investment and withdrawal functions. At present, some UHV hybrid DC transmission projects with DC fault ride-through capability have been built (such as the "Kunliulong" ±800kV UHV hybrid DC project) or are under construction (such as the Kubuqi North-Cangzhou ±800kV UHV hybrid DC project). These projects all use full / half-bridge hybrid MMCs with negative voltage output capability on the inverter side.

[0003] In the process of long-distance and large-capacity power transmission, the rapid recovery of DC line faults is crucial to ensure equipment safety, improve power supply reliability, and maintain the stability of the entire system. In actual projects, hybrid DC transmission systems have adopted the fault recovery strategies of traditional high-voltage DC transmission systems and flexible DC transmission systems. For the LCC side, fault clearing and recovery are achieved through the DC line fault restart function (DC line Fault Recovery Sequence, DFRS); for the MMC side with fault self-clearing capability, the fault recovery method is to unlock the converter again after the locked converter clears the fault current and undergoes the de-ionization process, and determine whether the fault disappears by detecting whether the DC voltage can be established normally. However, the above recovery strategies lack the pre-judgment of the nature of the fault. Restarting a permanent fault will cause a more serious secondary impact on the system, which is not conducive to the safe and stable operation of the system. In addition, in order to accurately distinguish between permanent / transient faults with large transition resistance, a sufficiently large voltage detection threshold must be set, which means that a longer detection (recovery) time is required. For example, the restart detection time of the Yongfu LCC-HVDC project is set to 75ms.

[0004] To solve this problem, many scholars have proposed methods for distinguishing the nature of faults in DC transmission lines. For the UHV hybrid DC transmission system using hybrid MMC on the inverter side, some experts pointed out that a sinusoidal current signal can be injected into the line through LCC, and the difference in the similarity of the currents at both ends of the line under different fault properties can be obtained based on Kirchhoff's current law, and a dynamic time warping algorithm is introduced to measure this difference. It has strong resistance to transition resistance and anti-interference capabilities, but this method requires the transmission of a large amount of current data, and the communication system is burdened with a large amount of data. In order to avoid the use of communication, some experts proposed that the hybrid MMC can be used to output low-frequency characteristic signals to the line, and the calculated voltage at the opposite end can be obtained by establishing a time domain distributed parameter model of the line, and the voltage at the local end can be compared to determine the nature of the fault. However, this method is not easy to distinguish between transient / permanent faults at the converter outlet, and the calculation accuracy is highly dependent on the accurate acquisition of line parameters. Other scholars have solved the fault distance by establishing a fault loop equation containing fault distance and fault impedance, and distinguished the nature of the fault according to the polarity of the fault distance. In theory, for port faults where the fault point is located at the head end of the line, the fault distance is zero, and the reliability of this method needs to be verified.

[0005] To address the above problems, from the perspective of control and protection coordination, a fault recovery strategy based on current injection is proposed with the goal of improving the success rate of system restart. Summary of the invention

[0006] The object of the present invention is to provide a UHV hybrid line fault recovery method based on direct current controllability, thereby meeting the requirements of the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The method for fault recovery of a UHV hybrid line based on direct current controllability comprises the following steps:

[0009] Step 1: The LCC station performs phase shifting after receiving the line protection action signal. After the phase shifting process is completed, the voltage u at the line head end protection installation is collected. dcR Data Substitution dcR (t)>u Δ (10) Check whether the voltage is rebuilt, u Δ The hybrid MMC station switches to the DC current suppression mode after receiving the line protection action signal, and sets the counter n=1, which means that the first injection is about to start;

[0010] Step 2: After 300ms of deionization, the hybrid MMC switches to the current injection mode and attempts to inject a step current signal with an amplitude of 0.5pu into the line;

[0011] Step 3: Collect the m output of PI in the period from 2τ to 2τ+T after injection dc Substitute the data into (9) to determine the nature of the fault. dc Failure to meet the formula within 3 consecutive sampling points In the formula, the DC modulation ratio m dc It is obtained from the output of the PI link in the control system, T is the data window length of the identification criterion, T = 5ms; τ = l / v, l is the line length, v is the speed of the traveling wave propagation, which is the speed of light; m Δ As the judgment threshold, take m Δ =0.85; and n <n max , n max If n is the maximum number of injections, then n=n+1, the hybrid MMC switches to the current suppression mode and jumps to step 2. dc Failure to satisfy equation (9) within three consecutive sampling points, and n = n max , then the hybrid MMC is locked, if m dc When there are three consecutive sampling points that all satisfy equation (9), the hybrid MMC immediately switches back to the original DC voltage control;

[0012] Step 4: When performing steps 2 to 3, determine u in real time dcR Whether it satisfies equation (10), once it satisfies equation (10), LCC immediately switches back to the original DC current control and restores the system current. max If the equation (10) cannot be satisfied within the period of 100 ms, the LCC is locked.

[0013] As a further technical solution of the present invention: based on a UHV hybrid DC transmission system, the UHV hybrid DC transmission system adopts a double valve group structure for each pole, the parameters of the two poles are symmetrical, LCC commutation is adopted on the rectifier side, that is, double 12-pulse series connection, and full / half-bridge sub-module hybrid MMC commutation is adopted on the inverter side. When the system operates normally, the LCC station adopts fixed DC current control, the MMC station adopts fixed DC voltage / reactive power control, and the voltage and current at the line head end protection installation are respectively recorded as u dcR 、i dcR , the voltage and current at the end-of-line protection installation are denoted as u dcI 、i dcI .

[0014] As a further technical solution of the present invention: for the LCC station, when a DC line fault occurs, the DC line protection should be quickly activated. For the LCC station, after the pole control system receives the line protection action signal, the thyristor trigger angle is urgently shifted to 120°, so that the LCC is switched from the rectification state to the inverter operation, so that the fault energy in the DC system is fed back to the AC power grid, reducing the current i flowing into the fault point of the LCC dcR , when i dcR When it decreases to near zero, the firing angle is further phase-shifted to above 150°.

[0015] As a further technical solution of the present invention: for the LCC station hybrid MMC station, active control of the fault current is achieved by switching the control mode. Thanks to the negative voltage output capability of the full-bridge submodule, the hybrid MMC DC voltage has a certain adjustment margin. On the one hand, during the DC fault ride-through, the DC voltage is adjusted by using the adjustment margin to suppress the fault current output to the short-circuit point. On the other hand, during the DC fault recovery, the DC line is injected with a small-amplitude characteristic signal by using the adjustment margin to assist in determining whether the fault point has disappeared.

[0016] As a further technical solution of the present invention: it also includes analysis of electrical quantity response characteristics under different fault properties, specifically including electrical quantity analysis of permanent faults and analysis of transient fault current characteristics.

[0017] As a further technical solution of the present invention: the electrical quantity analysis of the permanent fault includes current characteristic analysis and voltage characteristic analysis.

[0018] As a further technical solution of the present invention: the instantaneous fault current characteristic analysis includes current characteristic analysis and voltage characteristic analysis.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] In a hybrid DC transmission system that uses overhead lines as power transmission channels, DC line faults are inevitable, and a safe, fast, and reliable fault recovery strategy needs to be configured. This paper addresses the problem that the existing recovery strategy lacks a fault nature identification link, makes full use of the control characteristics of the hybrid MMC, and aims to improve the system restart success rate and reduce the secondary impact of the system. After the line is de-ionized, a step current is injected into the line to reflect the dynamic characteristics of the system under different fault properties, and then a DC line fault recovery strategy based on DC current controllability is proposed. The main contributions are as follows:

[0021] 1) The current is selected as the controlled quantity, which effectively prevents the overcurrent shock caused by the injection of characteristic signals when the converter output fails.

[0022] 2) Based on the system equivalent circuits under different fault properties, the response characteristics of electrical quantities after signal injection are analyzed: for transient faults, the current on the inverter side is uncontrollable and the voltage on the rectifier side is the system voltage; for permanent faults, the current on the inverter side is controllable and the voltage on the rectifier side is proportional to the transition resistance.

[0023] 3) Taking the control signal output by PI as the characteristic quantity, a fault property identification criterion based on the DC voltage modulation ratio is constructed on the inverter side.

[0024] 4) A large number of simulation results show that the proposed fault recovery strategy can adapt to various fault conditions occurring on the entire line, with a transient resistance of up to 500, can withstand noise with a signal-to-noise ratio of 30dB, and is insensitive to changes in system parameters, with good universality. In addition, the fault recovery strategy does not rely on the communication system, and has low requirements on the sampling rate of the protection device, making it easy to implement engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a topological diagram of the UHV hybrid DC transmission system.

[0026] Figure 2 It is a diagram of the hybrid MMC fault ride-through and current injection control strategy.

[0027] Figure 3 It is the system equivalent circuit diagram under permanent fault.

[0028] Figure 4 It is the control system block diagram.

[0029] Figure 5 Is a permanent fault dcI Waveform graph.

[0030] Figure 6 Is a permanent fault dcR Waveform graph.

[0031] Figure 7It is the equivalent circuit diagram of the system under transient faults.

[0032] Figure 8 It is the i dcI waveform diagram under transient faults.

[0033] Fig. 9 It is the equivalent circuit diagram when t0 < t < t0 + 2τ under transient faults.

[0034] Fig.10 It is the u dcR waveform diagram under transient faults.

[0035] Fig.11 It is the system recovery flow chart.

[0036] Fig.12 It is the m dc waveform diagram when transient faults occur at three locations.

[0037] Fig.13 It is the u dcR waveform diagram when transient faults occur at three locations.

[0038] Fig.14 It is the i dcR waveform diagram when transient faults occur at three locations.

[0039] Fig.15 It is the m dc waveform diagram of permanent faults.

[0040] Fig.16 It is the u dcR waveform diagram of permanent faults.

[0041] Fig.17 It is the frequency-dependent parameter and tower model diagram of the DC overhead line. Specific implementation manners

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] Such as Figure 1-17As shown in the figure: The fault recovery method of UHV hybrid line based on the controllability of DC current is proposed. First, the control response of LCC and hybrid MMC stations after the DC line protection is activated is analyzed, and the injection method of the current signal on the hybrid MMC side is given; secondly, based on the system circuit model and the control system model, the controllability of DC current after injection control switching is studied, and the difference in the response characteristics of voltage and current under different fault properties is deeply analyzed. Based on this difference, the fault property identification criterion is constructed, and the system recovery strategy is further formed; finally, the effectiveness of the proposed fault recovery strategy is verified by simulation in PSCAD.

[0044] Figure 1 The figure shows the topology of the UHV hybrid DC transmission system. The system adopts a dual valve group structure for each pole, with symmetrical parameters on both poles. LCC commutation (dual 12-pulse series connection) is used on the rectifier side, and full / half-bridge submodule hybrid MMC commutation is used on the inverter side. When the system is operating normally, the LCC station adopts constant DC current control, and the MMC station adopts constant DC voltage / reactive power control. In the figure, f1, f2, and f3 represent the fault points at the beginning, middle, and end of the line respectively (all are faults within the DC line area, and the nature of the fault needs to be determined). dcR 、i dcR are the voltage and current at the protection installation point at the head end of the line; u dcI 、i dcI They are the voltage and current at the end of the line protection installation. The positive direction of the measured current is specified to be from the busbar to the line.

[0045] The control response of the converter after the DC line protection is activated:

[0046] Phase shift control of LCC station: When a DC line fault occurs, the DC line protection should be activated quickly. For LCC station, after the pole control system receives the line protection action signal, the thyristor trigger angle is urgently shifted to 120°. The LCC is switched from rectification state to inverter operation, so that the fault energy in the DC system is fed back to the AC grid, reducing the current i flowing into the fault point of the LCC. dcR . when i dcR When it decreases to near zero, the firing angle is further phase-shifted to above 150°.

[0047] Hybrid MMC station DC current control:

[0048] For hybrid MMC stations, active control of fault current can be achieved by switching control modes. Thanks to the negative voltage output capability of the full-bridge submodule, the hybrid MMC DC voltage has a certain adjustment margin. On the one hand, during DC fault ride-through, adjusting the DC voltage using this adjustment margin can suppress the fault current output to the short-circuit point. On the other hand, during DC fault recovery, this adjustment margin can be used to inject a small-amplitude characteristic signal into the DC line to assist in determining whether the fault point has disappeared and avoid blind restart during fault recovery.

[0049] Some literature uses a step voltage signal with an amplitude of 0.1pu as the injection signal. According to the circuit law, when the voltage is determined, the fault current is inversely proportional to the impedance. Since the hybrid MMC has the ability to control fault current, in order to reduce investment costs and improve the dynamic response of the system, the inductance of the current limiting reactor configured at its DC outlet is generally small. Considering extreme cases, when a metallic bipolar short-circuit fault occurs at the hybrid MMC outlet, it is easy to cause overcurrent in the converter valve. Therefore, this paper selects the current signal as the injection signal to assist the protection in fault nature identification.

[0050] The hybrid MMC control strategy with current signal injection control is as follows: Figure 2 shown.

[0051] Figure 2 All quantities in the table are per unit values, among which U dcref is the DC voltage reference value. When the system is operating normally, U dcref =1.0pu;i dcIpu and i dcref are respectively the measured DC current and the DC current command value of the inverter station; u diffj and u cirj are the j-phase differential mode voltage reference value and the circulating current voltage reference value respectively; u p(n)jref is the voltage reference value of the upper (lower) bridge arm of phase j. The DC current control is implemented by PI controller, and the output of PI is m dc is the DC current modulation ratio, and its expression is shown in formula (1):

[18] :

[0052]

[0053] In the formula, u MMC It is the voltage of the upper and lower bridge arms of the mixed MMC and has a wide adjustable range; U N is the rated DC voltage.

[0054] Depend on Figure 2 It can be seen that the switching coordination of S1 and S2 enables the DC control loop to present three optional control modes. The following are the three control modes:

[0055] Mode I: DC voltage control mode. In this mode, switches S1 and S2 are both set to 0.

[0056] Mode II: DC current suppression mode. After a DC line fault occurs, the line protection will be activated in milliseconds, and the control switch S1 will switch from 0 to 1 (S2 is still set to 0), and the hybrid MMC will enter the DC current control mode. In this mode, the hybrid MMC DC voltage will adapt to the DC current change until the DC current reaches the preset value 0p.u.

[0057] Mode III: DC current injection mode. When the fault current drops to 0, after a certain deionization time, the control switch S2 is switched from 0 to 1 (S2 is still set to 1), and the hybrid MMC attempts to inject a step signal with an amplitude from 0p.u. to 0.5pu into the DC line. Generally, the maximum overcurrent capacity in the MMC is twice the rated current, so the selection of a current signal with an amplitude of 0.5pu can effectively reduce the overcurrent risk of the MMC.

[0058] The hybrid MMC can realize AC / DC decoupling control, and the switching between the three control modes in the DC control loop does not affect the AC control loop. In addition, the protection action strategy does not lock the hybrid MMC, so during the identification of the fault nature of the DC line, the hybrid MMC can continuously provide reactive power support for the receiving AC power grid.

[0059] Analysis of electrical quantity response characteristics under different fault properties: The measured current and voltage response characteristics should be determined by the control equation and the circuit equation. Obviously, the circuit structure under permanent faults is different from that under transient faults (the essence lies in whether there is a fault branch). Figure 2 When the injection control is shown, the electrical quantity response characteristics will inevitably be different.

[0060] Electrical quantity analysis of permanent faults:

[0061] Current characteristic analysis:

[0062] From the above analysis, it can be seen that LCC is before the hybrid MMC switches to the injection mode. dcR =0, so the rectifier side can be treated as an open circuit in the equivalent circuit diagram, and then the following is obtained: Figure 3 The equivalent circuit of the system under permanent fault is shown in the figure. In the figure, f represents the fault point, R f is the transition resistance of the fault point; terminals R and I represent the beginning and end of the line respectively; α is the ratio of the distance from R to the fault point to the total length of the line; R l and L l is the equivalent resistance and inductance per unit length of the line; L p and L x They are the inductance of the smoothing reactor of the LCC station and the inductance of the current limiting reactor of the hybrid MMC station; LMMC =2L arm / 3, R MMC =2R arm / 3,L arm and R arm are the hybrid MMC bridge arm inductance and its equivalent resistance respectively.

[0063] According to formula (1) and Figure 3 The circuit shown in the figure can obtain the DC current (per unit value) under permanent fault as shown in formula (1):

[0064]

[0065] In the formula, I N is the rated current; R eq =R MMC +(1-α)R x , L eq =L MMC +(1-α)L x ; R eqpu =R eq / Z N , L eqpu =L eq / Z N , Z N =U N / I N is the impedance reference value.

[0066] Combination Figure 2 The block diagram of the DC current control loop can be drawn by equation (2): Figure 4 shown.

[0067] Depend on Figure 4 It can be seen that by reasonably setting the PI controller parameters, closed-loop control of the DC current can be achieved, making the actual current i dcIpu Effective tracking dcref The control equation for DC current is (3):

[0068]

[0069] In the formula, k p and k i are the proportional coefficient and integral coefficient of the PI controller respectively. In this paper, k p =5, k i =10 3 .

[0070] Combining equations (2) and (3) eliminates M dc The closed-loop transfer function of current injection control under permanent fault is shown in formula (4):

[0071]

[0072] Considering that the step signal has a stable DC component and energy distribution under steady-state conditions, the step current signal is selected as the characteristic signal for injection in Section 2.2. 0- ), there is dcIpu (t 0- )=i dcref (t 0- )=0p.u., after injection control switching (denoted as t 0+ ), there is dcref (t 0+ )=0.5pu. Therefore, in formula (4), i dcref (s) = 0.5 / s. In summary, the DC current response under current injection control under permanent fault is a second-order system step response. By performing an inverse Laplace transform on equation (4), the time domain expression of the DC current can be obtained as shown in equation (5):

[0073]

[0074] Where t0 represents the injection control switching time; λ1 and λ2 are the characteristic roots of the closed-loop transfer function shown in equation (4); A1~A3 are constant coefficients.

[0075] Figure 5 The permanent fault time when different transition resistances occur at the line midpoint is given. dcI The simulation waveform of Figure 5 It can be seen that no matter what the transition resistance is, i dcI Always be able to quickly and effectively track the instruction value. It should be noted that Figure 5 The waveform of the medium current injection has a slight oscillation at the beginning, which is the high-frequency component caused by the folding and reflection of the traveling wave at the line port and the fault point. Since the propagation process is accompanied by the rapid attenuation of the traveling wave, in order to simplify the analysis Figure 3 The circuit diagram shown does not take into account the influence of wave processes. Figure 5 It can be seen from the observation that the transmission and attenuation of traveling waves do not affect i dcI Effective tracking of instruction values.

[0076] Voltage characteristic analysis:

[0077] according to Figure 3 It can be seen that because dcR =0, so u dcR Equal to the fault point voltage u f . Because I dcIpu Able to quickly track step signals dcref , so u dcR It can be obtained from formula (6):

[0078] udcR = u f = i dcref R f (6)

[0079] Figure 6 gives the simulation waveforms of u when a permanent fault with different transition resistances occurs at the midpoint of the line dcR . Observation Figure 6 shows that, due to the same amplitude of the injected current signal, the DC voltage is proportional to the transition resistance. Considering the maximum transition resistance of 500 Ω

[19] , according to Equation (6), u dcR = (1 × 2 × 500) kV = 500 kV, which is consistent with the blue line in the figure.

[0080] Analysis of the characteristics of transient fault current:

[0081] Analysis of current characteristics: As Figure 7 shown is the equivalent circuit of the system under transient fault.

[0082] In Figure 7 the circuit shown, i dcR = i dcI is always 0 and does not change with the change of the DC current command before and after t0. A transient fault means that in Equation (2), R eq → ∞, Figure 4 the denominator part of the red block diagram representing the actual circuit model in dcIpu is infinite, the original closed-loop control structure becomes open-loop, and i dcref cannot effectively track i

[0083] Figure 8 gives the simulation waveforms of i dcI under transient fault. Observation Figure 8 shows that i dcI eventually tends to 0, but it is not 0 but experiences a certain oscillation in the initial stage after the injection control switch. Also because Figure 7 the wave process is not considered. In fact, according to the principle of traveling wave protection of the line, if τ represents the time for the traveling wave to propagate along the entire length of the line, then in the time scale of t0 < t < t0 + 2τ (the first traveling wave), the line model is equivalent to the wave impedance, as Fig. 9 shown

[15] . In the figure, Z c represents the line wave impedance, and its expression is as shown in Equation (7):

[0084]

[0085] where R, L, C, and G are the resistance, inductance, conductance, and capacitance of the DC line per unit length respectively.

[0086] Due to the very small line resistance and conductance, the wave impedance Z in the traveling wave stage is usually regarded as a resistance, and its resistance value is c Therefore, in the initial stage after the injection control is switched, Equation (2) can be rewritten as: Therefore, in the initial stage after the injection control is switched, Equation (2) can be rewritten as:

[0087]

[0088] Combined with Figure 2 and Equation (8), it can be seen that at this time Figure 4 still remains in the closed-loop state, Figure 7 in which i within t0 < t < t0 + 2τ dcI towards i dcref The tracking process constitutes the traveling wave injected by the MMC into the DC line. Then, the injected traveling wave reflects back and forth at both ends of the line, and the traveling wave propagation process is accompanied by attenuation, which is reflected as an amplitude-reducing oscillation of the waveform in the time domain

[20] .

[0089] Analysis of voltage characteristics:

[0090] From Figure 7 it can be seen that after the faulty branch disappears, u dcR = u dcI . According to Figure 2 , u dcI is determined by the output m dc of the PI controller and is obtained from the current deviation i dcref -i dcIpu . Through the previous analysis, it can be seen that the physical model determines that the current deviation cannot be eliminated, and i dcref will be greater than i dcIpu , resulting in the forced saturation of the PI controller, and its output is clamped at the upper limit value. Considering that under instantaneous faults, the hybrid MMC needs to re-establish the voltage to restore the system, therefore, the upper limit of the PI output is set to 1 p.u., corresponding to m dc = 1, u dcR = u dcI = u MMC is equal to the normal operating voltage. Fig.10 gives the simulation waveforms of u dcR and u dcI under instantaneous faults. Observing Fig.10 it can also be seen that under instantaneous faults, u dcI tends to 800 kV.

[0091] Fault nature identification criterion and fault recovery strategy: By analyzing the controllability of the current signal, it is found that for instantaneous faults, the control of the hybrid MMC over i dcI is in a closed loop, i dcI is controllable, and u dcRProportional to the transition resistance. For permanent faults, the hybrid MMC dcI The control is open loop, i dcI Unable to reach the preset command value, u dcR Equal to the DC operating voltage. Given the asymmetry of the converters on both sides of the hybrid DC system line, corresponding criteria can be set on the rectifier side and the inverter side respectively. The advantage is that system recovery does not rely on communication. When a transient fault occurs and the communication link is damaged, power supply can still be effectively restored, which improves the reliability of fault adaptive recovery.

[0092] Hybrid MMC station fault nature identification criteria:

[0093] From the previous analysis, we can see that under permanent faults, dcIpu Always fast track i dcref , and the initial switching moment (t0 <t<t0+2τ),i dcIpu Able to track dcref , decreases after t>t0+2τ, and tends to the zero axis to oscillate with reduced amplitude. Therefore, the hybrid MMC station fault property identification criterion based on DC modulation ratio can be constructed as shown in formula (9):

[0094]

[0095] In the formula, the DC modulation ratio m dc It can be obtained from the output of the PI link in the control system. T is the data window length of the identification criterion. Considering the rapidity of control adjustment, T=5ms; τ=l / v, l is the line length, v is the speed of traveling wave propagation, which is approximately the speed of light; m Δ is the judgment threshold. Considering the control system tracking error, wave refraction reflection and other factors, m is taken Δ =0.85. At the same time, multi-point continuous judgment is adopted to ensure reliable identification of the fault nature, that is, during [t0+2τ, t0+2τ+T], if m dc When there are three consecutive sampling points that all satisfy equation (9), it is considered that the fault has disappeared and the hybrid MMC immediately switches back to the original DC voltage control. Otherwise, it is considered that the fault point has not disappeared and the next injection or direct converter blocking is prepared.

[0096] LCC station fault nature identification criteria:

[0097] From the previous analysis, it can be seen that the current command i should be set reasonably dcref It helps LCC distinguish the two fault properties. For the system in this paper, the normal operating voltage level is 800kV, which means that under transient faults, u dcR Tends to 800kV. Considering the maximum transition resistance is 500Ω, it means that under permanent fault, u dcR tends to (500×idcref )kV. Therefore, if we take i dcref =1kA (i.e. 0.5pu), the threshold u can be set Δ =(800+500×1) / 2kV=650kV. At the LCC station, the fault is judged whether the voltage can be established to determine whether the fault disappears, as shown in formula (10):

[0098] u dcR (t)>u Δ (10)

[0099] Formula (10) can be used to judge after the LCC phase shift process is completed: when u dcR When three consecutive sampling points satisfy equation (10), the fault is considered to have disappeared, and the LCC immediately switches back to the original DC current control. Otherwise, the fault point is considered to have not disappeared, and the converter needs to be directly locked after waiting for the next injection or exceeding a certain time limit.

[0100] System Failure Recovery Strategy:

[0101] System recovery strategy such as Fig.11 As shown, the specific steps include:

[0102] Step 1: The LCC station performs phase shifting after receiving the line protection action signal. After the phase shifting process is completed, u dcR Substitute the data into equation (10) to detect whether the voltage is rebuilt. After receiving the line protection action signal, the hybrid MMC station switches to the DC current suppression mode (mode II). At the same time, set the counter n=1, indicating that the first injection is about to start.

[0103] Step 2: After 300ms of deionization, the hybrid MMC switches to the current injection mode (mode III) and attempts to inject a step current signal with an amplitude of 0.5pu into the line.

[0104] Step 3: Collect the m output of PI in the period from 2τ to 2τ+T after injection dc Substitute the data into (9) to determine the nature of the fault. dc Failure to satisfy formula (9) within three consecutive sampling points, and n <n max (n max is the maximum injection number), then let n=n+1, the hybrid MMC switches to the current suppression mode (mode II), and jumps to step 2. dc Failure to satisfy equation (9) within three consecutive sampling points, and n = n max , then the hybrid MMC is locked. dc When there are three consecutive sampling points that all satisfy equation (9), the hybrid MMC immediately switches back to the original DC voltage control.

[0105] Step 4: When performing Steps 2 to 3, continuously judge whether u dcR satisfies Equation (10). Once Equation (10) is satisfied, the LCC immediately switches back to the original DC current control to restore the system current. If Equation (10) cannot be satisfied within the time period of (n max +1)×300 ms after the protection action, the LCC is blocked.

[0106] Case study implementation: To verify the effectiveness of the proposed fault recovery strategy, a hybrid HVDC system model as shown Figure 1 is built in PSCAD. The main system parameters are shown in Appendix Table A1. The hybrid MMC adopts the nearest level approximation modulation. The DC line adopts the frequency-dependent model, and the tower structure is shown in Appendix Figure A1. The simulation step size is 50 μs, and the sampling frequency is 10 kHz. Before the fault, the system operates under the rated conditions, and all faults occur at t = 1 s. The duration of all transient faults is 150 ms. Considering the fast-acting requirement of the DC line protection, it is assumed that the protections on both sides of the line trip at t = 1.002 s after the fault occurs. At t = t0 = 1.302 s, the hybrid MMC switches from Mode II to Mode III and starts injecting current into the DC line.

[0107] Transient fault: Taking the solid fault as an example, Figure 12-14 For Figure 1 the permanent positive pole grounding fault occurring at f3 in

[0108] In Fig.12 only the m dc waveform of the faulty pole is given. Since the line protection of the non-faulty pole does not operate, m dc is constantly 1. The sudden increase of m 0+ at time t dc in the figure is caused by the excessive current deviation due to sampling and control delays. After that, within the time period of t0 < t < t0 + 2τ, under the closed-loop control of the current, m dc gradually decreases. When t = t0 + 2τ = 1.306 s, the collected m dc is substituted into Equation (9) for identification. After t > 1.306 s, the hybrid MMC begins to lose control of the DC current, and the current approaches the zero axis. However, due to the existence of distributed capacitance, the measured current shows a damped oscillation trend, and m dc oscillates upward, and the PI tends to saturate. All three fault cases first satisfy Equation (9) at t = 1.3076 s and continuously satisfy Equation (9) within the next 2 sampling points. According to the protection judgment logic, the criterion shown in Equation (9) can accurately and reliably identify transient faults. The hybrid MMC switches back to the DC voltage control at t = 1.3078 s. After t > 1.3078 s, mdc =1.

[0109] Fig.13 It can be seen that when t = 1.312s, u dcR Formula (10) is satisfied for the first time and continues to be satisfied for the next two sampling points. According to the protection judgment logic, the criterion shown in Formula (10) can accurately and reliably identify instantaneous faults, and the LCC switches back to DC current control at t = 1.3122s. After that, under the control, the DC current gradually recovers to the rated value before the fault. Fig.13 and Fig.14 It can be seen that when t>1.5s, the system can be completely restored to the state before the fault.

[0110] Permanent failure: Taking metal properties as an example, Fig.15 and Fig.16 against Figure 1 The corresponding simulation waveforms are given for the permanent bipolar grounding faults occurring at f1 to f3. Since the electrical quantities are symmetrical under bipolar faults, only the positive pole waveform is given here for the convenience of observation and analysis.

[0111] exist Fig.15 In the three fault locations, m dc The waveform shows obvious segmented characteristics after t0, where the duration of the first segment is τ, τ / 2, and 0 (ignoring t 0+ The spike pulse at the moment of the fault is exactly the time when the first wave injected at each fault location exists, which verifies the correctness of the analysis in Section 3. When t = t0 + 2τ = 1.306s, the collected m dc Substitute into equation (9) for identification. At t = t0 + 2τ + T = 1.309s, m under three fault locations dc Formula (9) is never satisfied. According to the protection judgment logic, the criterion shown in Formula (9) can accurately and reliably identify permanent faults. The hybrid MMC switches back to DC current suppression control (mode II) at t = 1.309s and waits for the next injection or lockout according to the preset maximum number of injections.

[0112] Fig.16 It can be seen that u under the three fault locations dcR The waveforms are similar. Due to the short circuit effect at the fault point, the hybrid MMC cannot rebuild the voltage, u dcR If the voltage cannot be sensed, equation (10) cannot be satisfied. According to the protection judgment logic, the criterion shown in equation (10) can accurately and reliably identify permanent faults, and the LCC will not restart reliably.

[0113] Adaptability of the criterion under different fault conditions: In order to comprehensively examine the adaptability of the proposed fault recovery strategy to fault type, fault distance and transition resistance, different fault conditions are set for simulation. The specific simulation results are shown in Tables 1 and 2. Among them, Table 1 simulates transient faults, m dc and u dcR is the third consecutive sampling value greater than the threshold in the data window; t MMC and t LCC They represent the time when the hybrid MMC station and LCC station switch back to the normal operation control strategy. Table 2 Simulation for permanent faults, m dc and u dcR Take the maximum value in the data window.

[0114] Table 1. Simulation results under different transient fault conditions:

[0115]

[0116]

[0117] Table 2. Simulation results under different permanent fault conditions:

[0118] Fault type α Transition resistance <![CDATA[m dc ]]> <![CDATA[u dcR ]]> Identification results Positive ground 0％ 500Ω 0.38 0kV Permanent Bipolar short circuit 20％ 400Ω 0.38 163kV Permanent Negative ground 40％ 300Ω 0.53 235kV Permanent Positive ground 60％ 200Ω 0.40 174kV Permanent Bipolar short circuit 80％ 100Ω 0.14 49kV Permanent Negative ground 100％ 0Ω 0.01 1kV Permanent

[0119] As can be seen from Table 1, for instantaneous faults, since the system equivalent circuit is the same after the fault disappears, the m dc 、u dcR are similar, and are not greatly affected by the fault conditions, and are close to 1. dc can satisfy formula (9) within three consecutive points, u dcR All of them can satisfy formula (10) within three consecutive points. It can be seen that the criteria shown in formula (9) and formula (10) can accurately and reliably identify transient faults. The converter quickly switches back to the normal operation control strategy, and the system can quickly restore power supply under the regulation of the converter control action.

[0120] It can be seen from Table 1 that for permanent faults, under the same fault type, as the transition resistance increases and the fault distance decreases, u dcR According to the test results, the m measured under all permanent fault conditions is dc The maximum value is 0.53, and there is always m dc <m Δ ;u dcR The maximum value is 235kV, and there is always u dcR Δ It can be seen that the criteria shown in equations (9) and (10) can accurately and reliably identify permanent faults and the system cannot be restored reliably. ​

[0121] In summary, the proposed fault recovery strategy can adapt to different fault conditions.

[0122] Influence of noise: UHV DC transmission system has high voltage level and large electromagnetic interference, and noise influence is inevitable in the field sampling process. In order to investigate the effectiveness of fault recovery strategy under noise, the case where the sampling device is seriously interfered by noise is considered: Gaussian noise with a signal-to-noise ratio of 30dB is superimposed on the simulation data collected at the measuring point to simulate the actual sampling process. By traversing different fault conditions, Tables 3 and 4 give the simulation results with 30dB noise.

[0123] Table 3. Transient fault simulation results under 30dB noise:

[0124] Fault type α Transition resistance <![CDATA[m dc ]]> <![CDATA[u dcR ]]> Identification results <![CDATA[t MMC / s]]> <![CDATA[t LCC / s]]> Positive ground 0％ 500Ω 0.96 675kV Instantaneity 1.3078s 1.3132s Bipolar short circuit 20％ 400Ω 0.95 675kV Instantaneity 1.3077s 1.3132s Negative ground 40％ 300Ω 0.96 672kV Instantaneity 1.3077s 1.3131s Positive ground 60％ 200Ω 0.95 676kV Instantaneity 1.3078s 1.3132s Bipolar short circuit 80％ 100Ω 0.95 673kV Instantaneity 1.3076s 1.3131s Negative ground 100％ 0Ω 0.95 674kV Instantaneity 1.3077s 1.3131s

[0125] Table 4, permanent fault simulation results under 30dB noise:

[0126] Fault type α Transition resistance <![CDATA[m dc ]]> <![CDATA[u dcR ]]> Identification results Positive ground 0％ 500Ω 0.38 0kV Permanent Bipolar short circuit 20％ 400Ω 0.39 165kV Permanent Negative ground 40％ 300Ω 0.54 237kV Permanent Positive ground 60％ 200Ω 0.40 176kV Permanent Bipolar short circuit 80％ 100Ω 0.15 50kV Permanent Negative ground 100％ 0Ω 0.03 1kV Permanent

[0127] It can be seen from Tables 3 and 4 that the proposed fault recovery strategy still has sufficient margin to ensure the correct identification of the fault nature under the noise intensity of 30dB, and can quickly restore power supply under transient faults.

[0128] In a hybrid DC transmission system that uses overhead lines as power transmission channels, DC line faults are inevitable, and a safe, fast, and reliable fault recovery strategy needs to be configured. This paper addresses the problem that the existing recovery strategy lacks a fault nature identification link, makes full use of the control characteristics of the hybrid MMC, and aims to improve the system restart success rate and reduce the secondary impact of the system. After the line is de-ionized, a step current is injected into the line to reflect the dynamic characteristics of the system under different fault properties, and then a DC line fault recovery strategy based on DC current controllability is proposed. The main contributions are as follows:

[0129] 1) The current is selected as the controlled quantity, which effectively prevents the overcurrent shock caused by the injection of characteristic signals when the converter output fails.

[0130] 2) Based on the system equivalent circuits under different fault properties, the response characteristics of electrical quantities after signal injection are analyzed: for transient faults, the current on the inverter side is uncontrollable and the voltage on the rectifier side is the system voltage; for permanent faults, the current on the inverter side is controllable and the voltage on the rectifier side is proportional to the transition resistance.

[0131] 3) Taking the control signal output by PI as the characteristic quantity, a fault property identification criterion based on the DC voltage modulation ratio is constructed on the inverter side.

[0132] 4) A large number of simulation results show that the proposed fault recovery strategy can adapt to various fault conditions occurring on the entire line, with a transient resistance of up to 500Ω, can withstand noise with a signal-to-noise ratio of 30dB, is insensitive to changes in system parameters, and has good universality. In addition, the fault recovery strategy does not rely on the communication system, and has low requirements on the sampling rate of the protection device, making it easy to implement engineering applications.

[0133] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0134] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A UHV hybrid line fault recovery method based on DC current controllability, characterized in that: The following steps are involved: Step 1: The LCC station performs phase shifting after receiving the line protection action signal. After the phase shifting process is completed, the voltage u at the line head end protection installation is collected. dcR Data Substitution dcR (t)>u Δ (10) Check whether the voltage is rebuilt, u Δ The hybrid MMC station switches to the DC current suppression mode after receiving the line protection action signal, and sets the counter n=1, which means that the first injection is about to start; Step 2: After 300ms of deionization, the hybrid MMC switches to the current injection mode and attempts to inject a step current signal with an amplitude of 0.5pu into the line; Step 3: Collect the m output of PI in the period from 2τ to 2τ+T after injection dc Substitute the data into (9) to determine the nature of the fault. dc Failure to meet the formula within 3 consecutive sampling points In the formula, the DC modulation ratio m dc It is obtained from the output of the PI link in the control system, T is the data window length of the identification criterion, T = 5ms; τ = l / v, l is the line length, v is the speed of the traveling wave propagation, which is the speed of light; m Δ As the judgment threshold, take m Δ =0.85; and n <n max , n max If n is the maximum number of injections, then n=n+1, the hybrid MMC switches to the current suppression mode and jumps to step 2. dc Failure to satisfy equation (9) within three consecutive sampling points, and n = n max , then the hybrid MMC is locked, if m dc When there are three consecutive sampling points that all satisfy equation (9), the hybrid MMC immediately switches back to the original DC voltage control; Step 4: When performing steps 2 to 3, determine u in real time dcR Whether it satisfies equation (10), once it satisfies equation (10), LCC immediately switches back to the original DC current control and restores the system current. max If the equation (10) cannot be satisfied within the period of 100 ms, the LCC is locked.

2. The UHV hybrid line fault recovery method based on DC current controllability according to claim 1, characterized in that: Based on the UHV hybrid DC transmission system, the UHV hybrid DC transmission system adopts a double valve group structure for each pole, the parameters of the two poles are symmetrical, LCC commutation is adopted on the rectifier side, that is, double 12-pulse series connection, and full / half-bridge sub-module hybrid MMC commutation is adopted on the inverter side. When the system operates normally, the LCC station adopts fixed DC current control, and the MMC station adopts fixed DC voltage / reactive power control. The voltage and current at the line head end protection installation are respectively recorded as u dcR 、i dcR , the voltage and current at the end-of-line protection installation are denoted as u dcI 、i dcI .

3. The UHV hybrid line fault recovery method based on DC current controllability according to claim 2 is characterized in that: For LCC stations, when a DC line fault occurs, the DC line protection should be activated quickly. For LCC stations, after the pole control system receives the line protection action signal, the thyristor trigger angle is urgently shifted to 120°, so that the LCC switches from rectification to inverter operation, so that the fault energy in the DC system is fed back to the AC grid, reducing the current i flowing into the fault point of the LCC dcR , when i dcR When it decreases to near zero, the firing angle is further phase-shifted to above 150°.

4. The UHV hybrid line fault recovery method based on DC current controllability according to claim 2 is characterized in that: For the LCC station hybrid MMC station, active control of the fault current is achieved by switching the control mode. Thanks to the negative voltage output capability of the full-bridge submodule, the hybrid MMC DC voltage has a certain adjustment margin. On the one hand, during the DC fault ride-through, the DC voltage is adjusted using the adjustment margin to suppress the fault current output to the short-circuit point. On the other hand, during the DC fault recovery, the DC line is injected with a small-amplitude characteristic signal to assist in determining whether the fault point has disappeared.

5. The UHV hybrid line fault recovery method based on DC current controllability according to claim 4 is characterized in that: It also includes the analysis of the response characteristics of electrical quantities under different fault properties, specifically including the analysis of electrical quantities of permanent faults and the analysis of transient fault current characteristics.

6. The UHV hybrid line fault recovery method based on DC current controllability according to claim 5, characterized in that: The electrical quantity analysis of the permanent fault includes current characteristic analysis and voltage characteristic analysis.

7. The UHV hybrid line fault recovery method based on DC current controllability according to claim 5, characterized in that: The instantaneous fault current characteristic analysis includes current characteristic analysis and voltage characteristic analysis.

Images