Fault data generation and acquisition method of modular multilevel converter

Abstract

The invention discloses a fault data generation and acquisition method for a modular multilevel converter, and the method comprises the steps: dividing each acquisition period Tsam into a fault period and a recovery period, and enabling the duration of the acquisition period to be three fundamental frequency periods; the starting point t0 of the acquisition period is a zero crossing point at which the alternating current of the first fundamental frequency period is changed from negative to positive; the duration of the fault period is one fundamental frequency period T0, and the starting point tF of the fault period is [t0, t0 + T0]; in each acquisition period, except the fault period, the other time periods are recovery periods; in each fault period, a sub-module is randomly calibrated as a target sub-module, the target sub-module is controlled to simulate an open-circuit fault so as to obtain fault data, and meanwhile, the output voltage of the target sub-module is compensated through the redundant sub-module, so that the influence on normal operation of the modular multilevel converter is avoided. The method provided by the invention does not need to introduce any hardware, can be widely applied to various voltage levels and power levels, does not affect the normal operation of the modular multilevel converter in the data acquisition process, does not damage equipment, and can realize low-cost and high-reliability fault data acquisition.

Description

Technical Field

[0001] The present invention belongs to the technical field of direct current transmission, relates to the reliability of medium and high voltage power electronic converters, and in particular to a method for generating and collecting fault data of a modular multi-level converter. Background Art

[0002] The modular multi-level converter system is huge and contains many sub-modules, each of which contains several power switching devices. The open circuit fault of the power switching device is not easy to cause significant fault characteristics at the sub-module level in a short time, making it difficult to quickly identify and bypass the faulty sub-module. Existing fault location methods are generally divided into two categories: hardware-based fault location methods and software-based fault location methods. Among them, software-based fault location methods are popular due to their wide range of applications, no need for additional hardware introduction, and low cost.

[0003] Software-based fault location methods include model-based fault location methods and data-based fault location methods. Among them, the model-based fault location method needs to set a suitable threshold value, so as to determine the location of the faulty submodule by comparing the difference between the relevant variables of the faulty submodule and the reference value; since modular multi-level converters often face a more complex operating environment and a variety of different working conditions, it is difficult to fully meet the requirements of accurate and rapid location under different conditions. The data-based fault location method can automatically identify the faulty submodule by training the collected data, without setting a threshold, and can quickly and accurately locate the fault. However, the acquisition of fault data has become one of the bottlenecks for the widespread promotion of this method. Due to the small amount of fault data in actual projects, the data training requirements cannot be met; through laboratory prototype simulation, the prototype is easily damaged due to the degree of fault during the fault simulation process, resulting in a serious increase in costs; and the data training results obtained by laboratory prototype simulation may not meet the operation requirements of actual projects.

[0004] At present, most of the methods for collecting open-circuit fault data of modular multi-level converters are based on simulation models or laboratory small-power prototype data collection. The existing methods may not meet the positioning requirements of actual projects operating in complex scenarios and at multiple power levels. At the same time, the fault data acquisition process of small-power prototypes is destructive and harmful to the prototypes. Under the demand for large-scale fault data acquisition, the cost of data generation and collection is high.

[0005] Therefore, it is necessary to explore more efficient, lower-cost, more reliable and more adaptable fault data generation and collection methods to achieve accurate and rapid data-based fault location of modular multilevel converters and improve operational reliability. Summary of the invention

[0006] The purpose of the present invention is to provide a method for generating and collecting fault data of a modular multilevel converter, which does not require any hardware introduction and can be widely applied to various voltage levels and power levels. The data collection process does not affect the normal operation of the modular multilevel converter and does not damage the equipment, and can achieve low-cost and highly reliable fault data collection.

[0007] In order to achieve the above object, the solution of the present invention is:

[0008] A method for generating and collecting fault data of a modular multilevel converter, wherein the modular multilevel converter is a three-phase topology structure, each phase comprises an upper bridge arm from a positive pole to a connection point and a lower bridge arm from a connection point to a negative pole, wherein the upper bridge arm comprises N submodules and a first inductor sequentially connected in series from the positive pole to the connection point, and the lower bridge arm comprises N submodules and a second inductor sequentially connected in series from the negative pole to the connection point; each submodule comprises an upper tube, a lower tube and a capacitor; the method comprises,

[0009] Each acquisition cycle T sam It is divided into a failure period and a recovery period, where the duration of the collection period is T sam =3T0, T0 is a fundamental frequency cycle; the starting point t0 of the acquisition cycle is the zero-crossing point where the AC current of the first fundamental frequency cycle changes from negative to positive; the duration of the fault cycle is 1 fundamental frequency cycle T0, and its starting point t F =[t0,t0+T0]; in each acquisition cycle, except for the fault cycle [t F ,t F +t0], the rest of the time periods are recovery periods;

[0010] In each fault cycle, a submodule is randomly calibrated as the target submodule and controlled to simulate an open circuit fault to obtain fault data.

[0011] Among them, in the fault period [t F ,t F +t0], the process of generating and collecting upper tube open circuit fault data is as follows:

[0012] Step a1, when the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f ,T2=S f '; where x is the phase of the bridge arm, x = a, b, c, y is the y bridge arm of the x phase, y = p, n, p represents the upper bridge arm, n represents the lower bridge arm; when S f = 0 or 1, n xy =n xy0 , where n xy is the actual command signal corresponding to the number of bridge arm submodules put into operation, n xy0The reference command signal corresponding to the number of bridge arm submodules generated by the bridge arm control; S f It is the output command signal of the target submodule;

[0013] When the bridge arm current i xy ≤0, the upper and lower switch signals of the target submodule are: T1=0, T2=S f ';When S f = 0, n xy =n xy0 ; When S f =1, n xy =n xy0 +1;

[0014] Step a2, respectively collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

[0015] Among them, in the fault period [t F ,t F +t0], the process of generating and collecting the lower tube open circuit fault data is as follows:

[0016] Step b1, when the bridge arm current i xy When ≤0, the upper and lower switch signals of the target submodule are: T1=S f ,T2=S f ';When S f = 0 or 1, n xy =n xy0 ;

[0017] When the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f, T2=0; when S f = 0, n xy =n xy0 -1; when S f =1, n xy =n xy0 ;

[0018] Step b2, respectively collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

[0019] Among them, during the recovery cycle, the number of bridge arm submodules put into operation is always n xy =n xy0 , the switch instruction of the target submodule is always T1=S f ,T2=S f '.

[0020] Among them, each bridge arm collects the upper tube and lower tube open circuit fault data in turn, and the rotation period is not less than K0T0.

[0021] Among them, the starting point of the fault cycle is t F =t0+k T0 / K0, K0 is a positive integer, k=1,2,…,K0.

[0022] Among them, a submodule is randomly calibrated as a target submodule, and it is controlled to simulate an open circuit fault, including randomly calibrating a submodule as a target submodule, controlling it to simulate an upper tube open circuit fault when the bridge arm current is negative, and increasing the bridge arm submodule input to compensate for the fault submodule output; simulating a lower tube open circuit fault when the bridge arm current is positive, and reducing the bridge arm submodule input to compensate for the fault submodule output, so as to obtain fault data.

[0023] After adopting the above scheme, the present invention first sets a complete acquisition cycle, stipulates the fault cycle and recovery cycle within the acquisition cycle, and in different acquisition cycles, the fault cycle starts at different time points respectively; in the fault cycle, the selected target submodule simulates the upper tube open circuit fault when the bridge arm current is negative, and increases the bridge arm submodule input to compensate for the fault submodule output; simulates the lower tube open circuit fault when the bridge arm current is positive, and reduces the bridge arm submodule input to compensate for the fault submodule output; each acquisition cycle only simulates one fault. The present invention effectively solves the challenge that the amount of actual engineering fault data is small and difficult to obtain, and can obtain a large amount of fault data that meets the actual working conditions in a way that does not affect the normal operation of the modular multi-level converter, and realizes high reliability, low cost, and high adaptability of fault data generation and acquisition, which is conducive to improving the practicality, rapidity and accuracy of the data-based fault location method. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of the topological structure of a modular multi-level converter based on a half-bridge submodule used in the present invention;

[0025] Among them, (a) is the overall topology diagram, and (b) is the structural diagram of each module;

[0026] Figure 2 It is a schematic diagram of a collection cycle in a method for generating and collecting fault data of a modular multi-level converter, and a flow chart of a collection cycle setting method and a fault cycle setting method;

[0027] Figure 3It is the principle of switching submodules of the bridge arm of the modular multilevel converter during the fault period and the recovery period during the acquisition period;

[0028] Among them, (a) is a submodule switching method for simulating a submodule upper tube failure within a fault cycle, (b) is a submodule switching method for simulating a submodule lower tube failure within a fault cycle; and (c) is a submodule switching method within a recovery cycle. DETAILED DESCRIPTION

[0029] The technical solutions and beneficial effects of the present invention will be described in detail below with reference to the accompanying drawings.

[0030] Figure 1 It is a three-phase topology of the classic modular multilevel converter, with Figure 1 As shown in (a), each phase includes a symmetrically arranged upper bridge arm and a lower bridge arm. The upper bridge arm includes N half-bridge sub-modules and an inductor L connected in series from the positive pole to the midpoint. S The lower bridge arm includes N half-bridge sub-modules and an inductor L connected in series from the negative pole to the midpoint. S , the phase voltage of this phase is U dc , the phase current is I dc , let the voltage at the upper bridge arm be u ap , the bridge arm current is i ap , the voltage at the lower bridge arm is u an , the bridge arm current is i an ;like Figure 1 (b) shows the circuit structure diagram of the half-bridge submodule, which includes an upper tube, a lower tube, and a capacitor C. The upper tube switching signal is T1, and the lower tube switching signal is T2.

[0031] Figure 2 It is a schematic diagram of the collection cycle in the modular multilevel converter fault data generation and collection method.

[0032] like Figure 2 As shown in ②, during the normal operation of the modular multilevel converter, the fault data can be actively generated and collected based on the control without affecting the normal operation; among them, a complete cycle of generating and collecting fault data is the collection cycle T sam .

[0033] like Figure 2 As shown in ③, the acquisition cycle consists of 3 adjacent fundamental frequency cycles, with a duration of T sam =3T0, T0 is a fundamental frequency cycle, the value is 1 / f0, f0 is the fundamental frequency of the AC output voltage and voltage. The starting point t0 of the acquisition cycle is the zero-crossing point where the AC current of the first fundamental frequency cycle changes from negative to positive. The acquisition cycle includes a fault cycle and a recovery cycle. The fault cycle is 1 fundamental frequency cycle. In each acquisition cycle, except for the fault cycle [t F,t F +t0], the rest of the time periods are recovery periods.

[0034] like Figure 2 As shown in ④ and ⑤, the starting point of the fault cycle is t F The range of variation is [t0, t0+T0], and the value is t F =t0+kT0 / K0. K0 is a positive integer, ranging from [30,50]. k=1,2,…,K0.

[0035] like Figure 2 As shown in ①, each bridge arm collects the open circuit fault data of the upper and lower tubes of the submodule in turn, and the rotation period is not less than K0T0.

[0036] Figure 3 It is the submodule switching principle of the bridge arm of the modular multilevel converter in the fault cycle and the recovery cycle during the acquisition cycle. In each acquisition cycle, a target submodule is randomly calibrated, and the fault data is generated by the number of the submodule. The output command signal S of the target submodule f Generated by the bridge arm control signal, S f =0 or 1.

[0037] like Figure 3 As shown in (a), the failure period [t F ,t F +t0], the process of the modular multilevel converter actively generating and collecting the sub-module upper tube open circuit fault data is as follows:

[0038] Step 1: When the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f ,T2=S f '; where x is the phase of the bridge arm, x = a, b, c, y is the y bridge arm of the x phase, y = p, n, p represents the upper bridge arm, and n represents the lower bridge arm. f = 0 or 1, n xy =n xy0 , where n xy is the actual command signal corresponding to the number of bridge arm submodules put into operation, n xy0 It is the reference command signal for the number of bridge arm submodules generated by the bridge arm control. xy ≤0, the upper and lower switch signals of the target submodule are: T1=0, T2=S f ';When S f = 0, n xy =n xy0 ; When S f =1, n xy =n xy0 +1.

[0039] Step 2: Collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

[0040] like Figure 3 As shown in (b), the failure period [t F ,t F +t0], the process of the modular multilevel converter actively generating the open circuit fault data of the lower tube of the submodule is:

[0041] Step 1: When the bridge arm current i xy When ≤0, the upper and lower switch signals of the target submodule are: T1=S f ,T2=S f '. When S f = 0 or 1, n xy =n xy0 When the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f, T2=0; when S f = 0, n xy =n xy0 -1; when S f =1, n xy =n xy0 .

[0042] Step 2: Collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

[0043] like Figure 3 As shown in (c), during the recovery cycle, the number of bridge arm submodules put into operation is always n. xy =n xy0 , the switch instruction of the target submodule is always T1=S f ,T2=S f '.

[0044] It should be noted that the above is only an example of the embodiment of the present invention, and the data generation and collection method proposed in the present invention is applicable to a modular multi-level converter based on any sub-module, and any power electronic converter based on a cascade of multiple identical modules.

[0045] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0046] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for generating and collecting fault data of a modular multilevel converter, wherein the modular multilevel converter is a three-phase topology structure, each phase includes an upper bridge arm from a positive electrode to a connection point and a lower bridge arm from a connection point to a negative electrode, wherein: The upper bridge arm comprises N submodules and a first inductor connected in series from the positive electrode to the connection point, and the lower bridge arm comprises N submodules and a second inductor connected in series from the negative electrode to the connection point; each submodule comprises an upper tube, a lower tube and a capacitor; the method comprises: Each acquisition cycle T sam It is divided into a failure period and a recovery period, where the duration of the collection period is T sam =3T0, T0 is a fundamental frequency cycle; the starting point t0 of the acquisition cycle is the zero-crossing point where the AC current of the first fundamental frequency cycle changes from negative to positive; the duration of the fault cycle is 1 fundamental frequency cycle T0, and its starting point t F =[t0,t0+T0]; in each acquisition cycle, except for the fault cycle [t F ,t F +t0], the rest of the time periods are recovery periods; In each fault cycle, a submodule is randomly calibrated as the target submodule and controlled to simulate an open circuit fault to obtain fault data.

2. The method according to claim 1, characterized in that: During the fault period [t F ,t F +t0], the process of generating and collecting upper tube open circuit fault data is as follows: Step a1, when the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f ,T2=S f '; where x is the phase of the bridge arm, x = a, b, c, y is the y bridge arm of the x phase, y = p, n, p represents the upper bridge arm, n represents the lower bridge arm; when S f =0 or 1, n xy =n xy0 , where n xy is the actual command signal corresponding to the number of bridge arm submodules put into operation, n xy0 The reference command signal corresponding to the number of bridge arm submodules generated by the bridge arm control; S f It is the output command signal of the target submodule; When the bridge arm current i xy ≤0, the upper and lower switch signals of the target submodule are: T1=0, T2=S f ';When S f = 0, n xy =n xy0 ; When S f =1, n xy =n xy0 +1; Step a2, respectively collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

3. The method according to claim 1, characterized in that: During the fault period [t F ,t F +t0], the process of generating and collecting the lower tube open circuit fault data is as follows: Step b1, when the bridge arm current i xy When ≤0, the upper and lower switch signals of the target submodule are: T1=S f ,T2=S f ';When S f =0 or 1, n xy =n xy0 ; When the bridge arm current i xy >0, the upper and lower switch signals of the target submodule are: T1 = S f, T2=0; when S f = 0, n xy =n xy0 -1; when S f =1, n xy =n xy0 ; Step b2, respectively collect the fault cycle [t F ,t F +t0] the bridge arm current i xy , target submodule switch signal S f 、Target submodule capacitor voltage u Cf .

4. The method according to claim 1, characterized in that: During the recovery cycle, the number of bridge arm submodules put into operation is always n. xy =n xy0 , the switch instruction of the target submodule is always T1=S f ,T2=S f '.

5. The method according to claim 1, characterized in that: Each bridge arm collects the upper and lower tube open circuit fault data in turn, and the rotation period is not less than K0T0.

6. The method according to claim 1, characterized in that: The starting point of the fault cycle is t F =t0+kT0 / K0, K0 is a positive integer, k=1,2,…,K0.

7. The method according to claim 1, characterized in that: A submodule is randomly calibrated as a target submodule, and controlled to simulate an open circuit fault, including randomly calibrating a submodule as a target submodule, controlling it to simulate an upper tube open circuit fault when the bridge arm current is negative, and increasing the bridge arm submodule input to compensate for the fault submodule output; simulating a lower tube open circuit fault when the bridge arm current is positive, and reducing the bridge arm submodule input to compensate for the fault submodule output, so as to obtain fault data.

Images