A Reliability Calculation Method for Hybrid Converter Stations

Abstract

This application provides a method for calculating the reliability of a hybrid converter station, comprising: determining a first probability model based on a hybrid multilevel converter, wherein the first probability model is a probability model of the hybrid multilevel converter under various capacity states; determining a second probability model based on the grid-commutated converter, wherein the second probability model is a probability model of the first grid-commutated converter under various capacity states; constructing a reliability assessment model for the hybrid converter station based on the first probability model and the second probability model; and solving the reliability assessment model to determine the reliability of the hybrid converter station. The method proposed in this application not only significantly improves the accuracy of converter station reliability assessment but also simplifies the computational complexity in engineering applications, providing more comprehensive and reliable technical support for practical engineering.

Description

Technical Field

[0001] This application relates to the field of electronic equipment technology, and in particular to a method for calculating the reliability of a hybrid converter station. Background Technology

[0002] To promote the green transformation of the energy structure, it is necessary to accelerate the construction of large-scale wind and solar power bases in desert and Gobi regions. Although these areas possess abundant new energy resources, challenges in energy transmission technology, such as energy loss during long-distance transmission and difficulties in cross-regional resource dispatch, severely restrict the large-scale development and utilization of new energy. To address these constraints, ultra-high voltage, high-capacity, long-distance direct current (DC) transmission technology is urgently needed. The main technical solutions include line commutated converters (LCCs) and modular multilevel converters (MMCs).

[0003] Hybrid converter stations, as a crucial component of modern power systems, directly impact power transmission stability and reliability. To ensure long-term system operation, it's essential to assess the performance of converter stations under various fault modes and operating conditions, making reliability evaluation a critical issue. Since LCCs and MMCs exhibit distinct characteristics in technical performance and reliability, and their applicable scenarios differ significantly, the interaction between LCCs and MMCs during operation must be considered when evaluating the overall reliability of hybrid converter stations. This paper proposes a reliability evaluation model that comprehensively considers factors such as redundancy configuration and commutation failure by quantitatively analyzing their operating characteristics to accurately assess converter station reliability. Existing research on LCC reliability primarily employs analytical methods, including probability distribution methods and the frequency-duration (FD) method based on Markov processes, but neither fully considers commutation failure scenarios. Furthermore, some studies introduce virtual device of commutation failure (VDoCF) models to quantify the impact of the initial commutation failure, but fail to consider subsequent commutation failures. MMCs consist of a series of cascaded sub-modules with complex interactive coupling characteristics that are difficult to handle effectively using analytical methods. Monte Carlo simulation is the primary evaluation method. Within the Monte Carlo simulation framework, existing research has proposed reliability calculation methods from the perspectives of classical probability models and distribution functions. However, these analytical methods only consider single sub-modules and are not applicable to scenarios with mixed sub-modules. In studies focusing on mixed sub-modules, correlation functions are used to handle the correlation between sub-modules, achieving reliability assessments of mixed MMCs with and without redundant configurations. However, the selection criteria for these functions are insufficient, and the substitutability between different types of sub-modules is not considered. Summary of the Invention

[0004] To overcome the aforementioned technical deficiencies, this application provides a method for calculating the reliability of a hybrid converter station. To achieve the above objective, this application implements the following technical solution:

[0005] This application provides a reliability calculation method for a hybrid converter station, wherein the hybrid converter station includes a first subsystem and a second subsystem, the first subsystem being a hybrid multilevel converter and the second subsystem being a grid-commutated converter, comprising:

[0006] Based on the hybrid multilevel converter, a first probability model is determined, which is a probability model of the hybrid multilevel converter under each capacity state.

[0007] Based on the grid-commutated converter, a second probability model is determined. The second probability model is a probability model for each capacity state of the grid-commutated converter.

[0008] Based on the first probability model and the second probability model, a reliability assessment model for hybrid converter stations is constructed.

[0009] The reliability assessment model of the hybrid converter station is solved to determine the reliability of the hybrid converter station.

[0010] Optionally, determining the first probability model based on the hybrid multilevel converter includes:

[0011] Based on the hybrid multilevel converter, an optimal Copula function is determined, which is used to characterize the coupling relationship between the sub-modules of the hybrid multilevel converter;

[0012] Based on the optimal Copula function, the first probability model is determined.

[0013] Optionally, determining the optimal Copula function based on the hybrid multilevel converter includes:

[0014] Based on the hybrid multilevel converter, a first number of Copula correlation functions are selected;

[0015] The optimal Copula function is obtained by combining the first number of Copula-related functions.

[0016] Optionally, determining the first probabilistic model based on the optimal Copula function includes:

[0017] Based on the optimal Copula function, a reliability evaluation model for hybrid multilevel converters is constructed.

[0018] Based on the reliability assessment model of the hybrid multilevel converter, the first probability model is determined.

[0019] Optionally, the step of constructing a reliability assessment model for a hybrid multilevel converter based on the optimal Copula function includes:

[0020] Based on the optimal Copula function, a first reliability model for the hybrid multilevel converter is determined in the first case, considering the correlation of all sub-modules. The first case is that the half-bridge sub-modules and full-bridge sub-modules in the hybrid multilevel converter do not need to be replaced by each other.

[0021] Based on the optimal Copula function, a second reliability model for the hybrid multilevel converter is determined in the second case, considering the correlation of all submodules. In the second case, the half-bridge submodule and the full-bridge submodule in the hybrid multilevel converter need to be replaced by each other.

[0022] Based on the first reliability model and the second reliability model, a reliability model for a hybrid multilevel converter is constructed.

[0023] Optionally, determining the first probability model based on the hybrid multilevel converter reliability assessment model includes:

[0024] Based on the reliability assessment model of the hybrid multilevel converter, a first failure rate of the hybrid multilevel converter is determined. The first failure rate is the failure rate of a single submodule of the hybrid multilevel converter after considering the coupling and substitutability between submodules.

[0025] Based on the first failure rate, a first state transition diagram of the hybrid multilevel converter is constructed;

[0026] Based on the first state transition diagram, the first state transition matrix of the hybrid multilevel converter is determined.

[0027] Based on the first state transition matrix, a first probability model is determined.

[0028] Optionally, determining the second probabilistic model based on the grid commutator includes:

[0029] Based on the aforementioned power grid phase-commutator, a reliability assessment model for the power grid phase-commutator is constructed;

[0030] Based on the reliability assessment model of the power grid phase-commutator, a second probability model is determined.

[0031] Optionally, the reliability assessment model for the power grid commutator is one of the following:

[0032] Reliability assessment model for the rectifier side of a grid-commutated converter or reliability assessment model for the inverter side of a grid-commutated converter;

[0033] The step of constructing a reliability assessment model for the power grid phase-change converter based on the power grid phase-change converter includes:

[0034] Based on the aforementioned grid-commutated converter, a reliability assessment model for the inverter side of the grid-commutated converter is constructed.

[0035] The aforementioned reliability assessment model for the inverter side of the grid-commutated converter, based on the grid-commutated converter, includes:

[0036] Construct a model for calculating the probability of commutation failure;

[0037] Based on the grid commutation converter and the commutation failure probability calculation model, a reliability assessment model for the inverter side of the grid commutation converter is constructed.

[0038] Optionally, the construction process of the commutation failure probability calculation model is as follows:

[0039] Acquire comprehensive coverage of different fault modes of commutation failure;

[0040] Based on the different failure modes, the first influencing factor affecting the first commutation failure and the second influencing factor affecting subsequent commutation failures are determined.

[0041] Based on the first influencing factor and the second influencing factor, a sample space for the commutation process is established;

[0042] Based on the sample space, the shut-off angle under different influencing mechanisms is determined;

[0043] Based on the shut-off angle under the different influencing mechanisms, a commutation failure probability calculation model is constructed.

[0044] Optionally, determining the second probabilistic model based on the grid commutator reliability assessment model includes:

[0045] Based on the reliability assessment model of the grid commutator, the second failure rate and fault repair time of the grid commutator are determined, where the second failure rate is the failure rate of the grid commutator.

[0046] Based on the second failure rate and the fault repair time, a second state transition diagram of the power grid phase-commutation converter is constructed;

[0047] Based on the second state transition diagram, determine the second state transition matrix of the grid commutator;

[0048] Based on the second state transition matrix, the second probability model is determined.

[0049] This application has the following beneficial effects:

[0050] The method proposed in this application not only significantly improves the accuracy of converter station reliability assessment, but also simplifies the computational complexity in engineering applications, providing more comprehensive and reliable technical support for practical engineering.

[0051] In addition to the purposes, features, and advantages described above, this application has other purposes, features, and advantages. The application will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0052] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0053] Figure 1 This is a flowchart illustrating a method for calculating the reliability of a hybrid converter station provided in an embodiment of this application;

[0054] Figure 2 This is a main circuit topology diagram of the hybrid converter station provided in the embodiments of this application;

[0055] Figure 3 This is a schematic diagram of the main circuit of the hybrid multilevel converter provided in the embodiments of this application;

[0056] Figure 4(a) is a schematic diagram of the rectifier side of the grid commutator provided in an embodiment of this application, and Figure 4(b) is a schematic diagram of the rectifier side of the grid commutator provided in an embodiment of this application.

[0057] Figure 5 This is the first state transition diagram of the hybrid multilevel converter provided in the embodiments of this application;

[0058] Figure 6 This is the second state transition diagram of the grid phase-commutation converter provided in the embodiments of this application. Detailed Implementation

[0059] The embodiments of this application are described in detail below with reference to the accompanying drawings, but this application can be implemented in many different ways as defined and covered by the claims.

[0060] Therefore, in order to solve the above problems, such as Figure 1 As shown, this application proposes a reliability calculation method for hybrid converter stations, which is mainly applied to applications such as... Figure 2The hybrid converter station shown includes a first subsystem 1 and a second subsystem 2. The probability of the subsystems under different operating states is calculated to achieve mathematical equivalence. This allows the hybrid converter station, which is composed of different types of converter stations such as LCC and MMC, to be analyzed for overall reliability through mathematical formulas.

[0061] Here is a brief introduction to this hybrid converter station:

[0062] The first subsystem 1 is a hybrid multilevel converter (MMC). To more comprehensively analyze the interactions between submodules, this application selects, for example... Figure 3 The hybrid multilevel converter shown is based on a half-bridge submodule (HBSM) and a full-bridge submodule (FBSM).

[0063] The second subsystem 2 is a grid-commutated converter (LCC). The grid-commutated converter selected in this application can be distinguished according to the commutation direction. It can only be the rectifier side of the grid-commutated converter as shown in Figure 4(a) or the inverter side of the grid-commutated converter as shown in Figure 4(b). The difference between the inverter side and the rectifier side of the grid-commutated converter is that the inverter side of the grid-commutated converter has multiple commutation failure simulation modules connected in series.

[0064] Combination Figure 3 And Figures 4(a) and 4(b), here on Figure 1 The reliability calculation method for the hybrid converter station shown is explained in detail below:

[0065] Step S101: Based on the hybrid multilevel converter, determine the first probability model, which is the probability model of the hybrid multilevel converter under each capacity state;

[0066] Because single functions have limitations in fitting complex relationships, this application selects the first-order Copula related function under the Archimedesian Copula function family, for example, to more accurately describe the complex interactive coupling characteristics between submodules of a hybrid multilevel converter. N Using Gumbel, Clayton, and Frank functions as analytical tools, the first three correlation functions are combined with weights in a mathematical form. Then, the first number of Copula correlation functions are combined to obtain the optimal Copula function. The optimal Copula function is used to characterize the coupling relationships between submodules of the hybrid multilevel converter. The process of determining the optimal Copula function is as follows:

[0067] First, the maximum likelihood method is used to estimate the correlation coefficient of each Copula function. θ .

[0068] (1)

[0069] In the formula, C This represents a specific Copula function. n Number of data sets; N The number of submodules to be analyzed; F Let be the lifetime distribution function, and argmax represent finding the value that maximizes the log-likelihood function. θ Value. (For example: when analyzing 100 sets of cumulative distribution function data for 200 sub-modules, then n=100, N =200).

[0070] Secondly, calculate the weight coefficients of each copula function that makes up the optimal copula function. a i Each copula function generally includes empirical copula functions and theoretical copula functions. The formula for the empirical copula function is:

[0071] (2)

[0072] in, I (•) is an indicator function; Lifetime sample T 1,…, T N Rank statistics.

[0073] The Euclidean distance between the theoretical Copula function and the empirical Copula function is:

[0074] (3)

[0075] Then, choose the one that minimizes the Euclidean distance between the theoretical Copula function and the empirical Copula function. As the weighting coefficient of Copula.

[0076] Furthermore, by combining multiple Copula functions, the optimal Copula function for expressing the reliability of hybrid MMC is obtained, which can then be used to represent the relationships between selected submodules:

[0077] (4)

[0078] Among them, subscript Used to distinguish different types of Copula functions, and has:

[0079] (5)

[0080] Therefore, combining the results used in this article, we can obtain the following: N Using the Gumbel, Clayton, and Frank function forms, we obtain the optimal Copula function form as follows:

[0081] (6)

[0082] in, p equal to F ( T The number of ); These are the weights of the Gumbel, Clayton, and Frank functions, respectively. N The number of submodules to be analyzed, i.e., the distribution function. F The quantity.

[0083] After obtaining the optimal Copula function, a reliability assessment model for the hybrid multilevel converter can be constructed based on this optimal Copula function. Then, based on the hybrid multilevel converter reliability assessment model, the first probability model is determined. The process of constructing the hybrid multilevel converter reliability assessment model is as follows:

[0084] To ensure the reliability of hybrid multilevel converters (MMCs), redundant submodules are typically configured in practical engineering. This allows redundant submodules to replace the faulty ones if the initial configuration submodule fails, ensuring the normal operation of the MMC. Both the full-bridge submodule (FBSM) and the half-bridge submodule (HBSM) will cause MMC failure. Only when the FBSM and HBSM are in good working order can the MMC be reliably configured for normal operation. N F、 N H Only with this feature can the MMC function normally; and the FBSM redundant submodule with fault current interruption capability can replace the damaged HBSM.

[0085] about Figure 3 The reliability component modeling for a hybrid MMC (Multi-Module Configuration) with both full-bridge (FBSM) and half-bridge (HBSM) submodules is conducted in light of the more complex submodule replacement scenarios that arise during failures compared to a single MMC with only one type of submodule. Corresponding reliability analysis models will be established for each of these different operating modes.

[0086] First, based on the optimal Copula function, the first reliability model of the hybrid multilevel converter is determined in the first case, considering the correlation of all submodules. The first case assumes that the half-bridge submodules and full-bridge submodules in the hybrid multilevel converter do not need to be substituted for each other. The specific process is as follows:

[0087] When there is no need for substitution between FBSM and HBSM, the configuration for each bridge arm is as follows. N H + N 0H One and a half bridge modules, among which N H This represents the number of typical half-bridge sub-modules. N 0H This represents the number of redundant half-bridge submodules. The number of submodules currently in normal working order is set to [value missing]. j The submodule lifetime samples are arranged according to their working status. The normal submodules are: The fault submodule is: The reliability distribution function of the half-bridge submodule can be obtained as follows:

[0088] (7)

[0089] There exists a Copula This makes the joint distribution as:

[0090] (8)

[0091] In equation (8), p equal to F ( T The number of ).

[0092] By applying Sklar's theorem, equation (7) can be expressed as a Copula function composed of two joint distribution functions, which yields:

[0093] (9)

[0094] in, , , M The number of Monte Carlo samplings. for N The dimensional hybrid copula function is expanded as shown in equation (6).

[0095] The specific derivation of equation (9) is as follows:

[0096] (10)

[0097] "A Copula function expressed as two joint distribution functions" means that the two functions (min, max) in P subtracted in the second step can also be written as a Copula function.

[0098] Since the reliability distribution function of the aforementioned half-bridge submodule is the same as the reliability inference process of the full-bridge submodule under normal conditions, there also exists a similar reliability distribution function for the full-bridge submodule. Copula This makes the joint distribution as:

[0099] (11)

[0100] By expressing equation (11) as a Copula function composed of two joint distribution functions using Sklar's theorem, the reliability of the full-bridge submodule arm can be obtained:

[0101] (12)

[0102] The first reliability model for the hybrid multilevel converter, considering the correlation of all submodules, is as follows:

[0103] (13)

[0104] In the formula, This is the first reliability model.

[0105] Secondly, based on the optimal Copula function, a second reliability model for the hybrid multilevel converter is determined in the second case, considering the correlation of all submodules. The second case requires that the half-bridge submodules and full-bridge submodules in the hybrid multilevel converter need to be mutually replaceable. The specific process is as follows:

[0106] When FBSM and HBSM need to be interchanged, i.e., when the number of redundant half-bridge modules is insufficient to replace all faulty half-bridge sub-modules, the converter station needs to call upon the remaining redundant full-bridge modules to supplement them until the total number of operating half-bridge sub-modules is restored to the minimum number required for normal operation. In this case, the half-bridge sub-module is equivalent to the normally operating sub-module in Case 1. Substituting into equation (8), we can obtain the reliability distribution function of the half-bridge submodule as follows:

[0107] (14)

[0108] in, .

[0109] Following the first scenario, perform the same operation on the full-bridge submodule, and select... j F Each submodule is working normally, as follows: At the same time, the total number of full-bridge and half-bridge failures must not exceed the total redundancy to ensure that the replaced half-bridge can function normally. Therefore, the remaining spare submodules... N 0 varies with the number of faulty submodules. Similarly, the full-bridge submodule reliability formula can be obtained by calculating according to equation (8):

[0110] (15)

[0111] in, for j F Number of faulty submodules, remaining submodules .

[0112] Then, for the hybrid multilevel converter in the second case, considering the correlation of all submodules, the second reliability model is:

[0113] (16)

[0114] In the formula, This is the first reliability model.

[0115] Finally, a reliability evaluation model for the hybrid multilevel converter can be constructed based on the first and second reliability models:

[0116] (17)

[0117] In the formula, This is a reliability assessment model for hybrid multilevel converters.

[0118] After constructing the reliability assessment model for the hybrid multilevel converter, since the relationship between the reliability function and the failure rate follows an exponential change, the first failure rate of the hybrid multilevel converter can be determined. The first failure rate is the failure rate of a single submodule of the hybrid multilevel converter after considering inter-module coupling and replaceability. :

[0119] (18)

[0120] Changes in the state of a converter station are primarily triggered by events such as fault occurrence or repair completion. These dynamic processes are described using state variables, and parameters such as failure rate, repair rate, and installation rate are used to quantify the failure and recovery processes of sub-modules. Based on the state transition diagram, a system state transition rate matrix can be established. By combining this with solving a system of linear equations, the probability of the converter station in different states can be obtained, thereby enabling the assessment of the converter station's reliability.

[0121] Therefore, based on the first failure rate, a system can be constructed as follows: Figure 5The first state transition diagram of the hybrid multilevel converter is shown below. Based on this first state transition diagram, the first state transition matrix of the hybrid multilevel converter is determined as shown in the following equation:

[0122] (19)

[0123] in, N This represents the number of all submodules in one bridge arm of the MMC. For the recovery rate of a single submodule, For the installation rate of a single submodule ( (Refer to existing data).

[0124] After obtaining the first state transition matrix, the first probability model of equation (21) can be determined by combining the linear equations of equation (20). as follows:

[0125] (20)

[0126] in, This is the first probability model, which is the probability model for each capacity state of the hybrid multilevel converter. This refers to the current state of the converter station. P for Figure 5 The probability of each state.

[0127] (twenty one)

[0128] in, To calculate each operating state using equation (20) S 1 to The probability of being out of service in ( ) To calculate each operating state using equation (20) S 1 to The probability of it being in de-capacity operation. To calculate each operating state using equation (20) S 1 to The probability that it is in normal operation.

[0129] The number of sub-modules operating normally at the same available capacity level of the MMC converter station can be 100% capacity during normal operation, 50% capacity during reduced operation, and 0% capacity during shutdown. The corresponding number of sub-modules is determined according to the actual project. Based on the available capacity levels provided by LCC and MMC, the probabilities of each state calculated above are classified and added together, and then substituted into equation (44) to calculate the reliability of the hybrid converter station.

[0130] Step S102: Based on the grid-commutated converter, determine the second probability model, which is the probability model of the grid-commutated converter under each capacity state;

[0131] First, a reliability assessment model for the power grid phase-commutation converter is constructed based on the power grid phase-commutation converter. Then, a second probability model is determined based on the reliability assessment model for the power grid phase-commutation converter.

[0132] Grid-commutated converters can be categorized by their commutation direction, meaning they can only be classified as either the rectifier side or the inverter side. Therefore, the reliability assessment model for a grid-commutated converter can only be either a rectifier-side reliability assessment model or an inverter-side reliability assessment model. In other words, constructing a reliability assessment model for a grid-commutated converter involves two specific technical solutions: one is to construct a rectifier-side reliability assessment model, and the other is to construct an inverter-side reliability assessment model.

[0133] The process of constructing the reliability assessment model for the rectifier side of the power grid phase-commutation converter is as follows:

[0134] As shown in Figure 4(a), each pole of the LCC rectifier-side converter station is composed of series and parallel components. First, the series reliability parameters of the circuit breaker and transformer cascaded components are calculated for the positive pole. Then, the obtained parameters are calculated using the same method as the valve, and so on, to obtain the series reliability parameters of the four positive pole components. The formula is expressed as:

[0135] (twenty two)

[0136] in:

[0137] (twenty three)

[0138] in, The failure rate of the positive terminal of the rectifier side of the power grid commutator converter. The fault frequency of the positive terminal of the rectifier side of the power grid commutator is represented by . Together, these two parameters constitute the reliability assessment model for the positive terminal of the rectifier side of the power grid commutator. The subscripts a, b, c, and d refer to circuit breakers, transformers, valves, and reactors, respectively. The failure rate of the aforementioned components, This refers to the failure frequency of the aforementioned components. The repair time of each of the aforementioned components configured on the rectifier side is obtained based on actual engineering data.

[0139] Substituting the parameters of the circuit breaker, transformer, valve, and reactor on the negative terminal of the rectifier side into equations (22)-(23) above, we can obtain the failure rate of the negative terminal of the rectifier side of the power grid commutator. λ r负 Fault frequency of the negative terminal of the rectifier side of the power grid commutator Together, they constitute the negative electrode reliability assessment model for the rectifier side of the power grid commutator:

[0140] (twenty four)

[0141] Finally, since the positive and negative terminals of the rectifier side of the grid commutator are connected in parallel, the reliability assessment model for the rectifier side of the grid commutator can be obtained as follows:

[0142] (25)

[0143] in, For the failure rate of the rectifier side of the power grid commutator, This refers to the frequency of faults occurring on the rectifier side of the power grid's phase-commutation converter. To obtain the repair time of the positive and negative electrodes on the rectifier side, calculate according to the following formula:

[0144] (26)

[0145] in, These are the series repair times for two cascaded components: circuit breaker and transformer, and valve and reactor. for The series repair time of the four positive electrode components obtained by series calculation.

[0146] Similarly, substituting the parameters of the circuit breaker, transformer, valve, and reactor on the rectifier side negative terminal into equation (26), the repair time of the LCC negative terminal is:

[0147] (27)

[0148] in, These are the series repair times for the circuit breaker and transformer, and the valve and reactor, respectively, after substituting the negative pole parameters. for The series repair time of the four negative electrode components is calculated based on the series connection.

[0149] After a commutation failure occurs, the LCC will be forced to shut down. When constructing the inverter side reliability model, the fault simulation module is taken into consideration as a series element. The LCC inverter side converter station composition is shown in 4(b).

[0150] Figure 4(b) shows that the reliability parameter calculation for the LCC inverter side is the same as that in 4(a), except that an additional commutation failure simulation module is connected in series for both the positive and negative poles. The parameter calculation process will be explained in detail later. Finally, the positive pole reliability evaluation model of the grid commutation converter inverter side is obtained by the same calculation method as Equation (22):

[0151] (28)

[0152] in:

[0153] (29)

[0154] In the formula ， The failure rate of the positive terminal of the inverter side of the grid commutator converter. The frequency of faults occurring at the positive terminal of the inverter side of the grid commutator converter, along with the frequency of faults occurring at the positive terminal of the inverter side of the grid commutator converter, together constitute the reliability assessment model for the positive terminal of the inverter side of the grid commutator converter. (Subscript) These respectively refer to circuit breakers, transformers, valves, and reactors. The failure rate of the aforementioned components, This refers to the failure frequency of the aforementioned components. These are the respective repair times of the aforementioned components configured on the rectifier side, obtained based on actual engineering data. It is the failure rate of the commutation failure simulation module.

[0155] Similarly, by substituting the parameters of the circuit breaker, transformer, valve, reactor, and commutation failure simulation module of the inverter side negative terminal into equations (28)-(29), the failure rate of the inverter side negative terminal of the power grid commutation converter can be calculated. Fault frequency of the negative terminal of the inverter side of the grid commutator Together, they constitute the negative electrode reliability assessment model for the inverter side of the grid-commutated converter:

[0156] (30)

[0157] Finally, considering the parallel connection of the positive and negative terminals on the inverter side of the grid-commutated converter, the reliability assessment model for the LCC inverter side is obtained by the same calculation method as in equation (25):

[0158] (31)

[0159] in, For the failure rate of the inverter side of the grid commutator, The frequency of faults occurring on the inverter side of the power grid's phase-commutation converter. To obtain the repair times for the positive and negative electrodes on the inverter side, since the commutation failure simulation module is solely for quantifying fault probability and does not consider repair time, therefore... Similarly, by substituting the parameters of the circuit breaker, transformer, valve, and reactor on the positive and negative poles of the inverter side into formulas (26)-(27), the calculation is performed:

[0160] (32)

[0161] Similarly, the repair time is calculated by substituting the parameters of the circuit breaker, transformer, valve, and reactor on the negative side of the inverter into equation (32):

[0162] (33)

[0163] in, These are the series repair times for the circuit breaker and the two cascaded components of the transformer, valve and reactor after substituting the parameters of the negative pole on the inverter side; for The series repair time of the four negative electrode components is calculated based on the series connection.

[0164] The parameters of the aforementioned commutation failure simulation module are calculated based on the commutation failure probability calculation model. In other words, the grid commutation converter inverter-side reliability assessment model in this application is determined based on the grid commutation converter and the commutation failure probability calculation model. The construction process of the aforementioned commutation failure probability calculation model is as follows:

[0165] The main operating states of converter stations are symmetrical changes in three-phase voltage and asymmetrical faults. Both can result in successful or unsuccessful commutation, depending on the size of the turn-off angle and the minimum turn-off angle (in engineering applications, the minimum turn-off angle of the inverter is generally taken as 15° to 18°).

[0166] For symmetrical changes in three-phase voltage, the turn-off angle The calculation formula is:

[0167] (34)

[0168] in, For trigger angle, The electrical angle at which the voltage drop occurs; It is direct current; For commutation reactance; This is the effective value of the AC line voltage. The effective value of the voltage after the change. The time delay of the voltage change relative to the valve trigger moment is denoted as . ( The power frequency angular frequency, 100 π ).

[0169] For asymmetric faults, the shut-off angle... The calculation formula is:

[0170] (35)

[0171] in, A To determine the symmetrical component of the voltage waveform during commutation, take... ; B For asymmetric components, take .

[0172] Because the factors affecting commutation failure differ under different fault modes, namely the factors affecting the first commutation failure and subsequent commutation failures are different, we first obtain a comprehensive coverage of different fault modes that cover commutation failure, then identify the first influencing factor affecting the first commutation failure and the second influencing factor affecting subsequent commutation failures, consider the impact of these factors on the shut-off angle, and then determine whether commutation has failed based on the shut-off angle results, thereby accurately characterizing the commutation failure characteristics.

[0173] For the first commutation failure, the primary influencing factors can be summarized into four points: the magnitude of the commutation voltage drop, the rate of change of the commutation voltage magnitude, the time of the fault occurrence, and the DC current.

[0174] For subsequent commutation failures, the second influencing factor includes the factors affecting three-phase symmetry and asymmetrical faults, which differ in their influencing factors. The main causes of subsequent commutation failures under three-phase symmetry are excessively low commutation voltage amplitude and excessively high DC current. After the initial commutation failure, the LCC will use a low-voltage current limiting circuit; therefore, when analyzing subsequent commutation failures, the low-voltage current limiting strategy must also be considered, i.e.:

[0175] (36)

[0176] In the formula, DC voltage Current limit value All values ​​are expressed in per-unit format.

[0177] The main cause of asymmetric failure is the leading firing angle. The fluctuations contain both second-harmonic components and exhibit a regular fluctuation with a period of 6 commutations:

[0178] (37)

[0179] in, This represents the change in direct current. This is the DC component; The amplitude of the second harmonic component; The phase of the second harmonic component; The phase interval of adjacent phase voltages at the time of the fault. For the number of commutations, The power frequency angular frequency is equal to 100. π, For the first m The actual leading trigger angle of the second commutation. This is the initial value of the actual lead-out angle.

[0180] The above factors will affect the AC voltage amplitude in equations (34) and (35). U DC current I d These factors influence the shut-off angle result. The first influencing factor and the second influencing factor are established as a sample space for the commutation process. Different values ​​are taken under Monte Carlo simulation based on the nature of the influencing factors. Then, the shut-off angle under different influencing mechanisms is calculated by combining equations (34) and (35). The number of commutation failures during the entire Monte Carlo simulation is recorded to calculate the failure rate of the fault simulation module. .

[0181] (38)

[0182] in, The number of commutation failures. This represents the number of commutations, i.e., the number of Monte Carlo simulations.

[0183] By combining the obtained shut-off angle with equation (39), a commutation failure probability calculation model can be constructed. The numerator represents the case where the shut-off angle is less than the minimum shut-off angle, i.e., the commutation failure case; the denominator is all the calculated commutation cases. This index reflects the relative degree of commutation failure, rather than a simple proportion of failures. Even if a certain extinguishing angle is slightly less than the minimum extinguishing angle, it will be included in the calculation, reflecting the degree of commutation failure and providing a more detailed assessment.

[0184] (39)

[0185] in, For the arc extinguishing angle margin, For the minimum shut-off angle, if If the phase commutation is successful, then the phase commutation is successful. If the phase commutation fails, then the commutation will fail. Indicates the number of Monte Carlo simulations. .

[0186] Furthermore, in order to achieve a quantitative assessment of the risk of commutation failure, the probability result obtained from equation (39) is transformed into indicators of failure frequency and duration, thereby establishing a commutation failure simulation module.

[0187] (40)

[0188] in, The fault frequency of the commutation failure simulation module. The probability and average duration of commutation failure are given.

[0189] After obtaining the above reliability assessment model for the grid-commutated converter, the second probabilistic model can be determined. Since the state transitions on the rectifier side and inverter side of the LCC are the same, the resulting state transition space diagrams can be used... Figure 6 As shown, the resulting state transition rate matrix also has the same form. Taking the LCC rectifier side as an example, its determination process is explained as follows:

[0190] The second failure rate of the grid-commutated converter can be obtained from the above formula (22), that is, the failure rate of the rectifier side of the grid-commutated converter. Fault repair time It is based on the repair time of the positive and negative electrodes on the rectifier side. The process, obtained by integrating in parallel, is as follows:

[0191] (41)

[0192] Based on the second failure rate and the failure repair time, the second state transition diagram of the power grid commutator can be constructed, such as... Figure 6 As shown.

[0193] Then, based on the second state transition diagram, the second state transition matrix of the grid commutator is determined as shown in equation (42):

[0194] (42)

[0195] in, , ， The installation rate of the LCC rectifier side (refer to existing data).

[0196] Based on the aforementioned second state transition matrix, and combined with the linear equation system (Equation 20), the second probability model can be obtained. The second probability model is shown below, representing the probability models for each capacity state of the grid commutator:

[0197] (43)

[0198] in, To calculate each operating state using equation (20) S 1 to The probability of being in a bipolar shutdown state. To calculate each operating state using equation (20) S 1 to The probability of being in unipolar operation in ) To calculate each operating state using equation (20) S 1 to The probability of being in bipolar operation.

[0199] LCC converter stations have three available capacity levels: 100% capacity for bipolar operation, 50% capacity for unipolar operation, and 0% capacity for bipolar shutdown. The above calculations are based on the available capacity levels provided by the LCC for each state. S 1 to The probabilities under each category are summed to obtain the second probability model. .

[0200] Step S103: Based on the first probability model and the second probability model, construct a reliability assessment model for the hybrid converter station;

[0201] Considering the connections between subsystems, the operating state probabilities of all subsystems at different capacity levels are integrated using the convolution method of equation (40) to obtain the operating state probabilities of the entire hybrid converter station at different capacity levels.

[0202] (44)

[0203] in, This indicates that subsystems 1 and 2 are connected in series and parallel, respectively. For subsystems 1 and 2 at their respective capacity levels The probability of the running state under the given conditions, and ; This refers to the capacity level of the converter station after subsystem integration. The probability of non-zero values ​​was used as a reliability index for hybrid converter stations, thus establishing a reliability model for hybrid converter stations.

[0204] Step S104: Solve the reliability assessment model of the hybrid converter station to determine the reliability of the hybrid converter station.

[0205] The reliability of the hybrid converter station can be obtained by solving equation (44) above.

[0206] In summary, the method proposed in this application constructs a representation model of the interrelationships between MMC system submodules based on the optimal combination of multiple correlation functions; simultaneously, considering subsequent commutation failures in the LCC, a multi-scenario probabilistic model is constructed. By systematically integrating two types of converter reliability models, a multi-state reliability assessment system for hybrid converter stations is built. This method not only significantly improves the accuracy of converter station reliability assessment but also simplifies the computational complexity in engineering applications, providing more comprehensive and reliable technical support for practical engineering.

[0207] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for calculating the reliability of a hybrid converter station, wherein the hybrid converter station comprises a first subsystem and a second subsystem, the first subsystem being a hybrid multilevel converter and the second subsystem being a grid-commutated converter; characterized in that, include: Based on the hybrid multilevel converter, a first probability model is determined, which is a probability model of the hybrid multilevel converter under each capacity state. Based on the grid-commutated converter, a second probability model is determined. The second probability model is a probability model for each capacity state of the grid-commutated converter. Based on the first probability model and the second probability model, a reliability assessment model for hybrid converter stations is constructed. Solve the reliability assessment model of the hybrid converter station to determine the reliability of the hybrid converter station; The step of determining the first probability model based on the hybrid multilevel converter includes: Based on the hybrid multilevel converter, a first number of Copula correlation functions are selected; The first number of Copula related functions are combined to obtain the optimal Copula function, which is used to characterize the coupling relationship between the sub-modules of the hybrid multilevel converter. Based on the optimal Copula function, the first probability model is determined; The determination of the second probability model based on the grid commutator includes: Based on the aforementioned grid-commutated converter, a reliability assessment model for the grid-commutated converter is constructed; the reliability assessment model for the grid-commutated converter is one of either a rectifier-side reliability assessment model or an inverter-side reliability assessment model. Based on the aforementioned power grid commutator reliability assessment model, a second probability model is determined; The step of constructing a reliability assessment model for the power grid phase-change converter based on the power grid phase-change converter includes: Based on the aforementioned power grid phase-commutator, a reliability assessment model for the rectifier side of the power grid phase-commutator is constructed; The step of constructing a reliability assessment model for the power grid phase-change converter based on the power grid phase-change converter includes: Construct a model for calculating the probability of commutation failure; Based on the grid commutation converter and the commutation failure probability calculation model, a reliability assessment model for the inverter side of the grid commutation converter is constructed.

2. The method according to claim 1, characterized in that, The determination of the first probabilistic model based on the optimal Copula function includes: Based on the optimal Copula function, a reliability evaluation model for hybrid multilevel converters is constructed. Based on the reliability assessment model of the hybrid multilevel converter, the first probability model is determined.

3. The method according to claim 2, characterized in that, The reliability assessment model for hybrid multilevel converters, based on the optimal Copula function, includes: Based on the optimal Copula function, a first reliability model for the hybrid multilevel converter is determined in the first case, considering the correlation of all sub-modules. The first case is that the half-bridge sub-modules and full-bridge sub-modules in the hybrid multilevel converter do not need to be replaced by each other. Based on the optimal Copula function, a second reliability model for the hybrid multilevel converter is determined in the second case, considering the correlation of all submodules. In the second case, the half-bridge submodule and the full-bridge submodule in the hybrid multilevel converter need to be replaced by each other. Based on the first reliability model and the second reliability model, a reliability model for a hybrid multilevel converter is constructed.

4. The method according to claim 2, characterized in that, The determination of the first probability model based on the reliability assessment model of the hybrid multilevel converter includes: Based on the reliability assessment model of the hybrid multilevel converter, a first failure rate of the hybrid multilevel converter is determined. The first failure rate is the failure rate of a single submodule of the hybrid multilevel converter after considering the coupling and substitutability between submodules. Based on the first failure rate, a first state transition diagram of the hybrid multilevel converter is constructed; Based on the first state transition diagram, the first state transition matrix of the hybrid multilevel converter is determined. Based on the first state transition matrix, a first probability model is determined.

5. The method according to claim 1, characterized in that, The construction process of the commutation failure probability calculation model is as follows: Acquire comprehensive coverage of different fault modes of commutation failure; Based on the different failure modes, the first influencing factor affecting the first commutation failure and the second influencing factor affecting subsequent commutation failures are determined. Based on the first influencing factor and the second influencing factor, a sample space for the commutation process is established; Based on the sample space, the shut-off angle under different influencing mechanisms is determined; Based on the shut-off angle under the different influencing mechanisms, a commutation failure probability calculation model is constructed.

6. The method according to claim 1, characterized in that, The determination of the second probability model based on the reliability assessment model of the power grid commutator includes: Based on the reliability assessment model of the grid commutator, the second failure rate and fault repair time of the grid commutator are determined, where the second failure rate is the failure rate of the grid commutator. Based on the second failure rate and the fault repair time, a second state transition diagram of the power grid phase-commutation converter is constructed; Based on the second state transition diagram, determine the second state transition matrix of the grid commutator; Based on the second state transition matrix, the second probability model is determined.

Images