CN108509674B

CN108509674B - Improved hybrid MMC (modular multilevel converter) operation reliability evaluation model and method

Info

Publication number: CN108509674B
Application number: CN201810115615.7A
Authority: CN
Inventors: 余娟; 冯斐; 萨拉赫·卡莫
Original assignee: Chongqing University
Current assignee: Chongqing University
Priority date: 2018-02-06
Filing date: 2018-02-06
Publication date: 2021-10-26
Anticipated expiration: 2038-02-06
Also published as: CN108509674A

Abstract

The invention discloses an improved hybrid MMC operation reliability evaluation model and method based on multi-time scale heat damage, which mainly comprise the following steps: 1) and acquiring real-time data of the wind power transmission system. The real-time data mainly comprises wind speed, air temperature and equipment electrical parameters. 2) And establishing a reliability evaluation model of the hybrid MMC power device by utilizing the running characteristics of the hybrid MMC and a Miner's damage theory according to the original data. The power device of the wind power transmission system mainly comprises an insulated gate bipolar transistor IGBT and a crystal Diode. 3) And establishing a capacitor reliability evaluation model considering the voltage sharing of the fault SM according to the influence of the fault SM on the perfect SM capacitor. 4) And analyzing the loss distribution of the hybrid MMC device, thereby establishing an improved hybrid MMC operation reliability evaluation model considering SM multi-state. The method can be widely applied to the operation reliability evaluation of the hybrid MMC in the wind power and other renewable energy transmission grid connection.

Description

Improved hybrid MMC (modular multilevel converter) operation reliability evaluation model and method

Technical Field

The invention relates to the field of wind power transmission, in particular to an improved hybrid MMC operation reliability evaluation model and method based on multi-time scale heat damage.

Background

The modular multilevel converter type high-voltage direct-current transmission (MMC-HVDC) is a power transmission mode which has great prospect for realizing the connection of a large-scale wind power station and a power grid. The Modular Multilevel Converter (MMC) has the characteristics of modular design, high output voltage quality and the like, is key equipment in the VSC-HVDC, and the operational reliability of the Modular Multilevel Converter (MMC) is related to the safe and stable operation of the VSC-HVDC. However, the MMC has large transmission power variation due to random fluctuation of wind speed, and junction temperature amplitude and fluctuation variation of a power device in the MMC are obvious due to power and air temperature fluctuation, so that the reliability of the MMC is greatly influenced. Meanwhile, along with the increasing of wind power transmission power, the voltage level of the MMC rises, MMC devices are increased gradually, and the influence of power frequency, switching frequency and other electrical parameters on the operation reliability of the MMC is increased due to the increase of the MMC devices. However, conventional equipment reliability evaluation based on statistical data ignores the influence of current operating conditions on equipment reliability, resulting in a large deviation of the reliability evaluation result from actual operation. Therefore, the influence of environmental factors and electrical parameters needs to be considered, and an MMC operation reliability evaluation model in the wind power transmission system needs to be researched.

In recent years, many researches are made on MMC reliability research at home and abroad, but MMC reliability is evaluated based on a device constant fault rate model, or the fault rate of an MMC device is corrected by introducing a voltage correction factor. However, since the failure rate of the power device is time-varying and does not satisfy the basic condition of the constant failure rate model, it is necessary to further study the MMC operation reliability evaluation model based on the failure mechanism of the power device. In engineering applications, the hybrid MMC becomes a practical application scheme at present due to its good dc fault ride-through capability and low cost.

Disclosure of Invention

The present invention is directed to solving the problems of the prior art.

The technical scheme adopted for achieving the purpose of the invention is that the improved hybrid MMC operation reliability evaluation model based on multi-time scale heat damage mainly comprises the following steps:

1) and acquiring real-time data of the wind power transmission system. The real-time data mainly comprises wind speed, air temperature and equipment electrical parameters.

2) And establishing a reliability evaluation model of the hybrid MMC power device by utilizing the running characteristics of the hybrid MMC and a Miner's damage theory according to the original data. The power device of the wind power transmission system mainly comprises an insulated gate bipolar transistor IGBT and a crystal Diode.

Further, the main steps of establishing the reliability evaluation model of the hybrid MMC power device are as follows:

2.1) calculating the multi-time scale junction temperature of the mixed MMC power device. The method mainly comprises the following steps:

2.1.1) determining the output power P of the fan of the wind power transmission system_WToutAnd a sampling time t_nCorresponding to wind speed V_tnThe relationship (2) of (c).

Output power P of ith fan of wind power transmission system_WTout,iAnd a sampling time t_nCorresponding to wind speed V_tnThe relationship of (a) is as follows:

in the formula, P_ratedThe rated power of the fan. V_cutin、V_ratedAnd V_cutoutRespectively cut-in wind speed, rated wind speed and cut-out wind speed. k is a radical of_pRepresenting the coefficients related to air density and fan area. N is a radical of_WTThe total number of the fans in the wind power transmission system. n is the total number of samples up to the current operating time. t is t_nIs the sampling instant. V_tnIs the wind speed. i is any fan. 1, …, N_WT。

Transport power P of hybrid MMC_MMCoutAs follows:

in the formula, N_WTThe total number of the fans in the wind power transmission system. i is any fan. 1, …, N_WT。P_WTout,iAnd outputting power for the ith fan.

2.1.2) calculating to obtain the mixed MMC bridge arm current according to the formula 1.

A phase upper bridge arm current I_auAnd phase A lower bridge arm current I_adRespectively as follows:

in the formula I_dcIs a direct side current. I is_mIs the ac side current peak.

Is the power factor angle. f. of₀Is the fundamental frequency.

Direct side current I_dcAs follows:

in the formula of U_dcIs the dc side voltage. P_MMCoutIs the transport power of the hybrid MMC.

Peak value of AC side current I_mAs follows:

in the formula, P_MMCoutIs the transport power of the hybrid MMC.

Is the power factor angle. U shape_acThe effective value of the AC side voltage.

2.1.3) calculating the average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A.

Average loss of jth switching period of IGBT (insulated Gate Bipolar transistor) of upper bridge arm of A phase

As follows:

in the formula, R_ce、U_ceoAnd τ_TThe IGBT forward on-resistance, the threshold voltage and the duty cycle, respectively. a is_T、b_TAnd c_TFor simulating IGBT dynamic characteristic curveAnd (5) synthesizing parameters. U shape_ratedThe rated voltage of the IGBT. f. of_swAnd ρ_TThe switching frequency and the temperature coefficient of the IGBT, respectively. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle. I is_auIs the phase A upper bridge arm current. U shape_dcIs the dc side voltage.

Number of switching cycles n_swAs follows:

n_sw＝f_sw/f₀。 (7)

in the formula (f)_swThe switching frequency of the IGBT. f. of₀Is the fundamental frequency.

2.1.4) calculating the average loss of the j th switching period of the Diode

Average loss of the j-th switching period of the Diode

As follows:

in the formula: r_d、U_dAnd τ_DRespectively, Diode forward on-resistance, threshold voltage, and duty cycle. a is_D、b_DAnd c_DParameters were fitted to the Diode dynamics curve. U shape_DIs a Diode rated voltage. f. of_dwAnd ρ_DRespectively, the switching frequency and the temperature coefficient of the Diode. I is_auIs the phase A upper bridge arm current. U shape_dcIs the dc side voltage.

2.1.5) calculating the junction temperature of the jth switching period of the IGBT in the hybrid MMC

The junction temperature

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nThe corresponding air temperature.

The temperature difference of an RC parallel unit in the jth switching period IGBT junction-shell heat network of the ith order. r represents an arbitrary order. r is 1,2,3, 4.

The temperature difference of the RC parallel unit in the jth switching period IGBT shell-heat sink thermal network.

The temperature difference of the RC parallel unit in the jth switching period IGBT heat sink-environment thermal network. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

Temperature difference of RC parallel unit in jth switching period IGBT junction-shell network of nth order

As follows:

in the formula, R_Tjc,rIs the thermal resistance. Tau is_Tjc,rIs the thermal resistance R_Tjc,rThermal time constant of (2). T is_swIs a switching cycle.

Is the temperature difference of the RC parallel unit in the j-1 switching period IGBT junction-shell heat network of the r-th order.e is the base of the natural log function and is about 2.71828. .

Average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

Temperature difference of RC parallel unit in jth switching period IGBT shell-heat sink network

As follows:

in the formula, R_TchIs the thermal resistance. Tau is_TchIs the thermal resistance R_TchThermal time constant of (2). T is_swIs a switching cycle.

Is the temperature difference of the RC parallel unit in the j-1 switching period IGBT shell-heat sink thermal network. e is the base of the natural log function and is about 2.71828. .

The jth switching period IGBT is the temperature difference of RC parallel units in the radiating fin-environment network

As follows:

in the formula, R_haIs the thermal resistance. Tau is_haIs the thermal resistance R_haThermal time constant of (2). T is_swIs a switching cycle.

Is the cycle temperature difference of the RC parallel unit in the j-1 th switch IGBT thermal network. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

Average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A.

Average loss is the j-th switching period of the A-phase upper bridge arm Diode.

2.1.6) calculating the junction temperature of the j-th switching cycle of the Diode in the hybrid MMC

The junction temperature

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nThe corresponding air temperature.

The temperature difference of an RC parallel unit in a j-th switching period Diode junction-shell heat network of an r-th step. r represents an arbitrary order. r is 1,2,3, 4.

For the temperature of the RC parallel unit in the j switching period Diode shell-fin heat networkAnd (4) poor.

The temperature difference of the RC parallel unit in the jth switching period IGBT heat sink-environment thermal network.

j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

Temperature difference of RC parallel unit in jth switching period Diode junction-shell network

As follows:

in the formula, R_Djc,rIs the thermal resistance. Tau is_Djc,rIs the thermal resistance R_Djc,rThermal time constant of (2). T is_dwIs a switching cycle.

Is the temperature difference of the RC parallel unit in the j-1 switching period Diode junction-shell heat network. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

Temperature difference of RC parallel unit in jth switching period Diode shell-fin network

As follows:

in the formula, R_DchIs the thermal resistance. Tau is_DchIs the thermal resistance R_DchHot time ofA constant. T is_swIs a switching cycle.

Is the temperature difference of the RC parallel unit in the j-1 switching period Diade shell-fin heat network.

Average loss is the j-th switching period of the A-phase upper bridge arm Diode. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle.

2.2) hybrid MMC power device multi-time scale reliability evaluation. The method mainly comprises the following steps:

2.2.1) calculating the cycle failure period number N corresponding to the low-frequency period of the IGBT_{Tf_L}. Number of cycles to failure N_{Tf_L}As follows:

in the formula, t_onThe heating time is shown. U is 0.01 times the block voltage of the module. D is the diameter of the aluminum bonding wire. k is 9.3 × 10¹⁴。β₁＝-4.416。β₂＝1285。β₃＝-0.463。β₄＝-0.716。β₅＝-0.761。β₆＝-0.5。T_{Tjmax_L}And the maximum value of the low-frequency junction temperature of the IGBT. T is_{Tjmin_L}And the minimum value of the low-frequency junction temperature of the IGBT. I is_LThe effective value of the current of the IGBT low-frequency aluminum bonding wire is obtained. e is the base of the natural log function and is about 2.71828. . Cycle failure period number N corresponding to IGBT fundamental frequency period_{Tf_F}As follows:

in the formula, t_onThe heating time is shown. U is 0.01 times the block voltage of the module. D is the diameter of the aluminum bonding wire. k is 9.3 × 10¹⁴。β₁＝-4.416。β₂＝1285。β₃＝-0.463。β₄＝-0.716。β₅＝-0.761。β₆＝-0.5。T_{Tjmax_F}And the maximum value of the junction temperature of the IGBT fundamental frequency. T is_{Tjmin_F}And the minimum value of the IGBT fundamental frequency junction temperature is obtained. I is_FThe effective value of the current of the IGBT fundamental frequency aluminum bonding wire is obtained. e is the base of the natural log function and is about 2.71828. .

According to Miner's damage theory and formula (17), IGBT is in 0-t_nTime of day life consumption CL_T(t_n) As follows:

in the formula, N_{Tsum_L}Is 0-t_nThe total number of low frequency thermal cycles at that time. N is a radical of_{T_L,g}And N_{Tf_L,g}The number of thermal cycles and the number of cycle failure cycles corresponding to the g-th low-frequency thermal cycle of the IGBT are respectively. N is a radical of_{T_F,q}And N_{Tf_F,q}And respectively obtaining the fundamental frequency thermal cycle times and the cycle failure cycles corresponding to the q-th sampling time interval T of the IGBT. n is the total number of samples up to the current operating time. N is a radical of_{Tsum_L}Is the total number of low frequency thermal cycles.

IGBT sampling time t_nMean time to failure MTTF corresponding to time interval T_T(t_n) As follows:

in the formula, CL_T(t_n) Is IGBT at 0-t_nThe lifetime of the moment is consumed. T is the sampling time interval.

IGBT failure rate lambda_T(t_n) As follows:

in the formula, MTTF_T(t_n) For the IGBT sampling time t_nCorresponding to the average time to failure of time interval T.

IGBT reliability R_T(tn)As follows:

in the formula, λ_T(t_n) Is the IGBT failure rate. t is t_nIs the sampling instant.

2.2.2) calculating the cycle failure number N corresponding to the low-frequency period of the Diode_{Df_L}. Number of cycles to failure N_{Df_L}As follows:

in the formula, t_onThe heating time is shown. U is 0.01 times the block voltage of the module. D is the diameter of the aluminum bonding wire. k is 9.3 × 10¹⁴。β₁＝-4.416。β₂＝1285。β₃＝-0.463。β₄＝-0.716。β₅＝-0.761。β₆＝-0.5。T_{Djmax_L}Is the Diode low frequency junction temperature maximum. T is_{Djmin_L}Is the Diode low frequency junction temperature minimum. I is_DLThe effective value of the current of the Diode low-frequency aluminum bonding wire is obtained.

Number of cycle failure cycles N corresponding to the period of the Diode fundamental frequency_{Df_F}As follows:

in the formula, t_onThe heating time is shown. U is 0.01 times the block voltage of the module. D is the diameter of the aluminum bonding wire. k is 9.3 × 10¹⁴。β₁＝-4.416。β₂＝1285。β₃＝-0.463。β₄＝-0.716。β₅＝-0.761。β₆＝-0.5。T_{Djmax_F}Is the Diode fundamental frequency junction temperature maximum. T is_{Djmin_F}Is the Diode fundamental frequency junction temperature minimum. I is_DFThe effective value of the current of the Diode fundamental frequency aluminum bonding wire is obtained.

According to Miner's theory of injuryEquation (23), the Diade is between 0 and t_nTime of day life consumption CL_D(t_n) As follows:

in the formula, N_{Dsum_L}Is a Diode at 0-t_nTotal number of low frequency thermal cycles at the moment. N is a radical of_{D_L,α}And N_{Df_L,α}The number of thermal cycles and the number of cycle failure cycles corresponding to the alpha-th low-frequency thermal cycle of the Diode are respectively. N is a radical of_{D_F,ω}And N_{Df_F,ω}The thermal cycle number of the fundamental frequency and the cycle failure cycle number corresponding to the omega th sampling time interval T are respectively. n is the total number of samples up to the current operating time.

The Diode sampling time t_nMean time to failure MTTF corresponding to time interval T_D(t_n) As follows:

in the formula, CL_D(t_n) Is a Diode at 0-t_nThe lifetime of the moment is consumed. T is the sampling time interval.

Diade failure rate λ_D(t_n) As follows:

in the formula, MTTF_D(t_n) For the moment t of the Diade sampling_nCorresponding to the average time to failure of time interval T.

Diade reliability R_D(tn) is as follows:

in the formula, λ_D(t_n) Is the Diode failure rate. e is a natural logarithmic functionAbout 2.71828. . t is t_nIs the sampling instant.

3) And establishing a hybrid MMC capacitor reliability evaluation model considering the voltage sharing of the fault SM according to the influence of the fault SM on the perfect SM capacitor.

Further, the main steps of establishing the reliability evaluation model of the hybrid MMC capacitor are as follows:

3.1) calculating the capacitor reliability R_C(t_n) Namely:

in the formula, λ_cIs the capacitor failure rate. W_s0Is in an initial state. W_siSelf-healing energy corresponding to i SM faults. t is t_nIs the sampling instant.

Self-healing energy W_siAs follows:

U_Cicapacitor voltage for i SM faults. R_CIs a sheet resistance. C is capacitance. And f (P) is a sandwich pressure related function. k is a radical of_cTo calculate the correlation coefficient of the self-healing energy. a is the capacitor correlation coefficient. And b is the resistance correlation coefficient.

3.2) according to the capacitor reliability R_C(t_n) The hybrid MMC capacitor reliability was evaluated.

4) And analyzing the loss distribution of the hybrid MMC device, thereby establishing an improved hybrid MMC operation reliability evaluation model considering SM multi-state.

Further, the main steps of establishing the hybrid MMC operation reliability evaluation model considering SM multi-state are as follows:

4.1) calculating the reliability of the SM multi-state. The SM main circuit consists of an IGBT, a Diode and a capacitor.

According to the combination relation of IGBT, Diode and capacitor in SM, CSSM reliability R under intact state_CSu(t_n) As follows:

R_CSu(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·R_T3(t_n)·R_D3(t_n)·R_C(t_n)。 (30)

in the formula, R_T1(t_n)、R_T2(t_n) And R_T3(t_n) Reliability of the first IGBT, the second IGBT and the third IGBT, respectively. R_D1(t_n)、R_D2(t_n) And R_D3(t_n) Reliability of the first, second and third Diode, respectively. R_c(t_n) The reliability of the capacitor.

HBSM reliability R in good condition_HBu(t_n) As follows:

R_HBu(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·R_C(t_n)。 (31)

Reliability R of CSSM under half fault condition_CSp(t_n) As follows:

R_CSp(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·(1-R_T3(t_n)·R_D3(t_n))·R_C(t_n)。 (32)

in the formula, R_T1(t_n)、R_T2(t_n) And R_T3(t_n) Reliability of the first IGBT, the second IGBT and the third IGBT, respectively. R_D1(t_n)、R_D2(t_n) And R_D3(t_n) Reliability of the first, second and third Diode, respectively. Rc (t)_n) The reliability of the capacitor.

4.2) accounting for SM multistate hybrid MMC reliability

The reliability of the bridge arm is mainly divided into two cases.

The first case is: when neither CSSM nor HBSM damaged number exceeds its own redundancy number, i_CS≤M_CSAnd i is_HB≤M_HBMeanwhile, the hybrid MMC normally operates.

According to a binomial distribution probability formula, in the first case, the total CSSM reliability R of the upper bridge arm_CSs1(t_n) As follows:

in the formula, N_CSIs the baseline number of CSSM. R_CSu(t_n) The CSSM reliability in good condition. i.e. i_csIs the number of CSSM damages. M_CSIs the redundancy number of the CSSM.

In the first case, the total HBSM reliability R of the upper bridge arm_HBs1(t_n) As follows:

in the formula, N_HBThe baseline number for the HBSM. R_HBu(t_n) The HBSM reliability in the intact state. i.e. i_HBIs the number of HBSM breakages. M_HBIs the redundancy number of the HBSM.

Based on the series relation between CSSM and HBSM in bridge arm, the reliability R of single-phase upper bridge arm under the first condition_{Arm_u1}(t_n) As follows:

R_{Arm_u1}(t_n)＝R_CSs1(t_n)·R_HBs1(t_n)。 (35)

in the formula, R_HBs1(t_n) The total HBSM reliability for the upper arm in the first case. R_CSs1(t_n) The total CSSM reliability for the upper leg in the first case.

The second case is: when the CSSM defect number does not exceed the self-redundancy number, the HBSM defect number exceeds the self-redundancy number, and the good and half-fault CSSMs replace the damaged HBSM, i.e., i_CS≤M_CSAnd M is_HB<i_HB≤M_HB+M_CS-(i_CS-i_CSp) Meanwhile, the hybrid MMC normally operates. i.e. i_CSpIs the CSSM half-fault number.

According to a polynomial distribution probability formula, the total CSSM reliability R of the upper bridge arm_CSs2(t_n) As follows:

in the formula i_CSpIs the CSSM half-fault number. N is a radical of_CSIs the baseline number of CSSM. R_CSu(t_n) The CSSM reliability in good condition. i.e. i_csIs the number of CSSM damages. M_csIs the redundancy number of the CSSM. R_CSp(t_n) Reliability of the half fault state CSSM.

Based on the serial connection relation of all HBSM in the bridge arms and according to a binomial distribution probability formula, the total HBSM reliability R of the upper bridge arm_HBs2(t_n) As follows:

in the formula i_CSpIs the CSSM half-fault number. i.e. i_csIs the number of CSSM damages. N is a radical of_HBThe baseline number for the HBSM. R_HBu(tn)The HBSM reliability in the intact state. i.e. i_HBIs the number of HBSM breakages. M_HBIs the redundancy number of the HBSM.

Based on the series connection of CSSM and HBSM in the bridge arm, the second caseReliability R of single-phase upper bridge arm under condition_{Arm_u2}(t_n) As follows:

R_{Arm_u2}(t_n)＝R_CSs2(t_n)·R_HBs2(t_n)。 (38)

in the formula, R_HBs2(t_n) The total HBSM reliability for the upper arm in the second case. R_CSs2(t_n) The overall CSSM reliability for the upper leg in the second case.

Reliability function R of single-phase upper bridge arm_{Arm_u(tn)}As follows:

R_{Arm_u}(t_n)＝R_{Arm_u1}(t_n)+R_{Arm_u2}(t_n)。 (39)

in the formula, R_{Arm_u1}(t_n) The single-phase upper arm reliability in the first case. R_{Arm_u2}(t_n) The reliability of the single-phase upper bridge arm in the second condition.

Based on the symmetry of the mixed MMC and the series relation of each bridge arm, the reliability R (t) of the mixed MMC_n) As follows:

R(t_n)＝R_{Arm_u}(t_n)⁶。 (40)

in the formula, R_{Arm_u}(t_n) Is a single-phase upper bridge arm reliability function.

A method for using an improved hybrid MMC operational reliability assessment model based on multi-time scale thermal damage, comprising the steps of:

I) data is input in the improved hybrid MMC operational reliability evaluation model. The data mainly comprises environmental parameters and mixed MMC and fan electrical parameters. The environmental parameters mainly comprise wind speed V_tnAnd air temperature T_a,tn. The hybrid MMC and wind turbine electrical parameters mainly comprise switching frequency f_swThe hybrid MMC comprises a direct current side rated voltage, an alternating current side rated voltage, a power factor, a modulation ratio and a duty ratio.

II) according to the input data, the improved hybrid MMC operation reliability evaluation model calculates the loss of the power device. I.e. calculating samplesTime t_nAverage loss P of IGBT switching period in corresponding time interval T_T,avgAnd the average loss P of the switching period of the Diode_D,avg。

III) according to the input data, the improved hybrid MMC operation reliability evaluation model calculates the multi-time scale junction temperature of the power device.

Calculating the sampling time t_nJunction temperature mean value T of IGBT fundamental frequency period within corresponding time interval T_{Tjavg_F}Maximum value T_{Tjmax_F}Minimum value T_{Tjmin_F}And effective value of current I of aluminum bonding wire_F. Calculating 0-t according to the rain flow algorithm_nMaximum T of junction temperature curve of IGBT low-frequency period at moment_{Tjmax_L}Minimum value T_{Tjmin_L}And effective value of current I of aluminum bonding wire_L。

IV) according to the input data, the improved hybrid MMC operation reliability evaluation model calculates the service life consumption, the average failure time and the failure rate of the power device and the multi-time scale.

Calculating the number N of low-frequency cyclic failure cycles of the IGBT_{Tf_L}And the number N of cycles of failure of fundamental frequency cycle_{Tf_F}Thereby obtaining the IGBT at 0-t_nTime of day life consumption CL_T(t_n) Sampling time t_nMean time to failure MTTF corresponding to time interval T_T(t_n) Failure rate λ_T(t_n) And degree of reliability R_T(t_n)。

Calculating the number N of low-frequency cycle failure cycles of the Diode_{Df_L}And the number N of cycles of failure of fundamental frequency cycle_{Df_F}Thus obtaining a Diode in the range of 0-t_nTime of day life consumption CL_D(t_n) Sampling time t_nMean time to failure MTTF corresponding to time interval T_D(t_n) Failure rate λ_D(t_n) And degree of reliability R_D(t_n)。

V) according to the input data, the improved mixed MMC operation reliability evaluation model calculates the reliability R of the capacitor_C(t_n)。

VI) based on the input data, the improved hybrid MMC operational reliability assessment model calculates and accounts for the reliability of the SM multi-state hybrid MMCDegree R (t)_n) Thereby evaluating the reliability of improving the operation of the hybrid MMC.

The technical effect of the present invention is undoubted. The invention focuses on researching the coupling relation between the reliability of the hybrid MMC and the wind speed, the air temperature and the internal parameters of the electrical equipment, comprehensively considers the influence of the environmental factors such as the wind speed, the air temperature, the power frequency and the like and the electrical parameters on the operation reliability of the hybrid MMC from the failure mechanism of a device, and can effectively reflect the influence of the environmental factors and the internal electrical parameters of the equipment on the operation reliability of the hybrid MMC. The method couples the operational reliability of the hybrid MMC with the environmental factors and the electrical parameters, optimizes the topology of the hybrid MMC from the operation angle on the basis of loss, accurately describes the relationship between the operational reliability of the hybrid MMC and the environmental factors and the electrical parameters, and improves the operational reliability level of the hybrid MMC. Based on the running characteristic of the hybrid MMC, the invention provides a measure for reusing the fault SM aiming at the problem that the loss distribution of the SM in the hybrid MMC is seriously uneven, thereby improving the utilization rate of the SM in the hybrid MMC and achieving the effect of improving the running reliability of the hybrid MMC. The method can be widely applied to the operation reliability evaluation of the hybrid MMC in the wind power and other renewable energy transmission grid connection.

Drawings

FIG. 1 is a diagram of a hybrid MMC topology;

FIG. 2 is an HBSM topology;

FIG. 3 is a CSSM topology;

FIG. 4 is an annual wind speed data curve;

FIG. 5 is a plot of annual air temperature data;

FIG. 6 is a mixed MMC reliability curve before and after modification;

FIG. 7 is a hybrid MMC reliability curve;

FIG. 8 is an SM failure rate curve;

FIG. 9 is a graph of hybrid MMC reliability as a function of switching frequency and time.

Detailed Description

The present invention is further illustrated by the following examples, but it should not be construed that the scope of the above-described subject matter is limited to the following examples. Various substitutions and alterations can be made without departing from the technical idea of the invention and the scope of the invention is covered by the present invention according to the common technical knowledge and the conventional means in the field.

Example 1:

an improved hybrid MMC operation reliability evaluation model based on multi-time scale heat damage mainly comprises the following steps:

Further, the hybrid MMC topology mainly includes three a, b, c three phase cells.

The a-phase units are sequentially connected in series with N_CSCSSM module and N_HBAnd each HBSM module.

The b-phase units are sequentially connected in series with N_CSCSSM module and N_HBAnd each HBSM module.

The c-phase units are sequentially connected in series with N_CSCSSM module and N_HBAnd each HBSM module.

The HBSM is composed of power modules T1, T2, and a capacitor C.

The CSSM is comprised of power modules T1, T2, T3, and a capacitor C.

Preferably, the MMC topology structure may be an upper bridge arm and a lower bridge arm each formed by connecting N sub-modules and 1 bridge arm reactance L in series, where SM is formed by 2 switching devices and 1 energy storage capacitor. Each SM can be in 3 operative states of plunge, cut and latch.

Transport power P of hybrid MMC_MMCoutAs follows:

in the formula I_dcIs a direct side current. I is_mIs the ac side current peak.

Is the power factor angle. f. of₀Is the fundamental frequency.

Direct side current I_dcAs follows:

Peak value of AC side current I_mAs follows:

in the formula, P_MMCoutIs the transport power of the hybrid MMC.

As follows:

in the formula, R_ce、U_ceoAnd τ_TThe IGBT forward on-resistance, the threshold voltage and the duty cycle, respectively. a is_T、b_TAnd c_TAnd fitting parameters for the IGBT dynamic characteristic curve. U shape_ratedThe rated voltage of the IGBT. f. of_swAnd ρ_TThe switching frequency and the temperature coefficient of the IGBT, respectively. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle. I is_auIs the phase A upper bridge arm current. U shape_dcIs the dc side voltage.

Number of switching cycles n_swAs follows:

n_sw＝f_sw/f₀。 (7)

2.1.4) calculating the average loss of the j th switching period of the Diode

Average loss of the j-th switching period of the Diode

As follows:

The loss of the hybrid MMC power device causes the junction temperature of the device to rise, and the junction temperature of the power module can be effectively calculated by constructing a foster heat network model.

The junction temperature is determined from the foster thermal network model

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nThe corresponding air temperature.

As follows:

Is the temperature difference of the RC parallel unit in the j-1 switching period IGBT junction-shell heat network of the r-th order. e is the base of the natural log function and is about 2.71828. .

As follows:

As follows:

Because the power frequency and other electrical parameters of different switching periods are different, the electrical parameters are reflected in the fundamental frequency junction temperature of a short time scale. 0-t_nThe time instant comprises n sampling time intervals T, where T_nCorresponding to a sampling time interval T having N_{T_F}One period of fundamental frequency (N)_{T_F}＝T×f₀) The ith fundamental frequency period (i ═ 1, …, N_{T_F}) And taking the temperature difference of the RC parallel unit as an iterative variable to obtain a junction temperature curve of the fundamental frequency period, and further obtain a junction temperature mean value T corresponding to the fundamental frequency period_{Tjavg_F}Maximum value T_{Tjmax_F}Minimum value T_{Tjmin_F}And effective value of current I of aluminum bonding wire_FAnd the method is used for evaluating the reliability of the IGBT fundamental frequency time scale. Assuming that the wind speed and air temperature are constant within a sampling time interval T, then N is_{T_F}T of one fundamental frequency period_{Tjavg_F}、T_{Tjmax_F}、T_{Tjmin_F}And I_FAnd therefore, the junction temperature parameter of one fundamental frequency period can be extracted.

The wind speed and air temperature at different sampling time intervals T are different and are reflected in the low-frequency junction temperature of a long time scale. According to 0-t_nMean value T of junction temperature of fundamental frequency of each sampling time interval T at moment_{Tjavg_F}Obtaining N by rain flow algorithm_{Tsum_L}Obtaining a low-frequency period junction temperature curve, and further obtaining a g-th low-frequency period junction temperature curve (g is 1, …, N)_{Tsum_L}) Corresponding maximum value of junction temperature T_{Tjmax_L}Minimum value T_{Tjmin_L}And effective value of current I of aluminum bonding wire_LAnd the method is used for evaluating the reliability of the IGBT low-frequency time scale.

The junction temperature

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nThe corresponding air temperature.

The temperature difference of the RC parallel units in the j switching period Diode shell-fin heat network.

The temperature difference of the RC parallel unit in the jth switching period IGBT heat sink-environment thermal network. j is an arbitrary switching period. j is 1, …, n_sw。n_swIs the total number of switching cycles in one fundamental frequency cycle. Because the temperature difference of the RC parallel unit in the jth switching period IGBT radiating fin-environment heat network and the temperature difference of the RC parallel unit in the jth switching period Diode radiating fin-environment heat network are equal in value, in order to simplify the calculation steps, the Diode junction temperature is calculated

In time, the temperature difference can be directly used

As follows:

As follows:

in the formula, R_DchIs the thermal resistance. Tau is_DchIs the thermal resistance R_DchThermal time constant of (2). T is_swIs a switching cycle.

2.2) hybrid MMC power device multi-time scale reliability evaluation. In the reliability evaluation of the hybrid MMC power device, the influence of environmental factors such as wind speed and air temperature and electrical parameters such as power frequency on the reliability of the power device is quantified by taking the low-frequency junction temperature and the fundamental-frequency junction temperature of the power device into consideration, so that the reliability evaluation precision of the power device is improved.

The method mainly comprises the following steps:

Preferably, the Miner's damage theory mainly needs to calculate the damage caused by one cycle and N under the constant-amplitude load_{Tsum_L}Damage caused by individual cycles, N under variable amplitude load_{Tsum_L}Damage caused by one cycle.

IGBT failure rate lambda_T(t_n) As follows:

IGBT reliability R_T(t_n) As follows:

According to Miner's injury theory and equation (23), the Diode is at 0-t_nTime of day life consumption CL_D(t_n) As follows:

Diade failure rate λ_D(t_n) As follows:

Diade reliability R_D(tn) is as follows:

in the formula, λ_D(t_n) Is the Diode failure rate. e is the base of the natural log function and is about 2.71828. . t is t_nIs the sampling instant.

3.1) calculating the capacitor reliability R_C(t_n) Namely:

Self-healing energy W_siAs follows:

The hybrid MMC main circuit consists of three phase units, namely a phase unit, a phase unit and a phase unit, wherein each phase unit is divided into an upper bridge arm and a lower bridge arm, and each bridge arm is formed by connecting a plurality of SM in series. In order to take cost into consideration, the SM in each bridge arm comprises a half-bridge SM (HBSM) and an SM with direct current fault ride-through capability, and the single-clamping SM (CSSM) is selected as the SM with the direct current fault ride-through capability. Both SM main circuits consist of IGBTs, diodes (diodes) and capacitors. Meanwhile, in order to improve the reliability of the hybrid MMC, each bridge arm includes a redundant SM as a backup, and the embodiment adopts an active backup which is usually adopted in practice.

HBSM reliability R in good condition_HBu(t_n) As follows:

R_HBu(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·R_C(t_n)。 (31)

in the formula, R_T1(t_n)、R_T2(t_n) And R_T3(t_n) Of a first IGBT, a second IGBT and a third IGBT respectivelyAnd (7) reliability. R_D1(t_n)、R_D2(t_n) And R_D3(t_n) Reliability of the first, second and third Diode, respectively. R_c(t_n) The reliability of the capacitor.

At unpaired T₃Before the module is added with a bypass switch, when T₃And D₃When any one of the devices fails, the entire CSSM is considered to have failed and immediately exits operation.

If it is at this time to T₃The module carries out bypass processing, can make CSSM be in the half fault state who maintains voltage support function to the equivalence continues the operation for HBSM, and then promotes CSSM utilization ratio.

Reliability R of CSSM under half fault condition_CSp(t_n) As follows:

4.2) accounting for SM multistate hybrid MMC reliability

The hybrid MMC bridge arm is formed by connecting a plurality of CSSM and HBSM in series.

Because both the complete CSSM and the half-fault CSSM can be equivalent to the HBSM, the voltage supporting function is maintained, and when the damage number of the HBSM is excessive, the condition that the complete CSSM and the half-fault CSSM replace the damaged HBSM exists, so that the hybrid MMC keeps normal operation.

Therefore, the bridge arm reliability can be divided into two cases, taking the single-phase upper bridge arm reliability as an example:

the reliability of the single-phase upper bridge arm is mainly divided into two cases.

R_{Arm_u1}(t_n)＝R_CSs1(t_n)·R_HBs1(t_n)。 (35)

The second case is: when CSSM damage number does not exceed self redundancy number, and HBSM damage number exceeds self redundancy number, and good and half fault CSSMWhen displacing a damaged HBSM, i.e. i_CS≤M_CSAnd M is_HB<i_HB≤M_HB+M_CS-(i_CS-i_CSp) Meanwhile, the hybrid MMC normally operates. i.e. i_CSpIs the CSSM half-fault number.

Based on the series relation between CSSM and HBSM in bridge arm, the reliability R of single-phase upper bridge arm under the second condition_{Arm_u2}(t_n) As follows:

R_{Arm_u2}(t_n)＝R_CSs2(t_n)·R_HBs2(t_n)。 (38)

in the formula, R_HBs2(t_n) The total HBSM reliability for the upper arm in the second case. R_CSs2(t_n) Total C of upper bridge arm in the second caseSSM reliability.

Reliability function R of single-phase upper bridge arm_{Arm_u(tn)}As follows:

R_{Arm_u}(t_n)＝R_{Arm_u1}(t_n)+R_{Arm_u2}(t_n)。 (39)

R(t_n)＝R_{Arm_u}(t_n)⁶。 (40)

Example 2:

1) data is input in the improved hybrid MMC operational reliability evaluation model. The data mainly comprises environmental parameters and mixed MMC and fan electrical parameters.

The environmental parameter mainly comprises wind speed

And air temperature

The hybrid MMC and wind turbine electrical parameters mainly comprise switching frequency f_swThe hybrid MMC comprises a minimum cut-in wind speed, a maximum cut-off wind speed, a rated wind speed and the like, a direct current side rated voltage, an alternating current side rated voltage, a power factor, a modulation ratio, a duty ratio and the like.

2) And according to the input data, calculating the loss of the power device by the improved hybrid MMC operation reliability evaluation model.

I.e. calculating the sampling instant t_nAverage IGBT switching period in corresponding time interval TLoss P_T,avgAnd the average loss P of the switching period of the Diode_D,avg。

3) And according to the input data, calculating the multi-time scale junction temperature of the power device by the improved hybrid MMC operation reliability evaluation model.

Calculating the sampling time t_nJunction temperature mean value T of IGBT fundamental frequency period within corresponding time interval T_{Tjavg_F}Maximum value T_{Tjmax_F}Minimum value T_{Tjmin_F}And effective value of current I of aluminum bonding wire_F。

Calculating 0-t according to the rain flow algorithm_nMaximum T of junction temperature curve of IGBT low-frequency period at moment_{Tjmax_L}Minimum value T_{Tjmin_L}And effective value of current I of aluminum bonding wire_L。

4) According to input data, the improved hybrid MMC operation reliability evaluation model calculates the service life consumption, the average failure time and the failure rate of the power device in consideration of multiple time scales.

5) According to the input data, the improved hybrid MMC operation reliability evaluation model calculates the reliability RC (t) of the capacitor_n)。

6) According to the input data, the improved hybrid MMC operation reliability evaluation model calculates hybrid MMC reliability R (t) of SM multi-state_n) Thereby improving the hybrid MMCThe reliability of the operation was evaluated.

Example 3:

a test for verifying an improved hybrid MMC operation reliability evaluation model based on multi-time scale thermal damage mainly comprises the following steps:

1) and configuring experimental parameters. In the embodiment, an FF1000R17IE4 model power module of Infineon company is used as a hybrid MMC power module, as shown in Table 1, and 2016-year wind speed and temperature data of Ireland Dublin are used as external working conditions of a wind farm and the hybrid MMC, so that the operational reliability of the hybrid MMC is evaluated based on the model provided by the invention. The annual data curves of the wind speed and the air temperature are shown in figures 2 and 3, and the mixed MMC and fan parameters are shown in table 2.

TABLE 1 thermal parameter Table for FF1000R17IE4 model Power Module

Parameter(s)	Numerical value
		Rated capacity of system	400MW
Network side voltage	220kV
		Voltage on the direct current side	400kV
Reference number of HBSM	137
		CSSM reference number	113
Reference failure rate of capacitor	1ⅹ10^-8occ/hour
		Reference voltage of capacitor	1.6kV
Output frequency	50Hz
		Switching frequency	2kHz
Power factor	0.9
		Modulation ratio	0.8
Fan capacity	2MW
		Number of fans	200
Cut-in wind speed	3m/s
		Rated wind speed	8.5m/s
Cut-out wind speed	16m/s

TABLE 2 hybrid MMC and Fan parameters

2) And (4) analyzing the thermal damage of the hybrid MMC power device. In order to verify the accuracy of the multi-time scale reliability evaluation model of the hybrid MMC power device, the low-frequency and fundamental frequency annual life consumption conditions of the power device are respectively obtained based on the reliability evaluation model of the power device, which is provided by the invention, by taking wind speed and air temperature annual data as input, and the specific table is shown in Table 3.

As can be seen from Table 3, although the low frequency lifetime was consumed at T₁、T₂、T₃、D₁And D₂The medium ratio is large, but the consumption ratio of the life of the fundamental frequency is not negligible, and the consumption ratio is T₂、D₁And D₃The percentage of the component (A) is respectively 16.39%, 25.06% and 67.82%, especially in D₃In the middle, the consumption of the life of the fundamental frequency is dominant over the consumption of the low frequency. Therefore, compared with the reliability evaluation only considering the low-frequency service life consumption, the multi-time scale reliability evaluation model established by the invention can comprehensively take the influence of environmental factors and electrical parameters on the hybrid MMC power device into account, and lays a foundation for the operation reliability evaluation of equipment.

TABLE 3 hybrid MMC power device annual lifetime consumption

3) Improved measure to hybrid MMC reliability impact

As can be seen from Table 3, T₃And D₃The sum of the lifetime consumptions of (a) is up to 50.08% higher in CSSM lifetime consumption than in other devices, and thus, T is higher₃The module is most vulnerable. The invention passes the fault T₃Short-circuiting the modules to enable T₃CSSM equivalent of module trouble is HBSM, continues to maintain voltage support function to promote CSSM utilization ratio, reach and promote mixed MMC reliability level.

To verify the impact of improved CSSM on hybrid MMC reliability proposed by the present invention, the SM multi-state-taking hybrid MMC operational reliability assessment is proposed based on the conventional model and the present inventionModel to obtain improved front and back mixed MMC reliability R₁And R₂See, in particular, fig. 4 and table 4.

As can be seen from FIG. 4, the improved reliability R of the hybrid MMC₂Greater than reliability before improvement R₁. As can be seen from Table 4, R is the time when the mixed MMC is operated for 15 years₁And R₂0.7421 and 0.8838, respectively, in comparison to R₁Improved mixed MMC reliability R₂The improvement is 19.10%. And when the mixed MMC is 20 years, R₁And R₂0.2018 and 0.4978, respectively, in comparison to R₁The improved reliability R2 of the hybrid MMC is improved by 146.68%. Therefore, the reliability of the improved hybrid MMC is effectively improved, and the effectiveness of the improved measures is verified.

Time/year	Reliability R1	Reliability R2	Percentage of reliability improvement
				5	0.9999	0.9999	0.00％
10	0.9861	0.9935	0.75％
				15	0.7421	0.8838	19.10％
20	0.2018	0.4978	146.68％
				25	0.0113	0.1184	947.78％
30	0	0.0096	—

TABLE 4 MMC reliability Table with mix of front and rear

4) And (5) evaluating the operation reliability of the hybrid MMC.

In order to verify the effectiveness of the mixed MMC operation reliability model provided by the invention, the influence of the fluctuation of environmental factors such as wind speed and air temperature and the difference of electrical parameters of the mixed MMC such as switching frequency on the operation reliability of the mixed MMC is evaluated based on the reliability evaluation model provided by the invention.

4.1) analyzing the influence of environmental factors on the operation reliability of the hybrid MMC.

In order to analyze the influence of environmental factors such as wind speed, air temperature and the like on the reliability of the hybrid MMC, the hybrid MMC is operated according to a hybrid MMC operation reliability model and a conventional constant failure rate model provided by the text by reference numbers to respectively obtain the reliability R of the hybrid MMC₁And R₂See, in particular, fig. 5. At the same time, CSSM and HBSM failure rate curves are derived, as shown in fig. 6.

As can be seen from FIG. 5, the reliability of the hybrid MMC may be changed by the influence of environmental factors such as wind speed and air temperature. As shown in FIGS. 2 and 6, when the hybrid MMC is operated for 1000h and 500-h, the wind speed is high and is concentrated on 7-11m/s, the service life loss of the device is large, and the fault rates of HBSM and CSSM are concentrated on 0.9 multiplied by 10^-6occ/hour and 1.8X 10^-6occ/hour, mixed MMC decreased faster. When the hybrid MMC operates at 3500-4000h, the wind speed is low and is concentrated at 1-5m/s, the service life loss of the device is low, the fault rate of HBSM and CSSM is close to zero, and the reliability of the hybrid MMC is slowly reduced. Similarly, as shown in FIG. 3, during the period of 4000-. Therefore, the hybrid MMC operational reliability assessment model presented herein is in line with theoretical expectations. Furthermore, as can be seen from FIG. 5, since R₂Not considering the changes of environmental factors such as wind speed, air temperature and the like, and R is measured at 1000h, 4000h and 7000h₂And R₁The relative errors respectively reach 41.38%, 34.45% and 82.36%, and the reliability difference is large. Therefore, the influence of environmental factors such as wind speed and air temperature on the reliability of the hybrid MMC needs to be taken into consideration.

4.2) analyzing the influence of the electrical parameters on the operation reliability of the hybrid MMC.

In order to analyze the influence of electrical parameters such as switching frequency on the operation reliability of the hybrid MMC, the reliability change condition of the hybrid MMC under different switching frequencies is obtained based on the operation reliability model of the hybrid MMC provided by the invention, and the concrete figure is shown in FIG. 7. Table 5 shows the reliability operation of the hybrid MMC at different switching frequencies for 900 h.

As can be seen from fig. 7, the higher the switching frequency, the faster the reliability of the hybrid MMC decreases. As can be seen from Table 5, when the switching frequency of the power device is increased from 1000Hz to 3000Hz at time 900h, the switching loss is increased, resulting in CSSM and HBSM failure rates of 6.91 × 10^-7occ/hour and 1.56X 10^-7occ/hour rose to 2.17X 10^-6occ/hour and 1.63X 10^-6occ/hour, the mixed MMC reliability is finally reduced from 0.715 to 0.2572. Therefore, the operation reliability evaluation model provided by the invention can effectively depict the influence of the switching frequency on the operation reliability of the hybrid MMC.

Table 5900 h-time mixed MMC operation reliability table under different switching frequencies

In conclusion, the mixed MMC operation reliability evaluation model considering the multi-time scale accumulated damage effect can accurately depict the influence of environmental factors such as wind speed and air temperature and electrical parameters such as switches on the operation reliability of the mixed MMC, the reliability of the mixed MMC is improved from the operation angle, and the reliability level of the improved mixed MMC is proved to be remarkably improved by constructing the corresponding mixed MMC reliability evaluation model considering the SM multi-state.

Claims

1. A method for establishing an improved hybrid MMC operation reliability evaluation model based on multi-time scale heat damage is characterized by mainly comprising the following steps:

1) acquiring real-time data of a wind power transmission system; the real-time data mainly comprises wind speed, air temperature and equipment electrical parameters;

2) according to the real-time data, establishing a reliability evaluation model of the hybrid MMC power device by utilizing the running characteristics of the hybrid MMC and a Miner's damage theory; the power device of the wind power transmission system mainly comprises an Insulated Gate Bipolar Transistor (IGBT) and a crystal Diode (Diode);

3) establishing a mixed MMC capacitor reliability evaluation model considering the voltage sharing of the fault SM according to the influence of the fault SM on the intact SM capacitor;

4) analyzing loss distribution of the hybrid MMC device, and thus establishing an improved hybrid MMC operation reliability evaluation model considering SM multi-state;

the method mainly comprises the following steps of establishing an improved hybrid MMC operation reliability evaluation model considering SM multi-state:

4.1) calculating the reliability of the SM multi-state; the SM main circuit consists of an IGBT, a Diode and a capacitor;

R_CSu(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·R_T3(t_n)·R_D3(t_n)·R_C(t_n)； (1)

in the formula, R_T1(t_n)、R_T2(t_n) And R_T3(t_n) Reliability of the first IGBT, the second IGBT and the third IGBT respectively; r_D1(t_n)、R_D2(t_n) And R_D3(t_n) Reliability of the first, second and third Diode, respectively; r_c(t_n) Capacitor reliability;

HBSM reliability R in good condition_HBu(t_n) As follows:

R_HBu(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·R_C(t_n)； (2)

in the formula, R_T1(t_n)、R_T2(t_n) Reliability of the first IGBT and the second IGBT respectively; r_D1(t_n)、R_D2(t_n) Reliability of the first and second diodes, respectively; r_c(t_n) Capacitor reliability;

reliability R of CSSM under half fault condition_CSp(t_n) As follows:

R_CSp(t_n)＝R_T1(t_n)·R_D1(t_n)·R_T2(t_n)·R_D2(t_n)·(1-R_T3(t_n)·R_D3(t_n))·R_C(t_n)； (3)

in the formula, R_T1(t_n)、R_T2(t_n) And R_T3(t_n) Reliability of the first IGBT, the second IGBT and the third IGBT respectively; r_D1(t_n)、R_D2(t_n) And R_D3(t_n) Respectively, a first, a second and a third DiodeReliability; rc (t)_n) Capacitor reliability;

4.2) accounting for SM multistate hybrid MMC reliability

The reliability of the bridge arm is mainly divided into two cases;

the first case is: when neither CSSM nor HBSM damaged number exceeds its own redundancy number, i_CS≤M_CSAnd i is_HB≤M_HBWhen the hybrid MMC normally operates;

in the formula, N_CSA reference number for the CSSM; r_CSu(t_n) CSSM reliability under intact condition; i.e. i_csIs the number of CSSM defects; m_CSRedundancy number for CSSM;

in the formula, N_HBA reference number for HBSM; r_HBu(t_n) HBSM reliability under intact state; i.e. i_HBDamage number for HBSM; m_HBRedundancy number for HBSM;

R_{Arm_u1}(t_n)＝R_CSs1(t_n)·R_HBs1(t_n)； (6)

in the formula, R_HBs1(t_n) The total HBSM reliability of the upper bridge arm in the first case; r_CSs1(t_n) In the first caseThe total CSSM reliability of the lower and upper bridge arms;

the second case is: when the CSSM defect number does not exceed the self-redundancy number, the HBSM defect number exceeds the self-redundancy number, and the good and half-fault CSSMs replace the damaged HBSM, i.e., i_CS≤M_CSAnd M is_HB<i_HB≤M_HB+M_CS-(i_CS-i_CSp) When the hybrid MMC normally operates; i.e. i_CSpIs the CSSM half fault number;

in the formula i_CSpIs the CSSM half fault number; n is a radical of_CSA reference number for the CSSM; r_CSu(t_n) CSSM reliability under intact condition; i.e. i_csIs the number of CSSM defects; m_csRedundancy number for CSSM; r_CSp(t_n) Reliability for a half fault state CSSM;

in the formula i_CSpIs the CSSM half fault number; i.e. i_csIs the number of CSSM defects; n is a radical of_HBA reference number for HBSM; r_HBu(tn)HBSM reliability under intact state; i.e. i_HBDamage number for HBSM; m_HBRedundancy number for HBSM;

R_{Arm_u2}(t_n)＝R_CSs2(t_n)·R_HBs2(t_n)； (9)

in the formula, R_HBs2(t_n) The total HBSM reliability of the upper bridge arm under the second condition; r_CSs2(t_n) The total CSSM reliability for the upper leg in the second case;

reliability function R of single-phase upper bridge arm_{Arm_u(tn)}As follows:

R_{Arm_u}(t_n)＝R_{Arm_u1}(t_n)+R_{Arm_u2}(t_n)； (10)

in the formula, R_{Arm_u1}(t_n) The reliability of the single-phase upper bridge arm under the first condition; r_{Arm_u2}(t_n) The reliability of the single-phase upper bridge arm under the second condition;

R(t_n)＝R_{Arm_u}(t_n)⁶； (11)

2. The method for establishing the improved hybrid MMC operation reliability evaluation model based on multi-time scale heat damage according to claim 1, wherein: the method mainly comprises the following steps of establishing the reliability evaluation model of the hybrid MMC power device:

1) calculating the multi-time scale junction temperature of the hybrid MMC power device; the method mainly comprises the following steps:

1.1) determining the output power P of the fan of the wind power transmission system_WToutAnd a sampling time t_nCorresponding to wind speed V_tnThe relationship of (1);

in the formula, P_ratedRated power for the fan; v_cutin、V_ratedAnd V_cutoutRespectively the cut-in wind speed, the rated wind speed and the cut-out wind speed; k is a radical of_pRepresenting coefficients related to air density and fan area; n is a radical of_WTThe total number of fans in the wind power transmission system; n is the total number of sampling points until the current running time; t is t_nIs the sampling time;

is the wind speed; i is any fan; 1, …, N_WT；

Transport power P of hybrid MMC_MMCoutAs follows:

in the formula, N_WTThe total number of fans in the wind power transmission system; i is any fan; 1, …, N_WT；P_WTout,iOutputting power for the ith fan;

1.2) calculating to obtain a mixed MMC bridge arm current according to a formula 1;

in the formula I_dcIs direct current side current; i is_mIs the ac side current peak;

is a power factor angle; f. of₀Is the fundamental frequency;

direct side current I_dcAs follows:

in the formula of U_dcIs a direct current side voltage; p_MMCoutA transport power for a hybrid MMC;

peak value of AC side current I_mAs follows:

in the formula, P_MMCoutA transport power for a hybrid MMC;

is a power factor angle; u shape_acThe effective value of the voltage at the alternating current side;

1.3) calculating the average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A;

As follows:

in the formula, R_ce、U_ceoAnd τ_TThe IGBT positive on-resistance, the threshold voltage and the duty ratio are respectively; a is_T、b_TAnd c_TFitting parameters for the IGBT dynamic characteristic curve; u shape_ratedRated voltage for IGBT; f. of_swAnd ρ_TThe switching frequency and the temperature coefficient of the IGBT are respectively; j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle; i is_auIs the A phase upper bridge arm current; u shape_dcIs a direct current side voltage;

number of switching cycles n_swAs follows:

n_sw＝f_sw/f₀； (18)

in the formula (f)_swIs the switching frequency of the IGBT; f. of₀Is the fundamental frequency;

1.4) calculating the average loss of the j-th switching period of the Diode

Average loss of the j-th switching period of the Diode

As follows:

in the formula: r_d、U_dAnd τ_DRespectively, Diode forward on-resistance, threshold voltage and duty cycle; a is_D、b_DAnd c_DFitting parameters for the dynamic characteristic curve of the Diode; u shape_DIs a Diode rated voltage; f. of_dwAnd ρ_DRespectively, the switching frequency and the temperature coefficient of the Diode; i is_auIs the A phase upper bridge arm current; u shape_dcIs a direct current side voltage;

1.5) calculating the junction temperature of the jth switching period of the IGBT in the mixed MMC

The junction temperature

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nA corresponding air temperature;

the temperature difference of an RC parallel unit in an IGBT junction-shell thermal network in the jth switching period of the nth order; r represents an arbitrary order; r is 1,2,3, 4;

the temperature difference of RC parallel units in the jth switching period IGBT shell-radiating fin heat network is obtained;

the temperature difference of an RC parallel unit in the jth switching period IGBT radiating fin-environment thermal network is obtained; j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

As follows:

in the formula, R_Tjc,rIs the thermal resistance; tau is_Tjc,rIs the thermal resistance R_Tjc,rThermal time constant of (d); t is_swIs a switching cycle;

is the temperature difference of RC parallel units in the ith order of switching period (j-1) IGBT junction-shell thermal network;

average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A; j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

As follows:

in the formula, R_TchIs the thermal resistance; tau is_TchIs the thermal resistance R_TchThermal time constant of (d); t is_swIs a switching period

The temperature difference of RC parallel units in the j-1 switching period IGBT shell-radiating fin heat network;

As follows:

in the formula, R_haIs the thermal resistance; tau is_haIs the thermal resistance R_haThermal time constant of (d); t is_swIs a switching cycle;

is the cycle temperature difference of the RC parallel unit in the j-1 th switch IGBT thermal network; j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

average loss of the jth switching period of the IGBT of the upper bridge arm of the phase A;

average loss of the jth switching period of the A-phase upper bridge arm Diode;

1.6) calculating the junction temperature of the Diode jth switching cycle in the mixed MMC

The junction temperature

As follows:

in the formula (I), the compound is shown in the specification,

is the sampling time t_nA corresponding air temperature;

the temperature difference of an RC parallel unit in a j-th switching period Diode junction-shell heat network of an r-th order; r represents an arbitrary order; r is 1,2,3, 4;

the temperature difference of RC parallel units in the j switching period Diode shell-radiating fin heat network is obtained;

for the temperature difference of RC parallel unit in jth switching period IGBT heat sink-environment heat network(ii) a j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

As follows:

in the formula, R_Djc,rIs the thermal resistance; tau is_Djc,rIs the thermal resistance R_Djc,rThermal time constant of (d); t is_dwIs a switching cycle;

is the temperature difference of the RC parallel unit in the j-1 switching period Diode junction-shell heat network; j is any switching period; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

average loss of the jth switching period of the A-phase upper bridge arm Diode;

As follows:

in the formula, R_DchIs the thermal resistance; tau is_DchIs the thermal resistance R_DchThermal time constant of (d); t is_swIs a switching cycle;

is the temperature difference of RC parallel units in the j-1 switching period Diade shell-fin heat network;

average loss of the jth switching period of the A-phase upper bridge arm Diode; j is 1, …, n_sw；n_swThe total number of switching cycles in a fundamental frequency cycle;

2) evaluating the multi-time scale reliability of the hybrid MMC power device; the method mainly comprises the following steps:

2.1) calculating the cycle failure period number N corresponding to the low-frequency period of the IGBT_{Tf_L}(ii) a Number of cycles to failure N_{Tf_L}As follows:

in the formula, t_onIs the heating time; u is 0.01 times of the blocking voltage of the module; d is the diameter of the aluminum bonding wire; k is 9.3 × 10¹⁴；β₁＝-4.416；β₂＝1285；β₃＝-0.463；β₄＝-0.716；β₅＝-0.761；β₆＝-0.5；T_{Tjmax_L}The maximum value of the IGBT low-frequency junction temperature is obtained; t is_{Tjmin_L}The minimum value of the IGBT low-frequency junction temperature is obtained; i is_LThe effective value of the current of the IGBT low-frequency aluminum bonding wire is obtained;

cycle failure period number N corresponding to IGBT fundamental frequency period_{Tf_F}As follows:

in the formula, t_onIs the heating time; u is 0.01 times of the blocking voltage of the module; d is the diameter of the aluminum bonding wire; k is 9.3 × 10¹⁴；β₁＝-4.416；β₂＝1285；β₃＝-0.463；β₄＝-0.716；β₅＝-0.761；β₆＝-0.5；T_{Tjmax_F}The maximum value of the IGBT fundamental frequency junction temperature is obtained; t is_{Tjmin_F}Junction temperature of IGBT fundamental frequencyA minimum value; i is_FThe effective value of the current of the IGBT fundamental frequency aluminum bonding wire is obtained;

in the formula, N_{T_L,g}And N_{Tf_L,g}Respectively corresponding thermal cycle times and cycle failure cycles of the IGBTs in the g-th low-frequency thermal cycle; n is a radical of_{T_F,q}And N_{Tf_F,q}Respectively corresponding to the q sampling time interval T of the IGBT, the fundamental frequency thermal cycle times and the cycle failure cycles; n is the total number of sampling points until the current running time; n is a radical of_{Tsum_L}Is the total number of low frequency thermal cycles;

in the formula, CL_T(t_n) Is IGBT at 0-t_nThe consumption of the lifetime at a moment; t is a sampling time interval;

IGBT failure rate lambda_T(t_n) As follows:

in the formula, MTTF_T(t_n) For the IGBT sampling time t_nThe average time to failure of the corresponding time interval T;

IGBT reliability R_T(tn)As follows:

in the formula, λ_T(t_n) Is the IGBT failure rate; t is t_nIs the sampling time;

2.2) calculating the cycle failure period number N corresponding to the Diode low-frequency period_{Df_L}(ii) a Number of cycles to failure N_{Df_L}As follows:

in the formula, t_onIs the heating time; u is 0.01 times of the blocking voltage of the module; d is the diameter of the aluminum bonding wire; k is 9.3 × 10¹⁴；β₁＝-4.416；β₂＝1285；β₃＝-0.463；β₄＝-0.716；β₅＝-0.761；β₆＝-0.5；T_{Djmax_L}Is the maximum value of the Diode low-frequency junction temperature; t is_{Djmin_L}Is the minimum value of the Diode low-frequency junction temperature; i is_DLThe effective value of the current of the Diode low-frequency aluminum bonding wire is obtained;

in the formula, t_onIs the heating time; u is 0.01 times of the blocking voltage of the module; d is the diameter of the aluminum bonding wire; k is 9.3 × 10¹⁴；β₁＝-4.416；β₂＝1285；β₃＝-0.463；β₄＝-0.716；β₅＝-0.761；β₆＝-0.5；T_{Djmax_F}Is the maximum value of the junction temperature of the Diode fundamental frequency; t is_{Djmin_F}Is the minimum value of the junction temperature of the Diode fundamental frequency; i is_DFThe effective value of the current of the Diade fundamental frequency aluminum bonding wire is obtained;

in the formula, N_{Dsum_L}Is a Diode at 0-t_nTotal number of low frequency thermal cycles at the moment; n is a radical of_{D_L,α}And N_{Df_L,α}Respectively corresponding thermal cycle times and cycle failure cycles of the alpha-th low-frequency thermal cycle of the Diode; n is a radical of_{D_F,ω}And N_{Df_F,ω}Respectively corresponding to the omega th sampling time interval T, the fundamental frequency thermal cycle times and the cycle failure cycles; n is the total number of sampling points until the current running time;

in the formula, CL_D(t_n) Is a Diode at 0-t_nThe consumption of the lifetime at a moment; t is a sampling time interval;

diade failure rate λ_D(t_n) As follows:

in the formula, MTTF_D(t_n) For the moment t of the Diade sampling_nThe average time to failure of the corresponding time interval T;

diade reliability R_D(tn) is as follows:

in the formula, λ_D(t_n) Is the rate of Diade failures; t is t_nIs the sampling instant.

3. The method for establishing the improved hybrid MMC operation reliability evaluation model based on multi-time scale heat damage according to claim 1, wherein: the method mainly comprises the following steps of establishing a reliability evaluation model of the hybrid MMC capacitor:

1) calculating capacitor reliability R_C(t_n) Namely:

in the formula, λ_cIs the capacitor failure rate; w_s0Is in an initial state; w_siSelf-healing energy corresponding to the i SM faults; t is t_nIs the sampling time;

self-healing energy W_siAs follows:

U_Cicapacitor voltages corresponding to i SM faults; r_CIs a thin film resistor; c is capacitance; (p) is a sandwich pressure related function; a is the capacitor correlation coefficient; b is a resistance correlation coefficient; k is a radical of_cCalculating the correlation coefficient of self-healing energy;

2) according to capacitor reliability R_C(t_n) The hybrid MMC capacitor reliability was evaluated.

4. A method for modeling an improved hybrid MMC operational reliability assessment model based on multi-time scale thermal damage according to any one of claims 1-3, and mainly comprising the steps of:

1) inputting data in an improved hybrid MMC operation reliability evaluation model; the data mainly comprises environmental parameters, mixed MMC and fan electrical parameters; the environmental parameter mainly comprises wind speed

And air temperature

The hybrid MMC and wind turbine electrical parameters mainly comprise switching frequency f_swThe hybrid MMC has a direct current side rated voltage, an alternating current side rated voltage, a power factor, a modulation ratio and a duty ratio;

2) according to input data, the improved hybrid MMC operation reliability evaluation model calculates the loss of a power device; i.e. calculating the sampling instant t_nAverage loss P of IGBT switching period in corresponding time interval T_T,avgAnd the average loss P of the switching period of the Diode_D,avg；

3) According to input data, the improved hybrid MMC operation reliability evaluation model calculates the multi-time scale junction temperature of the power device;

calculating the sampling time t_nJunction temperature mean value T of IGBT fundamental frequency period within corresponding time interval T_{Tjavg_F}Maximum value T_{Tjmax_F}Minimum value T_{Tjmin_F}And effective value of current I of aluminum bonding wire_F(ii) a Calculating 0-t according to the rain flow algorithm_nMaximum T of junction temperature curve of IGBT low-frequency period at moment_{Tjmax_L}Minimum value T_{Tjmin_L}And effective value of current I of aluminum bonding wire_L；

4) According to input data, the improved hybrid MMC operation reliability evaluation model calculates the service life consumption, the average failure time and the failure rate of the power device and the multi-time scale;

calculating the number N of low-frequency cyclic failure cycles of the IGBT_{Tf_L}And the number N of cycles of failure of fundamental frequency cycle_{Tf_F}Thereby obtaining the IGBT at 0-t_nTime of day life consumption CL_T(t_n) Sampling time t_nMean time to failure MTTF corresponding to time interval T_T(t_n) Failure rate λ_T(t_n) And degree of reliability R_T(t_n)；

Calculating the number N of low-frequency cycle failure cycles of the Diode_{Df_L}And the number N of cycles of failure of fundamental frequency cycle_{Df_F}Thus obtaining a Diode in the range of 0-t_nTime of day life consumption CL_D(t_n) Sampling time t_nMean time to failure MTTF corresponding to time interval T_D(t_n) Failure rate λ_D(t_n) And degree of reliability R_D(t_n)；

5) According to input data, the improved hybrid MMC operation reliability evaluation model calculates the reliability R of the capacitor_C(t_n)；

6) According to the input data, the improved hybrid MMC operation reliability evaluation model calculates hybrid MMC reliability R (t) of SM multi-state_n) Thereby evaluating the reliability of improving the operation of the hybrid MMC.