Modular Multilevel Power Regulator Sub-module Lower Switch Open-circuit Fault Location Method

Abstract

The invention discloses a method for locating the open - circuit fault of the lower switch of a modular multilevel power regulator sub - module, which includes two steps: pre - screening of faulty modules and accurate location of faulty modules. The pre - screening of faulty modules preliminarily screens out possible faulty sub - modules according to the maximum voltage value of each fundamental wave period. The location of faulty modules finally determines the fault location based on the given pulse signal and the change trend of its output voltage. The fault location method of the invention is applicable to the fault location in the rectification state where the voltage change is not obvious, and realizes the highly reliable operation of the modular multilevel power regulator.

Description

Technical Field

[0001] The present invention relates to the field of power electronics, and particularly to a method for locating an open-circuit fault of a lower switch of a sub-module of a modular multilevel power regulator. Background Art

[0002] With the rapid development of high-speed railways, large-capacity single-phase loads are connected to the power grid, resulting in unbalanced three-phase grid loads, generating negative sequence components, polluting the power grid, increasing line losses, and causing equipment heating. The modular multilevel power regulator has the characteristics of reactive power and negative sequence compensation, and has received extensive attention.

[0003] The modular multilevel power regulator realizes high-voltage output through the cascading of low-voltage devices. As the voltage level increases, the number of cascaded sub-modules also increases significantly. With so many sub-modules cascaded, the possibility of sub-module failures increases, and its reliability has become a hot topic of concern. Sub-module failure types can be divided into short-circuit faults and open-circuit faults. Among them, short-circuit faults will cause the current of the device to increase rapidly, which may cause instantaneous damage to the device. Therefore, usually a corresponding short-circuit fault detection function is configured in the drive circuit design to quickly isolate the fault. In contrast, the impact of an open-circuit fault on the sub-module is less than that of a short-circuit fault, and the voltage characteristics of the open-circuit fault vary with different faulty devices. Even under certain working conditions, the sub-module with an open-circuit fault can be normally inserted or removed like a healthy sub-module, and will not affect the operation of the system. However, continuous open-circuit faults may cause an increase in circulating current and voltage rise, affecting the normal operation of other components.

[0004] The modular multilevel power regulator is divided into two converters, an Alfa converter and a Beta converter. When the load changes, the two converters alternately operate in the rectification state and the inversion state. When the converter is in the inversion working state, there is a positive DC component in the current of each bridge arm of the converter. After an open-circuit fault occurs in the sub-module, the voltage of the faulty sub-module shows an upward trend, which is significantly different from the voltage characteristics of normal sub-modules. By extracting the voltage characteristic values, the faulty sub-module and the faulty device can be effectively located. However, for the sub-module in the rectification working condition, there is a negative DC bias in the current of each bridge arm. When the bridge arm current is greater than 0, the voltage of the sub-module with an open-circuit fault in the lower switch rises, while when the bridge arm current is less than 0, the voltage of the sub-module with an open-circuit fault in the lower switch drops. The time of voltage drop is controllable and longer than the time of voltage rise, which will cause the voltage characteristics of its faulty module to be the same as those of normal sub-modules, and the port characteristics are not affected. Therefore, it will be difficult to distinguish the open-circuit fault in the lower switch in the rectification state. And the two converters of the system are always alternately in the rectification and inversion states, and the unclear discrimination characteristics will bring great challenges to the safe and reliable operation of the system. Therefore, it is necessary to further consider the analysis and extraction of the fault characteristics in the rectification state, and there is an urgent need for a fault location method that meets the rectification state. Summary of the Invention

[0005] The object of the present invention is to provide a method for locating the open - circuit fault of the lower switch tube of a modular multilevel power regulator sub - module in view of the deficiencies of the prior art, to solve the problem that it is difficult to identify the open - circuit fault of the lower switch tube in the rectification state, and to meet the fault location in the rectification state.

[0006] The technical solution adopted by the present invention is as follows:

[0007] A method for locating the open - circuit fault of the lower switch tube of a modular multilevel power regulator sub - module includes two steps: pre - screening of fault modules and accurate positioning of fault modules;

[0008] The pre - screening of the fault modules includes the following steps:

[0009] 1.1) Sample the capacitor voltages udc1, udc2,..., udcN of each sub - module of each bridge arm respectively, and within a fundamental wave period of 0 - 0.02 s, compare to obtain the maximum capacitor voltages of each sub - module, which are respectively denoted as udcmax[1], udcmax[2],..., udcmax[N];

[0010] 1.2) Arrange the maximum values of the voltages of each sub - module within the fundamental wave period in descending order of voltage, calculate the voltage difference between the two largest sub - module voltages. If udcmax[1] - udcmax[2]>uth noise (uth noise is the noise voltage value) holds, record the sub - module number k of the highest voltage in the descending voltage arrangement. If it does not hold, return to the previous step;

[0011] 1.3) The counting arrays of the highest voltages of the 1st - Nth sub - modules are Ncnt[1], Ncnt[2],..., Ncnt[N]. When the voltage of the kth sub - module is continuously the highest, the corresponding Ncnt[k] is incremented by 1, and the count values of other sub - modules are cleared to 0;

[0012] When the highest - voltage continuous cycle number Ncnt[k] of the kth sub - module ≥ 5, the kth sub - module is determined as a pre - fault sub - module. When the highest - voltage continuous cycle number Ncnt[k] of the kth sub - module < 5, return to step 1.1);

[0013] The accurate positioning of the fault module includes the following steps:

[0014] 2.1) Give the pulse signal S2[k]=1 to the lower switch tube T2 of the kth pre - fault sub - module. To prevent shoot - through, the pulse signal of the corresponding upper switch tube T1 is S1[k]=0, and record the sub - module voltage value udcf corresponding to Ncnt[k]=5 of the pre - fault sub - module;

[0015] 2.2) Within the time t < Ts / 4 (where Ts is the fundamental wave period), sample the voltage value udck of the k-th sub-module respectively, calculate its absolute value |udck - udcf| with udcf, and compare it with the voltage threshold U dcth If |udck - udcf| > U dcth , then the corresponding voltage difference counter NT2[k] is incremented by 1;

[0016] 2.3) If t > Ts / 4 and |udck - udcf| > U dcth , then the corresponding uth noise increases;

[0017] 2.4) Determine whether NT2[k] is less than the counter threshold Nth. If it is greater than the threshold, it can be determined that the lower switch T2 of the k-th sub-module is open-circuited. If it is less than the counter threshold Nth, it is determined that the sub-module has no fault, and its flag bits Ncnt[k] and NT2[k] are cleared, and the corresponding uth noise increases.

[0018] By repeatedly executing the above two steps in the main program, accurate positioning of the open-circuit fault of the sub-module can be effectively achieved.

[0019] The method for locating the open-circuit fault of the lower switch of the sub-module of the modular multilevel power regulator in the present invention pre-screens the faulty modules according to the maximum voltage value of each fundamental wave period to initially screen out the possible faulty sub-modules. The location of the faulty module is finally determined based on the given pulse signal and the change trend of its output voltage. The fault location method of the present invention is applicable to the fault location in the rectification state where the voltage change is not obvious, and realizes the highly reliable operation of the modular multilevel power regulator. Description of the Drawings

[0020] Figure 1 is a schematic diagram of the topological structure of the modular multilevel power regulator;

[0021] Figure 2 is a schematic diagram of the flow for locating the open-circuit fault of the lower switch of the sub-module of the modular multilevel power regulator. Detailed Embodiment

[0022] The following further elaborates in detail the content of the present invention in conjunction with the embodiments and the drawings, but it is not a limitation to the present invention.

[0023] Refer to Figure 1 , the modular multilevel power regulator is divided into two converters, the Alfa converter and the Beta converter. When the load changes, the two converters alternately operate in the rectification state and the inversion state.

[0024] The primary side of the V / v traction transformer leads to three terminals, which are connected to the A, B, and C phases of the public grid respectively. The current flowing into the transformer is i SA , i SB , i SC The transformer converts the 220kV or 110kV three-phase electricity in the public power grid into two-phase 27.5kV single-phase electricity to supply different power loads. Definition Figure 1 The left side of the V / v traction transformer is connected to the power supply arm with phase b, and the right side is connected to phase a. The power supply voltage of the left and right power supply arms is u sb = u sa =27.5kV, the transformer secondary outflow current is defined as i sb , i sa ,in i sb = i sa The single-phase H-bridge type modular multi-level power conditioners are connected back-to-back to form RPCs, which are directly connected to the power supply arm. i Lb ( i La )、 i cb ( i ca ) are the b-phase (a-phase) locomotive load current and the b-phase (a-phase) current flowing into the converter respectively.

[0025] definition u sb = u sa Indicates the AC voltage of the power supply arms on both sides. i cb = i ca It represents the AC current flowing into the inverter from the power supply arm. Figure 1 The structures of the bridge arms of each phase are exactly the same. The subscripts of the quantities related to the upper bridge arm are represented by p, and the subscripts of the quantities related to the lower bridge arm are represented by n. U dc 、i dc are the DC side voltage and current respectively.

[0026] Figure 1Among them, SM1, SM2, …, SMN represent each sub-module (sub-modular, SM). Wherein, T1 is the power switch device on the sub-module, T2 is the power switch device under the sub-module, and D1 and D2 are freewheeling diodes. When the switch T1 is turned on and the switch T2 is turned off, the sub-module is put into use, and the module voltage is equal to the capacitor voltage; otherwise, it is cut off.

[0027] See Figure 2 , the open-circuit fault location method for the lower switch of the sub-module of the modular multilevel power regulator includes two steps: pre-screening of the fault module and accurate location of the fault module;

[0028] Assume that the open-circuit fault is in the switch T2 of the Nth sub-module (i.e., the last sub-module). The following is to determine the accurate fault location through the location method of the present invention.

[0029] First, perform pre-screening of the fault module, including the following steps:

[0030] 1.1) Sample the capacitor voltages udc1, udc2, …, udcN of the sub-modules of each arm respectively, and within a fundamental wave period of 0 - 0.02 s, compare to obtain the maximum capacitor voltages of each sub-module, which are respectively denoted as udcmax[1], udcmax[2], …, udcmax[N];

[0031] 1.2) Arrange the maximum values of the voltages of each sub-module within the fundamental wave period in descending order of voltage, and calculate the voltage difference between the two largest sub-module voltages. Since udcmax[1] - udcmax[2] > uth noise holds, record the sub-module numbers of the highest voltages in the descending order of voltage as N - 1 and N;

[0032] 1.3) The counting arrays of the highest voltages of the 1st to Nth sub-modules are Ncnt[1], Ncnt[2], …, Ncnt[N]. When the voltage of the kth sub-module is continuously the highest, the corresponding Ncnt[k] is incremented by 1, and the count values of other sub-modules are cleared to 0;

[0033] When the number of consecutive highest voltage periods Ncnt[k] of the kth sub-module ≥ 5, the kth sub-module is determined as a pre-fault sub-module. When the number of consecutive highest voltage periods Ncnt[k] of the kth sub-module < 5, return to step 1.1);

[0034] At this time, it is found that for the last two sub-modules, the number of consecutive highest voltage periods of the (N - 1)th and Nth sub-modules are both greater than 5, and the two modules are determined as pre-fault sub-modules.

[0035] Next, accurately locate the two pre-fault modules, including the following steps:

[0036] 2.1) Given that the pulse signal S2[N - 1] of the lower transistor T2 of the (N - 1)-th pre-fault sub-module is S2[N - 1]=1, to prevent shoot-through, the pulse signal of the corresponding upper transistor T1 is S1[N - 1]=0, and record the sub-module voltage value udcf corresponding to when the pre-fault sub-module Ncnt[N - 1]=5;

[0037] 2.2) Within the time t<Ts / 4, sample the voltage value udc(N - 1) of the (N - 1)-th sub-module, and calculate its absolute value |udc(N - 1)-udcf| with udcf, and compare it with the voltage threshold U dcth If |udc(N - 1)-udcf|>U dcth , then the corresponding voltage difference counter NT2[N - 1] is incremented by 1;

[0038] 2.3) If t>Ts / 4 and |udc(N - 1)-udcf|>U dcth , then the corresponding uth noise increases;

[0039] 2.4) Since NT2[N - 1] is less than the counter threshold Nth, it is determined that the (N - 1)-th sub-module has no fault, and its flag bits Ncnt[N - 1] and NT2[N - 1] are cleared, and the corresponding uth noise increases;

[0040] 2.5) Given that the pulse signal S2[N] of the lower transistor T2 of the N-th pre-fault sub-module is S2[N]=1, to prevent shoot-through, the pulse signal of the corresponding upper transistor T1 is S1[N]=0, and record the sub-module voltage value udcf corresponding to when the pre-fault sub-module Ncnt[N]=5;

[0041] 2.6) Within the time t<Ts / 4, sample the voltage value udc(N) of the N-th sub-module, and calculate its absolute value |udc(N)-udcf| with udcf, and compare it with the voltage threshold U dcth If |udc(N)-udcf|>U dcth , then the corresponding voltage difference counter NT2[N] is incremented by 1;

[0042] 2.7) Since t>Ts / 4 and |udc(N)-udcf|>U dcth , then the corresponding uth noise increases;

[0043] 2.8) Since NT2[N] is greater than the counter threshold Nth, it is determined that the lower transistor T2 of the N-th sub-module has an open-circuit fault.

[0044] The fault location method of the present invention is applicable to fault location in the rectification state where the voltage change is not obvious, and can effectively achieve accurate positioning of the open-circuit fault of the sub-module.

Claims

1. A method for locating the open - circuit fault of the lower switch of a sub - module in a modular multilevel power regulator, characterized in that, the method includes two steps: pre - screening of faulty modules and accurate location of faulty modules; The pre - screening of faulty modules includes the following steps: 1.1) Sample the capacitor voltages udc1, udc2, …, udcN of each arm of the bridge respectively, and within a fundamental wave period of 0 - 0.02 s, compare to obtain the maximum capacitor voltages of each sub - module, denoted as udcmax[1], udcmax[2], …, udcmax[N] respectively; 1.2) Arrange the maximum values of the voltages of each sub-module within the fundamental wave period in descending order of voltage, calculate the voltage difference between the two largest sub-module voltages. If \(u_{dc_{max}}[1]-u_{dc_{max}}[2]>u_{th}\), noise If it holds, record the serial number \(k\) of the sub-module with the highest voltage in the descending voltage arrangement. If it does not hold, return to the previous step; 1.3) The counting arrays for the 1st to Nth sub - modules with the highest voltage are Ncnt[1], Ncnt[2], …, Ncnt[N]. When the voltage of the kth sub - module remains the highest, the corresponding Ncnt[k] is incremented by 1, and the count values of other sub - modules are cleared to 0; When the number of consecutive periods Ncnt[k] of the highest voltage of the kth sub - module is ≥ 5, the sub - module is determined as a pre - faulty sub - module. When the number of consecutive periods Ncnt[k] of the highest voltage of the kth sub - module is < 5, return to step 1.1); The accurate location of faulty modules includes the following steps: 2.1) Given the pulse signal S2[k]=1 for the lower switch T2 of the kth pre - faulty sub - module, and to prevent shoot - through, the pulse signal for the corresponding upper switch T1 is S1[k]=0, and record the sub - module voltage value udcf corresponding to Ncnt[k]=5 of the pre - faulty sub - module; 2.2) Within the time t < Ts / 4, sample the voltage value udck of the k-th sub-module respectively, calculate its absolute value |udck - udcf| from udcf, and compare it with the voltage threshold U dcth If |udck - udcf| > U dcth , then the corresponding voltage difference counter NT2[k] is incremented by 1; 2.3) If t > Ts / 4 and |udck - udcf| > U dcth , then the corresponding uth noise increases; 2.4) Determine whether NT2[k] is less than the counter threshold value Nth. If it is greater than the threshold value, it can be determined that the lower transistor T2 of the k-th sub-module is open-circuited. If it is less than the counter threshold value Nth, it is determined that the sub-module has no fault, and its flag bits Ncnt[k] and NT2[k] are cleared, and the corresponding uth noise Increase; By repeatedly executing the above two steps of pre - screening of faulty modules and accurate location of faulty modules through the main program, the accurate location of the open - circuit fault of the lower switch of the sub - module can be effectively achieved.

Images