A sub-module of MMC with DC fault ride-through capability

Abstract

The application discloses a novel MMC sub-module with DC fault ride-through capability, which comprises three IGBT tubes T1, T2 and T3, three independent diodes D4, D5 and D6, and one capacitor C, the emitter of the T1 is connected with the positive pole of the D4, the negative pole of the D4 is connected with the collector of the T2; the collector of the T1 is connected with the positive pole of the C, the negative pole of the C is connected with the emitter of the T3, the emitter of the T2 is connected with the collector of the T3; the positive pole of the D5 is connected with the negative pole of the C, the negative pole of the D5 is connected with the emitter of the T1; the positive pole of the D6 is connected with the collector of the T3, the negative pole of the D6 is connected with the positive pole of the C, one end of an output port is connected with the positive pole of the D4 and the emitter of the T1, the other end of the output port is connected with the collector of the T3 and the emitter of the T2. The sub-module has the advantages of less devices, simple control, low operation loss, more balanced capacitor voltage and being beneficial to fast restart.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power electronics, in particular to an MMC sub-module with DC fault ride-through capability and a fault ride-through control method thereof. BACKGROUND

[0002] Modular Multilevel Converter (MMC) has the advantages of easy expansion, high output waveform quality, strong fault handling capability and high transmission efficiency, and is one of the main converter topologies in the development and innovation of flexible DC power transmission technology today. In actual engineering, if cables are used to form a flexible DC power grid, the fault that occurs is generally a permanent fault, and power outage for maintenance cannot be avoided. However, compared with overhead lines, cables have the problems of high cost and low transmission capacity, which will seriously limit the establishment and development of flexible DC power grids. The current development trend is to build a flexible DC power grid based on overhead lines. The flexible DC power grid formed by overhead lines has a high rate of non-permanent faults. In order to ensure the power supply reliability of the DC power transmission system, the converter station should not be shut down after a non-permanent fault occurs on the DC side. Therefore, improving the fault ride-through capability of the DC power grid through measures such as limiting fault current is one of the key technologies to meet the conditions of engineering application.

[0003] Fault ride-through refers to continuing to maintain operation without leaving the power grid in the event of a power grid fault, until the fault is resolved and the system returns to normal and smooth operation. When it comes to the field of new energy power generation, wind power, power grid faults can bring a series of transient processes to wind turbine generators and other electrical equipment in wind farms, such as overcurrent, low voltage, overspeed, etc. In the past, wind turbine generators did not have fault ride-through capability, so when a power grid fault occurred, wind turbine generators would generally automatically disconnect from the power grid for safety reasons, increasing the difficulty of restoring the local power grid, worsening the stability of the power grid, and even exacerbating the fault and causing the system to collapse. Therefore, in order to maintain the safe and stable operation of the power grid, power grid departments in various countries have put forward strict technical requirements for the power connection of wind farms according to their actual conditions, among which the fault ride-through capability requirement (Grid Fault Ride Through GFRT) of wind turbines is considered to be the most challenging and important technical requirement. If wind turbines have fault ride-through capability, the number of repeated grid connections of wind turbine generators during faults can be greatly reduced, and the impact on the power grid can be reduced.

[0004] According to different equipment for crossing DC side fault, the fault crossing method can be divided into three types of AC circuit breaker isolation, DC circuit breaker isolation and converter self-clearing. The method of crossing non-permanent fault by tripping AC side circuit breaker cannot meet the performance requirements of flexible DC power grid in action time, power supply reliability and the like, and is only used in cooperation with permanent fault to perform power-off maintenance operation. The DC circuit breaker has certain feasibility, but increases the system cost and power loss, and the DC circuit breaker is not mature for high voltage grade and large capacity engineering scene technology, and cannot meet the actual DC engineering application requirements at present.

[0005] The method of converter self-clearing is based on the inherent topology structure of the converter, and cooperates with the corresponding fault crossing control strategy to achieve the purpose of suppressing and clearing the fault current through the converter. Compared with the method of crossing fault through the DC circuit breaker, this method does not need additional equipment and avoids long switch action time, has strong fault current blocking capability, short fault recovery time after crossing, and is the research hotspot of the fault handling method of the DC power transmission system at present.

[0006] In the traditional MMC sub-module, the half bridge type sub-module (HBSM) has simple control and low cost, but the locked sub-module cannot realize fault crossing after the occurrence of DC side fault; the full bridge type sub-module (FBSM) has fault crossing capability on the premise of retaining the simple control characteristic, but has high cost; the clamp double sub-module (CDSM) simultaneously considers economy and fault crossing capability to a certain extent, but the control becomes more complex and there is the problem of unbalanced capacitor energy after fault crossing, and the CDSM cannot be restarted quickly.

[0007] On the premise of ensuring fault crossing capability, many new sub-module topologies with different performances are derived at present based on the three traditional MMC sub-modules, but all performances required by engineering cannot be considered.

[0008] Therefore, it is urgent to design an MMC sub-module with simple control and low cost, and capable of ensuring fault crossing capability. SUMMARY

[0009] (1) Technical problems to be solved

[0010] In view of the defects of the prior art, the application provides a novel MMC sub-module with DC fault ride-through capability, which is referred to as a bidirectional switch bypass sub-module (BSBSM) in the following, the BSBSM can limit the fault current by reversely connecting the capacitor into the fault loop after fast locking, and the control mode is simple and the BSBSM has the capability of fast restarting.

[0011] (II) Technical solutions

[0012] In order to achieve the above object, the main technical solutions adopted by the application include:

[0013] The application provides a novel MMC sub-module with DC fault ride-through capability, the MMC sub-module comprises three IGBT tubes T1, T2 and T3, three independent diodes D4, D5 and D6, and one capacitor C, the emitter of the T1 is connected with the positive pole of the D4, the negative pole of the D4 is connected with the collector of the T2; the collector of the T1 is connected with the positive pole of the C, the negative pole of the C is connected with the emitter of the T3, the emitter of the T2 is connected with the collector of the T3; the positive pole of the D5 is connected with the negative pole of the C, the negative pole of the D5 is connected with the emitter of the T1; the positive pole of the D6 is connected with the collector of the T3, the negative pole of the D6 is connected with the positive pole of the C, one end of an output port is connected with the positive pole of the D4 and the emitter of the T1, the other end of the output port is connected with the collector of the T3 and the emitter of the T2, wherein the D1-D3 are anti-parallel diodes corresponding to the T1-T3.

[0014] Further, the MMC sub-module does not comprise the independent diode D4 and the anti-parallel diode D2, that is, the D4 and the D2 are replaced by a power transmission line.

[0015] Further, the MMC sub-module comprises a positive terminal "+" and a negative terminal "-", the positive terminal "+" is one end of the output port, and the negative terminal "-" is the other end of the output port.

[0016] Further, the MMC sub-module comprises the following three working modes:

[0017] (1) Mode 1: the T2 and the T3 give on signals, the T1 gives an off signal, the capacitor C is bypassed, and 0 level is outputted;

[0018] (2) Mode 2: the T1 and the T3 give on signals, the T2 gives an off signal, the capacitor C is put into, and 1 level is outputted;

[0019] (3) Locking mode: the T1, the T2 and the T3 all give off signals, the converter is locked, and the capacitor C is reversely put into the fault loop to limit the fault current to perform fault ride-through.

[0020] In another aspect, the application also discloses a control method of the MMC converter with the novel MMC sub-module with DC fault ride-through capability as described above, and the control method comprises the following steps S1-S3:

[0021] Step S1: judging whether the change of the DC side current exceeds a threshold value; if yes, entering step S2, otherwise, cyclically executing S1;

[0022] Step S2: locking all IGBTs of the BSBSM sub-module, and judging whether the DC side fault current is cleared within a specified time; if yes, executing step S3, if no, marking as a permanent fault, and tripping the AC side circuit breaker for power-off maintenance;

[0023] Step S3: marking as the DC side fault clearance and unlocking the IGBTs of the BSBSM sub-module, judging whether the DC side current is overcurrent, if yes, repeatedly executing step S2, if no, indicating that the system returns to normal and the non-permanent fault ride-through is completed.

[0024] (Three) beneficial effects

[0025] From the above technical solution, the application has the following beneficial effects:

[0026] 1) The BSBSM of the application can limit the fault current by quickly locking the capacitor to be reversely connected to the fault loop after the fault, and the control mode of the fault ride-through is simple, and the capacitor voltage is more balanced, so that the MMC converter formed has a fast restart capability, uses few devices and has a low cost, and is suitable for large-scale application.

[0027] 2) The two structures of the BSBSM of the application, the original structure and the derived structure, have outstanding advantages: the original structure has high reliability, and the economy has been greatly improved compared with the full-bridge sub-module with DC fault ride-through capability; and the derived structure has slightly poor reliability, but the economy is further improved on the original structure. BRIEF DESCRIPTION OF DRAWINGS

[0028] The features and advantages of the application will be more clearly understood through reference to the following drawings, which are schematic and should not be understood as limiting the application, in which:

[0029] Figure 1 is a circuit topology structure diagram of the BSBSM described in the application.

[0030] Figure 2 is a fault current flow mode diagram of the BSBSM after being locked, wherein, Fig. (a) is an input state diagram of the BSBSM sub-module when working normally, and Fig. (b) is a working state diagram of the BSBSM sub-module when in bypass state.

[0031] Figure 3 is a schematic diagram of the flow direction of fault current after the BSBSM constituting the MMC of the application locks.

[0032] Figure 4 is a system simulation result diagram after the BSBSM constituting the converter of the application, wherein, Fig. (a) is an AC side A phase output voltage waveform diagram, Fig. (b) is a DC side voltage waveform diagram, Fig. (c) is a DC side current waveform diagram, and Fig. (d) is an A phase bridge arm sub-module capacitor voltage diagram.

[0033] Figure 5 is a flow chart of the fault ride-through control method of the BSBSM constituting the MMC of the application. DETAILED DESCRIPTION

[0034] To make the objects, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be described below in connection with the drawings of the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0035] As shown in Figure 1 , the application designs a new MMC sub-module with DC fault ride-through capability, i.e. a bidirectional switch bypass sub-module (BSBSM), which comprises three IGBTs T1, T2 and T3, wherein D1, D2 and D3 are the corresponding anti-parallel diodes, three independent diodes D4, D5 and D6, and one capacitor C. The emitter of T1 is connected to the positive pole of D4, and the negative pole of D4 is connected to the collector of T2. The collector of T1 is connected to the positive pole of C, and the negative pole of C is connected to the emitter of T3. The emitter of T2 is connected to the collector of T3. The positive pole of D5 is connected to the negative pole of C, and the negative pole of D5 is connected to the emitter of T1. The positive pole of D6 is connected to the collector of T3, and the negative pole of D6 is connected to the positive pole of C. One end of an output port is connected to the positive pole of D4 and the emitter of T1, and the other end of the output port is connected to the collector of T3 and the emitter of T2.

[0036] Of course the BSBSM can also be implemented by other similar transformation structures, which include three groups of switches S1, S2, S3, three diodes D4, D5, D6 and a DC capacitor C, each group of switches including an insulated gate bipolar transistor (T1, T2, T3) and a diode (D1, D2, D3) connected in anti-parallel therewith. Without considering diodes D4, D5, D6, the remaining devices are distributed in a H-bridge topology, in which switches S1, S2 are connected in series, capacitor C and reverse switch S3 are connected in series (IGBT and its anti-parallel diode are referred to as switch S), and then the two groups of series-connected devices are connected in parallel. Diode D4 is connected in series with switch S2, and the direction is consistent with that of T2, which is to combine T2 into a unidirectional switch (the anti-parallel diode of IGBT is a companion device and plays a protective role), that is, when T2 is turned on, the combined unidirectional switch S2 will be turned on, and in other cases it will be in an off state; diode D5 is connected across the midpoint of the two parallel branches, and the direction is consistent with that of T3, that is, when T1 is locked and current cannot flow through capacitor C and switch S1 circuit, D5 and T3 combine to form a unidirectional switch S3, which has the same effect as unidirectional switch S2, in addition, D5 also provides a fault current flow path; diode D6 is connected in parallel with the H-bridge and combines with D5 to form a complete reverse fault current flow path (the current reference direction of the sub-module is shown in the figure), and the output port of the sub-module is connected across the unidirectional switch S2.

[0037] Further consideration, BSBSM also has a derivative structure, which can be removed for economic considerations two diodes (D2, D4), and the structure is not affected in terms of functionality, that is, T2 alone as a unidirectional switch S2. But the reliability of BSBSM sub-module will decrease slightly, in addition to the main function of freewheeling, the IGBT anti-parallel diode also has the function of protecting the IGBT from reverse voltage, and after removal, it is easy to break down when bearing reverse voltage, and the reliability will decrease. But after analysis, T2 does not bear reverse voltage in normal operation, and the freewheeling function of diodes D2 and D4 is not applied at this time, so removing them in normal operation does not affect the basic function, and the impact on reliability is not great. At the same time, it can further improve the economy, and can reduce the cost when the BSBSM sub-module is used on a large scale.

[0038] As shown in Figure 2 , the BSBSM circuit includes two normal working states and one fault ride-through state:

[0039] 1) Normal working state

[0040] When working normally, the sub-module is in the state as shown in the attached Figure 2As shown in (a) of the figure, control T2 is blocked, T1 and T3 are turned on, when the current is positive, it flows through D1, C, D3, when the current is reversed, it flows through T3, C, T1, and the output voltage is the voltage U of the capacitor C C .

[0041] The bypass state is shown in the figure Figure 2 As shown in (b) of the figure, control T1 is blocked, T2 and T3 are turned on, that is, a set of reverse-parallel unidirectional switches S2 and S3 provide a bypass path, when the current is positive, it flows through D4, T2, when the current is reversed, it flows through T3, D5, and the output voltage at this time is 0.

[0042] 2) Fault ride-through state

[0043] After the fault occurs, all IGBTs of the sub-module are blocked, and the fault current flows as shown in the figure Figure 3 When the fault current is positive, it flows through D1, C, D3, when the fault current is reversed, it flows through D6, C, D5, regardless of the direction of the fault current, the capacitor can be charged to the sub-module, that is, after blocking all IGBTs of the sub-module, the capacitor is reversely connected to the fault loop to limit the fault current for fault ride-through.

[0044] Therefore, it can be seen that the BSBSM includes the following three working modes:

[0045] (1) Mode 1: T2 and T3 give on signals, T1 gives an off signal, the capacitor C is bypassed, and the output is 0 level;

[0046] (2) Mode 2: T1 and T3 give on signals, T2 gives an off signal, the capacitor C is put into, and the output is 1 level;

[0047] (3) Blocking mode: T1, T2 and T3 all give off signals, the converter is blocked, and the capacitor C is reversely put into the fault loop to limit the fault current for fault ride-through.

[0048] When the fault occurs, the capacitor reversely put into the blocking mode can clear the fault current, and can realize fast restart of the system after fault removal.

[0049] In addition, the fault ride-through strategy of the MMC converter based on the BSBSM sub-module of the application is shown in the figure Figure 5 .

[0050] Step S1: judge whether the change of the direct current is greater than the threshold value; if yes, go to step S2, otherwise, execute S1 in a loop;

[0051] Step S2: all IGBTs of the blocking BSBSM sub-module are locked, and it is determined whether the DC side fault current is cleared within a specified time; if yes, step S3 is executed, and if no, permanent fault is marked, and the AC side breaker is tripped for power failure maintenance;

[0052] Step S3: the IGBTs of the BSBSM sub-module are unlocked after the DC side fault is cleared, and it is determined whether the DC side current is overcurrent; if yes, step S2 is repeatedly executed, and if no, the system is restored to normal and the non-permanent fault ride-through is completed.

[0053] One of the main differences between the shift full-bridge sub-modular multilevel converter (MMC) of the present application and other MMC topologies in the prior art is that the blocking can achieve the blocking of the fault current, and the capacitor voltage is more balanced, which is beneficial to rapid restart: according to the attached Figure 2 sub-module fault current path and attached Figure 3 In the fault ride-through process, all sub-modules of the MMC are connected in series to share the voltage during the fault ride-through process, and the energy is evenly distributed on each sub-module capacitor. And only when the blocking condition is met under the condition of the MMC three-phase six-bridge arm voltage, the fault ride-through process can be considered to be completed, so after the non-permanent fault on the DC side is eliminated, the voltages of the MMC sub-modules are basically the same, and rapid restart can be achieved without the need for operations such as sub-module capacitor voltage balancing.

[0054] Based on the MATLAB / Simulink simulation platform, an 11-level single-ended MMC system based on a bidirectional switch bypass sub-module is built, a closed-loop current decoupling control is adopted, and nearest level modulation (NLM) is used, and the system parameters are shown in Table 1, and the simulation results are shown in the attached Figure 4 Figures, which cover the normal working process, the fault occurrence and ride-through process, and the system restart process. Among them, the (a)~ (d) in the attached Figure 4 Figures are the waveforms of the AC side A-phase output voltage, the DC side voltage, the DC side current, and the A-phase upper bridge arm sub-module capacitor voltage. It can be seen that before the fault occurs, the AC measured A-phase output voltage is a staircase wave of the equivalent modulated sinusoidal signal, the DC side voltage is stable, the DC side current is quickly stabilized at about 200A after the system starts, and the sub-module capacitor voltage has an additional fluctuation of not more than ±5% from the rated value of 1kV; before the sub-module is blocked after the fault occurs, the AC measured A-phase output voltage is severely distorted, the DC side voltage drops sharply, the DC side current quickly rises to about 950A, and the sub-module capacitor voltage is quickly discharged; after the sub-module is blocked, the DC side current quickly decreases, and the fault current is cleared after 8.6ms, at the same time, the sub-module capacitor is charged to about the rated value of 1kV, and the capacitor voltages of each sub-module are very small, and the system has the ability to restart; after the fault current is cleared and the simulated fault point disappears, the system is restarted, and the process and the waveforms of each electrical quantity are basically the same as the initial system start.

[0055] Therefore, the simulation result verifies that the MMC system based on the bidirectional switch bypass sub-module BSBSM has excellent normal working, fault ride-through and restart capabilities.

[0056] Table 1. Parameters of single-ended MMC-HVDC test system

[0057]

[0058] Note: The time from fault occurrence to sub-module IGBT blocking is 25 ms. In actual engineering, this process is about 2-3 ms. The time is prolonged here mainly to more intuitively reflect the sub-module capacitor voltage discharge and the charging and discharging process during fault ride-through.

[0059] The main difference between the shift full-bridge sub-modular multilevel converter (MMC) of the present application and other MMC topologies in the prior art is the simplicity of control and low switching loss, which saves investment cost. In normal operation, the most involved is the switching of the input and bypass states of the sub-module. After the fault occurs, the switching process from the normal working mode to the blocking mode is mainly concerned. Therefore, the control signals of the three working modes of the BSBSM are first given, as shown in Table 2, 0 represents an off signal, and 1 represents an on signal.

[0060] From Table 1, it can be further analyzed that in normal operation, T3 is in the open state and does not involve signal changes. Therefore, the switching of the input and bypass states of the BSBSM is only the switching of T1 and T2 states, which is basically the same as the simple control mode of the traditional HBSM. When switching from the normal working state to the blocking state, only the switching of two IGBT states is involved, such as the switching from the input state to the blocking state, keeping the original T2 off state, and switching T1 and T3 from the on state to the off state. The bypass state switches to the blocking state in the same way. Therefore, the BSBSM also has certain advantages in the process of blocking the sub-module due to its simple control.

[0061] The switching of the working state of the BSBSM only involves the switching of two IGBT switch states, which is consistent with the traditional HBSM. Therefore, low switching loss is also a big advantage of the BSBSM.

[0062] Table 2. Control signals of three working modes of BSBSM

[0063]

[0064] The determination of the BSBSM economy is mainly determined from two aspects of the pre-engineering and operation. Since the BSBSM and the traditional full-bridge FBSM are basically consistent in working performance (the BSBSM control is simpler), both have excellent fault ride-through capability and fast restart capability, therefore, the comparison is made with the FBSM.

[0065] The pre-engineering investment cost includes the number of IGBTs, diodes, capacitors and thyristors required per level;

[0066] The operation cost includes the on-state loss (the number and types of power electronic devices in normal operation), and the dynamic loss (the number and frequency of the corresponding action switches under different control strategies and different topological structures).

[0067] According to the attached Figure 2 It can be analyzed that in the three working states of the sub-module, the maximum voltage resistance of all devices is the capacitor voltage U C of the BSBSM, as shown in the attached Figure 1 Each sub-module needs at most 3 IGBTs and 6 diodes, and the cost is lower after removing D2 and D4. The cost of large-capacity diodes is much lower than that of IGBTs. Compared with the FBSM which needs 4 IGBTs and 4 diodes, one IGBT is reduced and two diodes are increased, but the performance is better.

[0068] 2) Operation cost

[0069] The dynamic loss, i.e. the switching loss, has been analyzed together with the control simplicity. In normal operation, each switching of the working state only involves the switching of the switching state of two IGBTs, which belongs to the type of sub-module with the lowest dynamic loss.

[0070] Through the above analysis, the BSBSM of the MMC sub-module has good economy and simple control mode.

[0071] Although the embodiments of the present application are described in combination with the attached drawings, various modifications and variations can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. An MMC submodule with DC fault ride-through capability, characterized in that, The MMC submodule includes three IGBT transistors T1, T2, and T3, three independent diodes D4, D5, and D6, and one capacitor C. The emitter of T1 is connected to the positive terminal of D4, and the negative terminal of D4 is connected to the collector of T2. The collector of T1 is connected to the positive terminal of C, and the negative terminal of C is connected to the emitter of T3. The emitter of T2 is connected to the collector of T3. The positive terminal of D5 is connected to the negative terminal of C, and the negative terminal of D5 is connected to the emitter of T1. The positive terminal of D6 is connected to the collector of T3, and the negative terminal of D6 is connected to the positive terminal of C. One end of the output port is connected to the positive terminal of D4 and the emitter of T1, and the other end of the output port is connected to the collector of T3 and the emitter of T2. D1 to D3 are anti-parallel diodes corresponding to T1 to T3. The MMC submodule includes the following three working modes: (1) Mode 1: T2 and T3 provide turn-on signals, T1 provides turn-off signals, capacitor C is bypassed, and the output level is 0. When the current is in the positive direction, it flows through D4 and T2, and when the current is in the reverse direction, it flows through T3 and D5. At this time, the output voltage is 0. (2) Mode 2: T1 and T3 provide the turn-on signal, T2 provides the turn-off signal, capacitor C is connected, and the output is level 1; (3) Blocking mode: T1, T2 and T3 all give a shutdown signal, the converter is blocked, and capacitor C is reverse-connected into the fault circuit to limit the fault current and allow fault ride-through. During normal operation, T3 is in the on state and does not involve any change in the applied signal. Therefore, the switching between the on and bypass states of the MMC submodule is only a switching of the T1 and T2 states, which is the same as the simple control mode of the traditional HBSM. When switching from the normal operation state to the lockout state, it only involves the switching of the two IGBT states. When switching from the on state to the lockout state, the original T2 is off, and T1 and T3 are switched from the on state to the off state. When switching from the bypass state to the lockout state, the original T1 is off, and T2 and T3 are switched from the on state to the off state. The control method for the MMC converter with DC fault ride-through capability of the MMC submodule specifically includes the following steps S1-S3: Step S1: Determine whether the change in DC side current exceeds the threshold; If yes, proceed to step S2; otherwise, repeat step S1. Step S2: Lock out all IGBTs of the MMC submodule and determine whether the DC side fault current is cleared within the specified time; if yes, proceed to step S3; otherwise, mark it as a permanent fault and trip the AC side circuit breaker for power outage maintenance. Step S3: Mark the DC side fault cleared and unlock the IGBT of the MMC submodule. Determine whether the DC side current is overcurrent. If yes, repeat step S2. Otherwise, it means that the system has returned to normal and the non-permanent fault ride-through has been completed.

2. The MMC submodule with DC fault ride-through capability according to claim 1, characterized in that, The MMC submodule does not include independent diode D4 and anti-parallel diode D2, meaning that D4 and D2 are replaced by power lines.

3. The MMC submodule with DC fault ride-through capability according to claim 1, characterized in that, It includes a positive terminal "+" and a negative terminal "-", wherein the positive terminal "+" is one end of the output port and the negative terminal "-" is the other end of the output port.

Images