Direct current and trigger angle cooperative control strategy for suppressing continuous commutation failure

Abstract

The invention discloses a cooperative control strategy for suppressing continuous commutation failure of new energy through a direct current delivery system and a parameter setting method thereof, which are suitable for the field of high-voltage direct current transmission. The strategy comprises the following steps: determining key factors influencing the continuous commutation failure according to a direct current system control equation and an operation characteristic curve; on the basis of judgment of alternating current fault severity based on commutation failure fault prediction control, a cooperative control strategy for improving a direct current and an inversion side trigger angle instruction is provided, the direct current is reduced and the trigger angle is increased at the same time according to the alternating current fault severity, and a weight value is dynamically adjusted to achieve a better control effect. The strategy can effectively suppress the DC continuous commutation failure problem caused by the AC system fault at the inverter side of the high-voltage DC power transmission system, increase the turn-off angle margin, retain the voltage support capability, reduce the active power loss during the fault period, increase the consumption of the DC system to the surplus reactive power of the sending end, and effectively suppress the overvoltage level of the sending end.

Description

Technical Field

[0001] The present application belongs to the technical field of high-voltage direct current transmission, and specifically relates to a strategy for suppressing continuous commutation failure of a high-voltage direct current transmission system of a grid-commutated converter and a parameter setting method. Background Art

[0002] Renewable energy sources such as wind power and photovoltaic power will become the main source of my country's energy growth. In the face of the reverse distribution of energy and load in my country, high-voltage direct current transmission has become one of the ways to solve the problem of new energy transmission and consumption. Commutation failure is one of the most common faults of the grid-commutated converter high voltage direct current transmission system (line commutated converter high voltage direct current, LCC-HVDC), which will cause the DC voltage to drop rapidly and the DC current to increase sharply, seriously impacting the AC system. If the DC system is not properly controlled, after the first commutation failure, the DC system will be induced to fail for the second or multiple times due to excessive changes in voltage, current, phase angle, etc., aggravating the transient voltage fluctuations of the commutation busbar at the new energy machine end and the DC sending end, and even causing large-scale new energy units to disconnect from the grid due to overvoltage, seriously threatening the safe and stable operation of the sending and receiving end systems.

[0003] The first commutation failure of DC occurs at the initial moment of AC fault, which is difficult to avoid. The subsequent commutation failure suppression caused by it has attracted the attention of scholars. At present, scholars at home and abroad have carried out a lot of research on the suppression of subsequent commutation failure of high-voltage DC transmission. Among them, the method of optimizing the control and protection strategy has received widespread attention because it is more economical and operational. In order to improve the arc extinction angle margin, some scholars proposed to consider the severity of AC faults and propose a control method of adaptive dynamic adjustment of the trigger angle in combination with the power recovery speed during the recovery process. A nonlinear dynamic current deviation control method was also proposed, which replaced the original linear function with a nonlinear curve to improve the sensitivity of the turn-off angle increment to the current deviation. However, in a weak receiving-end AC system, this type of trigger angle advance control may cause excessive DC current and excessive reduction of the receiving-end AC voltage, increasing the risk of commutation failure. Some scholars analyzed the disadvantages of trigger angle advance control and proposed to prevent continuous commutation failure by reducing the current command. Some scholars proposed a maximum current limit based on the quasi-steady-state formula by measuring the converter station voltage and control angle constraints. Some scholars proposed a control strategy to reduce or restore DC transmission power in real time based on the critical voltage risk prediction results. Some scholars have also proposed to improve the current control performance by improving the low-voltage current limiting link. Therefore, the existing strategy mainly suppresses subsequent commutation failures by reducing the DC power or trigger angle advance control during fault recovery. However, suppressing continuous commutation failures by reducing part of the DC power will worsen the AC overvoltage level at the new energy grid connection point to a certain extent; and although the trigger angle advance control will not aggravate the overvoltage level, the AC voltage recovery performance at the receiving end will be affected. Summary of the invention

[0004] In view of the shortcomings or improvement needs of the existing strategies for suppressing continuous commutation failures in DC systems, the present invention proposes a DC current and trigger angle coordinated suppression strategy that can effectively suppress continuous commutation failures in DC transmission systems and overvoltages caused by new energy grid connection at the sending end.

[0005] The present invention is based on the correction of the DC current command value and the trigger angle of the inverter side when there is a risk of continuous commutation failure. It uses the DC power quasi-steady-state formula, the fault severity measurement module and the weight coefficient distribution to calculate the correction command of the DC current and the trigger angle, so as to achieve the function of continuous commutation failure and the overvoltage suppression of the sending end grid connection. The strategy control block diagram is shown in the figure. Figure 7 As shown in the figure, by using the commutation failure prediction control principle, the DC current and trigger angle are corrected when the risk of secondary commutation failure is detected during the fault recovery period, and different control effects are achieved by changing the weight value λ.

[0006] DC current and trigger angle maximum correction instruction calculation module. Figure 7The lower part shows the calculation module of the DC current correction command ΔΙ and the trigger angle correction command Δβ. According to the DC power quasi-steady-state formula (1), the γ and β constraints based on the fault AC voltage component are set to calculate the maximum value command.

[0007]

[0008] In order to prevent continuous commutation failure during the fault recovery period, it is hoped that γ is close to the normal level and combined with U in the commutation failure prediction control. CF The margin is further improved, as shown in formula (2). In the formula, k is the tuning parameter.

[0009] γ=γ0+kU CF (2)

[0010] The dynamic weight value calculation of overvoltage is considered. In the cooperative control strategy, Δβ max The weight value is λ, ΔI max The weight value is 1-λ. The smaller the weight value λ is, the smaller the proportion of trigger angle control is, the smaller the required exchange reactive power at the receiving end is, and the less the AC voltage drops; the larger the λ is, the smaller the proportion of current control is, the smaller the transmission power loss is, and the DC system consumes more reactive power during fault recovery, which can suppress the AC overvoltage at the sending end. Therefore, the two controls are coordinated to achieve the best control effect and the appropriate λ value is designed. Specifically, Figure 8 As shown in formula (3-5).

[0011] λ=ΔU Ⅰ ΔU Ⅱ (3)

[0012] ΔU Ⅰ =λ max +(U acr -U acrref )+(U aci -U aciref )(4)

[0013]

[0014] The present invention provides a DC current and trigger angle coordinated suppression strategy that can effectively suppress continuous commutation failure of DC power transmission system and overvoltage of new energy grid-connected at the sending end. The strategy increases the turn-off angle margin during fault recovery by simultaneously reducing the DC current command value and increasing the inverter-side trigger angle command value, and dynamically adjusts the weight values ​​of the two to achieve a better control effect. The continuous commutation failure of the DC power transmission system and the overvoltage at the sending end are suppressed, and the problem that reducing part of the DC power to suppress continuous commutation failure will deteriorate the AC overvoltage level of the new energy grid-connected point to a certain extent is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the DC system steady-state operation curve and fault transient curve.

[0016] Figure 2 Schematic diagram for comparing the control characteristic curves before and after reducing the DC current.

[0017] Figure 3 This is a schematic diagram comparing the control characteristic curves before and after increasing the trigger angle.

[0018] Figure 4 Schematic diagram of the DC system operation curve under coordinated control.

[0019] Figure 5 Schematic diagram of the collaborative control strategy.

[0020] Figure 6 Schematic diagram of the weight value calculation module.

[0021] Figure 7 This is a schematic diagram of the structure of the LCC-HVDC transmission system for a large wind power base.

[0022] Figure 8 The three-phase fault simulation waveforms under three different control strategies are shown in Figure 2. Figure 8 (a) is the turn-off angle waveform. Figure 8 (b) is the DC current waveform. Figure 8 (c) is the AC voltage waveform of the wind turbine grid connection point. Figure 8 (d) is the AC voltage waveform on the inverter side.

[0023] Fig. 9 The single-phase fault simulation waveforms under three different control strategies are shown below. Fig. 9 (a) is the turn-off angle waveform. Fig. 9 (b) is the DC current waveform. Fig. 9 (c) is the AC voltage waveform of the wind turbine grid connection point. Fig. 9 (d) is the AC voltage waveform on the inverter side.

[0024] Fig.10 This is a waveform diagram comparing the control effects under different fault levels. Fig.10 (a) is a comparison of single-phase faults. Fig.10 (b) is a comparison of three-phase faults. DETAILED DESCRIPTION

[0025] The present invention is further described below in conjunction with the accompanying drawings and implementation examples.

[0026] Figure 1The figure shows the DC system operation curve and fault transient curve. Point A in the figure is the steady-state operation point of the DC system, which is located at the intersection of the constant current control curve on the rectifier side and the constant turn-off angle control curve on the inverter side. When a fault occurs on the AC busbar on the inverter side, the system operates according to the fault curve in the figure. The first commutation failure ends and the fault recovery phase begins. The DC current I d It recovers to the reference value along the low voltage current limiting (VDCOL) control curve. After reaching point E, since I d The reference value has not yet been restored, and the current deviation (CEC) control begins to work. The operating point moves along the fixed firing angle control characteristic curve EF, and the turn-off angle γ continues to decrease. d It is generally believed that when γ is less than the threshold (7°), commutation failure will occur. Therefore, the EF characteristic curve directly affects whether the second commutation failure occurs.

[0027] According to the above analysis, in order to prevent subsequent commutation failure, it is necessary to reduce the γ reduction during EF, that is, to adjust downward Figure 1 The factors that affect γ during EF are mainly the DC current I d and the inverter side trigger angle β. The relationship between the three is shown in formula (6), where U aci is the inverter side AC voltage, T1 is the converter transformer ratio, x c is the equivalent commutation reactance. Further analysis of the control characteristic curve can reduce I d The characteristic curve changes after increasing β as shown in Figure 2 Figure 3 shown.

[0028]

[0029] Figure 2 In the process of reducing the DC current reference value under VDCOL control, I d , that is, the original VDCOL control curves of the rectifier side and the inverter side are simultaneously shifted to the left, such as Figure 2 As shown by the red line in the middle, after translation, the original stable operation point F moves to point F', γ increases, and the possibility of subsequent secondary commutation failure is reduced. Figure 3 Increasing β is equivalent to increasing the DC current reference value on the inverter side during constant current control, and the control characteristic curve, i.e., the inverter side VDCOL, moves to the right. Assuming that the DC voltage U d can be restored to the same level, increasing β can make I d Return to a value closer to the steady state, that is, Figure 3The midpoint E moves to the right to point E”, and then moves down to F” along the current deviation control point F, γ increases to prevent the subsequent secondary commutation failure. However, since there is only a 0.1pu margin between the original inverter side VDCOL and the rectifier side VDCOL, the adjustable range of β is small. In addition, increasing β will increase the reactive power consumption level of the DC system during fault recovery, and the AC voltage is at a low level, which puts higher requirements on the reactive power configuration of the receiving system. However, reducing I d It will reduce the reactive power consumed by the DC converter, and the reactive power generated by the AC filter on the rectifier side will be more difficult to be consumed, which will cause the overvoltage at the sending end to further increase during the fault recovery period, and worsen the operating conditions of the new energy units at the sending end.

[0030] From the perspective of suppressing overvoltage of renewable energy grid connection, the DC side needs to increase the DC current I during fault recovery. d From the perspective of suppressing continuous commutation failure, it is also necessary to reduce the fault recovery period I d Obviously, only control I d It is impossible to simultaneously meet the requirements of suppressing overvoltage and continuous commutation failure. Increasing β at the same time is essentially to increase the arc extinction angle margin and reduce the risk of commutation failure by controlling the DC system to trigger in advance. However, increasing the starting voltage threshold will also cause the inverter station to absorb more reactive power from the AC system, which is not conducive to AC voltage recovery. If the two control methods are reasonably combined and I is reduced at the same time, d Increasing β, the system operation curve is as follows Figure 4 As shown. The VDCOL control curve on the rectifier side moves to the left to reduce I d The reference value, the VDCOL control curve on the inverter side moves to the right to increase the β reference value, and the stable operation point is located at point F''. It can be seen that point F'' is different from the single control I d At the lower point F', the DC current is closer to the steady-state value, which reduces the active power loss during the fault and increases the reactive power consumption of the sending-end DC converter. Compared with point F", F"' under the single control of β control, the voltage drop is less, the reactive power demand of the receiving-end converter is lower during the fault, and it is more conducive to fault recovery.

[0031] Therefore, a coordinated control strategy is proposed by combining DC current control with inverter side trigger angle control. Figure 5 As shown in the figure, by using the commutation failure prediction control principle, the DC current and trigger angle are corrected when the risk of secondary commutation failure is detected during the fault recovery period, and different control effects are achieved by changing the weight value λ. Figure 5 The lower part is the calculation module of the DC current correction command ΔI and the trigger angle correction command Δβ. The maximum values ​​of the two correction commands are coordinated and controlled by the weight value λ, and combined with the commutation failure prediction control principle, the γ and β constraints based on the fault AC voltage component are set according to the quasi-steady-state formula (1) to calculate the maximum value command. dcis the DC power, x c is the equivalent commutation reactance.

[0032] Taking the CIGRE standard test model as an example, the shutdown angle γ0 is set to 15° and β0 is 38° when the system is operating normally, and the shutdown angle γ' is 0° when the commutation fails. In order to suppress the continuous commutation failure during the fault recovery period, it is hoped that γ is close to the normal level, combined with the U in the commutation failure prediction control. CF The margin is further improved, as shown in formula (2). In the formula, k is the tuning parameter.

[0033] The larger RMS value of the three-phase zero-sequence voltage for asymmetric fault and the two-phase stationary coordinate voltage for symmetric fault is used in the commutation failure prediction control. CF To describe the severity of the fault, the calculation formula is as follows (7-9). The three-phase zero-sequence voltage of an asymmetric fault is the sum of the voltages, as shown in (8); the symmetric fault detection transforms the instantaneous value of the three-phase voltage into the voltage value under the two-phase stationary coordinate through coordinate transformation, and takes its root mean square, as shown in (9).

[0034] U CF =max(U0,U αβ )(7)

[0035] U0=U a +U b +U c (8)

[0036]

[0037] Correcting the DC current to improve the γ margin will result in a partial loss of DC power. Assuming that during the fault recovery phase, the inverter side AC voltage U aci is approximately equal to the normal value, and the inverter side trigger angle is also kept near the normal value β0. Substituting equation (2) into (1) yields equation (10), where P dc ' is the DC power after increasing the γ margin. Combining equation (10) with equation (1) where the turn-off angle is γ' when the commutation fails, we can solve for ΔI max As shown in formula (11).

[0038]

[0039]

[0040] Correcting the trigger angle to increase the γ margin has little effect on the DC active power. Substituting equation (2) into equation (1) yields equation (12), where β' is the inverter trigger angle after increasing the γ margin. Combining equation (12) with equation (1) with a turn-off angle of γ0, we can solve for Δβ max As shown in formula (13).

[0041]

[0042]

[0043] Finally, from equation (11) and equation (13), we can get Figure 5 Medium ΔI max With Δβ max The instruction computes a value.

[0044] Figure 6 The figure shows the schematic diagram of the weight value calculation module. In the collaborative control strategy, Δβ max The weight value is λ, ΔI max The weight value is 1-λ. The smaller the weight value λ is, the smaller the proportion of trigger angle control is, the smaller the required exchange reactive power at the receiving end is, and the less the AC voltage drops; the larger the λ is, the smaller the proportion of current control is, the smaller the transmission power loss is, and the DC system consumes more reactive power during fault recovery, which can suppress the AC overvoltage at the sending end. Therefore, in order to coordinate the two controls to achieve a better control effect, it is necessary to design a suitable λ value. The specific details are shown in formula (3-5).

[0045] In formula (3), λ is given by ΔU Ⅰ With ΔU Ⅱ It consists of two parts. Ⅰ is the voltage deviation component of the DC system, which is dynamically adjusted by the AC voltage deviation between the sending and receiving ends; ΔU Ⅱ is the voltage change rate deviation component. On the basis of the former, λ is adjusted twice to improve the sensitivity of overvoltage suppression and subsequent commutation failure suppression. k1 is the setting parameter, T1 is the measurement link time constant, and G is the gain coefficient, which has the same value as the CIGRE test model.

[0046] ΔU Ⅰ Calculated by formula (4). In formula (4), U acr with U aci are the rectifier side AC voltage and the inverter side AC voltage, U acrref with U aciref is the corresponding reference value. In the fault recovery stage, due to the AC filter reverse reactive power and continuous commutation failure suppression measures, the DC reactive power consumption is reduced, resulting in U acr Further increase. λ also increases under its influence, the proportion of current control in the coordinated control decreases, slowing down the DC current drop speed, and the DC system can consume more surplus reactive power at the sending end to achieve the effect of suppressing overvoltage. At the same time, the proportion of trigger angle control begins to increase, compensating for the loss of the continuous commutation failure effect caused by the reduction of current control proportion. On the other hand, when λ is too large and the trigger angle compensation amount on the inverter side is too much, the reactive power consumption of the receiving-end DC converter increases, and U aci will decrease slightly again. From formula (3) and (4), we can know that ΔU Ⅰwill decrease, and λ will also decrease accordingly, which adaptively reduces the proportion of trigger angle control, reduces the reactive power demand of the receiving-end converter station, and makes the AC voltage recovery more stable.

[0047] ΔU Ⅱ Introducing the first-order differential of the sending-end voltage in the λ calculation can make the coordinated control more sensitive to overvoltage suppression and subsequent commutation failure suppression. As shown in formula (5), when the voltage at the sending-end grid connection point rises, suppressing overvoltage is the main goal. At this time, ΔU Ⅱ >1,ΔU Ⅰ With ΔU Ⅱ All of them increase λ, the proportion of current control in the coordinated control strategy decreases, and the DC system absorbs more reactive power to suppress overvoltage. When the voltage at the sending end grid connection point begins to drop, subsequent commutation failure gradually becomes a key issue, and suppressing overvoltage becomes a secondary goal. At this time, ΔU Ⅱ <1, and the growth rate of λ is appropriately reduced under its influence, which can more smoothly and quickly transition to the subsequent process of λ reduction, and more effectively suppress subsequent commutation failure.

[0048] The specific implementation of the present invention is now described in detail with reference to the accompanying drawings and embodiments, but the present invention is not limited to the described embodiments. The steps are as follows:

[0049] 1) In the PSCAD electromagnetic simulation software, a large wind power base is constructed through the LCC-HVDC transmission system, that is, Figure 7 The simulation model of the test system is shown. Two working conditions, symmetrical fault and asymmetrical fault, are set to simulate and test the three control strategies respectively;

[0050] 2) Working condition 1: When the fault is set to 5s, a three-phase short circuit fault with a grounding inductance of 0.2H occurs on the inverter-side AC busbar, and the fault duration is 0.2s;

[0051] 3) Working condition 2: When the fault is set to 5s, a single-phase short circuit fault with a grounding inductance of 0.25H occurs on the AC busbar on the inverter side, and the fault duration is 0.2s;

[0052] 4) Three control strategies: Control scheme 1 is the CIGRE standard test model, control scheme 2 is the CIGRE standard test model plus the continuous commutation failure suppression method proposed in the existing literature, and control scheme 3 is the CIGRE standard test model plus the continuous commutation failure suppression method proposed in the present invention;

[0053] 4) Under the two working conditions of three-phase short circuit fault and single-phase short circuit fault, a fault with a duration of 0.2s is set for simulation verification, and the fault level F shown in formula (14) is introduced L Where P dN is the rated DC power, ω is the AC system angular frequency, U L is the AC bus voltage.

[0054]

[0055] Depend on Figure 8 It can be seen that in the CIGRE standard test model of control scheme 1, the DC system will have two consecutive commutation failures, the sending-end AC voltage at the wind turbine grid connection point will produce two overvoltage peaks, and the receiving-end AC voltage will also drop twice. Both control schemes 2 and 3 can effectively suppress continuous commutation failures, and the secondary drop of the receiving-end voltage has been significantly improved, increasing the voltage by 0.06pu during the fault recovery period. However, control scheme 2 did not show effective improvement in the grid-connected overvoltage peak, and the control scheme 3 proposed in this article has a larger DC current during the fault recovery period. While the DC system has a stronger power transmission capability, it consumes more reactive power from the sending-end AC filter. Therefore, the transient overvoltage level at the renewable energy grid connection point under control scheme 3 is lower, and the overall decrease is 0.08pu during the fault recovery period. .

[0056] Depend on Fig. 9 It can be seen that both control scheme 2 and control scheme 3 can effectively suppress continuous commutation failure, and the secondary drop amplitude of the AC voltage at the receiving end of the inverter side is significantly improved. The voltage of control scheme 3 is 0.045pu higher than that of control scheme 1 during fault recovery. Moreover, similar to working condition 1, compared with control scheme 2, the DC current of control scheme 3 proposed in this paper is larger during fault recovery, consuming more surplus reactive power at the sending end. Therefore, the transient overvoltage level at the renewable energy grid connection point under control scheme 3 is lower, 0.07pu lower than that of control scheme 2.

[0057] Depend on Fig.10 It can be seen that the CIGRE standard test system has subsequent commutation failure after the single-phase fault level reaches 45%, while the collaborative control strategy proposed in the present invention only has the first commutation failure, effectively suppressing subsequent commutation failure. Under three-phase faults, the CIGRE standard test system has subsequent commutation failure after the fault level reaches 55%, while the collaborative control strategy proposed in the present invention has a stable ability to suppress subsequent commutation failure.

Claims

1. An improved DC current and firing angle command value coordinated control strategy for suppressing continuous commutation failure, characterized in that: When the risk of continuous commutation failure is detected in the DC system, the improved instructions of the DC current on the rectifier side and the trigger angle on the inverter side are calculated according to the fault severity obtained by the continuous commutation failure prediction control; and the two are coordinated and controlled through weight parameters.

2. According to the requirements of claim 1, the collaborative control strategy includes the selection logic of the correction parameters.

3. According to the requirements of claim 1, the improved instruction calculation method for DC current and inverter side trigger angle includes an AC fault severity calculation method, a maximum instruction value calculation method, and a weight parameter setting method.

Images