Method and system for predicting service life of traction converter based on probabilistic damage model
By constructing a probability damage model, combining the physics of solder cracking failure and heat conduction characteristics, the complexity problem of traction converter life prediction in the existing technology is solved, and more accurate and true life prediction is achieved, suitable for traction in complex environments Converter.
Patent Information
- Application Number
- CN202411109496.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-13
AI Technical Summary
In the prior art, the traction converter life prediction method ignores complex environmental factors and the variability and dependence of the degradation process of internal semiconductor devices, resulting in a large deviation from the prediction results from the actual situation.
Using a method based on the probability damage model, the power loss model, thermal model, degradation model and interdependence degradation relationship model of semiconductor devices are constructed. Combined with solder cracking failure physics, the probability damage model is constructed and the limit state equation analysis is performed to achieve life prediction.
Improves the accuracy and authenticity of life prediction, can be closer to actual system operation, and is suitable for traction converters in complex environments, helping to deeply understand system behavior and improve maintenance strategies.
Smart Images

Figure CN119047161A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of prognosis and health management of traction converters, and in particular to a method and system for predicting the life of a traction converter based on a probabilistic damage model. Background Art
[0002] As the core component of the electric traction system, the traction inverter is responsible for converting the high-voltage power of the power grid into low-voltage power suitable for electric traction. Its performance and life directly affect the operating stability and safety of the entire system. However, in actual operation, the traction inverter faces a variety of complex and severe challenges. The traction inverter is usually composed of multiple power semiconductor devices. These components will be affected by various factors such as current, voltage, and temperature during long-term operation, and gradually age. This aging process has significant randomness and uncertainty, and there is also interdependence between different components, which makes the life prediction of the entire traction inverter more complicated.
[0003] In the prior art, the life prediction of traction converters mostly adopts simplified models or data-driven fitting methods. These models often ignore the variable environmental factors in the actual system, and fail to reflect the complex mechanism of the entire degradation process of the traction converter, including the variability and dependence of the degradation process of internal semiconductor devices, as well as the randomness of failures, resulting in a large deviation between the prediction results and the actual situation.
[0004] Therefore, a more realistic life prediction method that is closer to the actual system is needed to improve the accuracy of the prediction and help to deeply understand the system behavior, thereby improving the maintenance strategy and ensuring the continuous and stable operation of the electric traction system. Summary of the invention
[0005] The present invention provides a method and system for predicting the life of a traction converter based on a probabilistic damage model, so as to solve the problems existing in the prior art.
[0006] In order to achieve the above object, the present invention is implemented by the following technical solutions:
[0007] In a first aspect, the present invention provides a method for predicting the life of a traction converter based on a probabilistic damage model, comprising:
[0008] S1: constructing a power loss model of a semiconductor device in a bridge arm of a traction converter, and constructing a thermal model of the semiconductor device based on the power loss model;
[0009] S2: Based on the physics of solder cracking failure, a degradation model of the semiconductor device is constructed using temperature as an input of the model;
[0010] S3: According to the heat conduction characteristics, the degradation factor, the correlation variable relationship matrix and the dependence coefficient matrix are constructed; and the interdependent degradation relationship model is constructed;
[0011] S4: constructing a probabilistic damage model based on the thermal model, the degradation model and the interdependent degradation relationship model;
[0012] S5: constructing a limit state equation based on the probabilistic damage model; and predicting the life of the traction converter based on the limit state equation.
[0013] In a second aspect, the present application provides a traction inverter life prediction system based on a probabilistic damage model, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in the first aspect when executing the computer program.
[0014] The present invention has the following beneficial effects:
[0015] The traction converter life prediction method based on the probabilistic damage model provided by the present application comprehensively considers multiple factors, including the variability and randomness of the degradation process of internal semiconductor devices, as well as the randomness of failures, and is suitable for the multi-variable factor environment of complex actual systems; the use of a probabilistic modeling method makes the model closer to the operation of the actual system, improving the authenticity and practicality of the model; by more realistically simulating the internal interdependent degradation relationship of the traction converter, it surpasses the simplification of component independence in the traditional model, improves the accuracy of life prediction, and helps to deeply understand the system behavior and improve maintenance strategies. At the same time, the traction converter life prediction method based on the probabilistic damage model of the present invention can also be applied to the life prediction of other converters with different topological structures. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0017] Figure 1 is a flow chart of a method for predicting the life of a traction converter based on a probabilistic damage model according to a preferred embodiment of the present invention;
[0018] Figure 2 is a schematic diagram of a traction converter topology according to a preferred embodiment of the present invention;
[0019] Figure 3 is a circuit diagram of a thermal network model of a traction converter according to a preferred embodiment of the present invention;
[0020] Figure 4is a cross-sectional schematic diagram of a traction converter heat dissipation system according to a preferred embodiment of the present invention;
[0021] Figure 5 is a schematic diagram of the operating condition of a traction converter according to a preferred embodiment of the present invention;
[0022] Figure 6 is a diagram of average temperatures of various units in operation of a traction converter according to a preferred embodiment of the present invention;
[0023] Figure 7 is a diagram showing changes in safety margin during operation of a traction converter according to a preferred embodiment of the present invention;
[0024] Figure 8 This is a diagram of the life prediction results of the traction converter according to the preferred embodiment of the present invention. DETAILED DESCRIPTION
[0025] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0026] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.
[0027] See also Figure 1 The present invention provides a method for predicting the life of a traction converter based on a probabilistic damage model, comprising:
[0028] S1: constructing a power loss model of a semiconductor device in a bridge arm of a traction converter, and constructing a thermal model of the semiconductor device based on the power loss model;
[0029] S2: Based on the physics of solder cracking failure, a degradation model of the semiconductor device is constructed using temperature as an input of the model;
[0030] S3: According to the heat conduction characteristics, the degradation factor, the correlation variable relationship matrix and the dependence coefficient matrix are constructed; and the interdependent degradation relationship model is constructed;
[0031] S4: constructing a probabilistic damage model based on the thermal model, the degradation model and the interdependent degradation relationship model;
[0032] S5: constructing a limit state equation based on the probabilistic damage model; and predicting the life of the traction converter based on the limit state equation.
[0033] The above-mentioned traction converter life prediction method based on the probabilistic damage model can accurately predict the remaining service life of various traction converters. This method can flexibly set a variety of power device models and test conditions; according to different prediction requirements, accurate life prediction can be performed for different operating conditions and operating loads without the need for physical damage experiments or relying on a large amount of historical data of the degradation process; this method can provide extremely realistic and reliable life prediction results for the monitoring and operation and maintenance of various converter power systems containing power devices. At the same time, the traction converter life prediction method based on the probabilistic damage model of the present invention shows a wide range of applicability. It is not only suitable for traction converters, but can also be applied to other converter life prediction scenarios containing power device structures and with different topological structures, providing strong support for the health management and maintenance strategy formulation of various converter power systems.
[0034] In the complete example, first, see Figure 2 and Figure 3 , construct the power loss model and thermal model of the semiconductor devices in the traction converter bridge arm, specifically:
[0035] The power conversion of the traction converter mainly includes two parts, namely the rectifier part and the inverter part. The phase bridge arm of the rectifier part and the inverter part is denoted as a finite set W = {1, 2, Lz, ..., Z}, where z is the phase bridge arm serial number and Z is the total number of phase bridge arms. The value of z can be 1 to Z, representing the phase bridge arm of the rectifier part and the phase bridge arm of the inverter part respectively. The phase bridge arm is composed of semiconductor devices such as power devices and diodes. The semiconductor devices in the zth phase bridge arm are denoted as a finite set W z ={1,2,Ln,…,N}, n is the serial number of the semiconductor device, N is the total number of semiconductor devices, and the value of n can be 1 to N, which represent the power device, diode and clamping diode of the phase bridge arm from top to bottom.
[0036] Constructing a power loss model for semiconductor devices in the traction converter bridge arm:
[0037] Calculate the power loss of the nth semiconductor device in the zth phase bridge arm using the formula:
[0038]
[0039] in, is the power loss of the nth semiconductor device in the zth phase bridge arm, i z is the conduction current at the AC end of the z-phase bridge arm; f sw is the switching frequency; and They are respectively the conduction current i of the nth semiconductor device in the zth phase bridge arm z The conduction voltage drop and switching loss under Γ z,n and χ z,n are the average duty cycle and conduction state of the nth semiconductor device in the zth phase bridge arm operating mode, respectively.
[0040] Building a thermal model of semiconductor devices in the traction converter bridge arm:
[0041] Considering the junction temperature of the semiconductor device as the temperature of the solder under the chip, the thermal model of the nth semiconductor device in the zth phase bridge arm is constructed, and the formula is:
[0042] T z,n =P loss,z,n (Z z,n,jc +Z z,n,ch )+T z h (2)
[0043] Among them, T z,n , P loss,z,n , Z z,n,jc and Z z,n,ch are the temperature, power loss, thermal resistance from chip to substrate, and thermal resistance from substrate to heat sink of the nth semiconductor device in the zth phase bridge arm, respectively. z h is the temperature at the heat sink of the z-th phase bridge arm.
[0044] Based on the physics of solder cracking failure, a degradation model of semiconductor devices is constructed with temperature as the input of the model:
[0045] Taking the temperature of semiconductor devices as input, the Manson-Coffin life equation is used to describe the degradation process of semiconductor devices. The formula is:
[0046]
[0047] Among them, L z,n is the crack length of the nth semiconductor device in the zth phase bridge arm after degradation, ΔT z,n , and ΔN z,n They are the temperature of the nth semiconductor device in the zth phase bridge arm at T z,iThe temperature fluctuation range, average temperature and crack length under the condition of L z,i The number of cycles required when L,z,n is the estimated error of the crack length of the nth semiconductor device in the zth phase bridge arm; a and b are the estimated coefficients of the crack length respectively; Δε p,z,n is the solder of the nth semiconductor device in the zth phase bridge arm at ΔT z,n , The accumulated plastic strain under , expressed as:
[0048]
[0049] Where A, B, Q, and R are the estimated coefficients of accumulated plastic strain; Δα is the difference in thermal expansion coefficient between solder and chip; θ ε,z,n is the estimated error of the accumulated plastic strain of the nth semiconductor device in the zth phase bridge arm.
[0050] Formulas (3) to (4) constitute the degradation model of semiconductor devices in the bridge arm of the traction converter.
[0051] See also Figure 4 ,According to the heat conduction characteristics, the degradation factor, the correlation variable relationship matrix and the dependence coefficient matrix are constructed;,and the mutual dependence degradation relationship model is constructed;
[0052] Construct the degradation factor:
[0053] Construct the semiconductor device degradation factor, the formula is:
[0054]
[0055] Among them, Θ z.n represents the degradation factor of the nth semiconductor device in the zth phase bridge arm; It means to take the derivative of the variable in the next bracket with respect to the variable in the previous bracket.
[0056] Construct the association variable relationship matrix:
[0057] The degraded semiconductor device is used as the source device, and the semiconductor device affected by degradation is used as the target device. When the nth semiconductor device in the zth phase bridge arm is used as the target device, the associated variable relationship matrix is constructed. It is expressed as:
[0058]
[0059] in, is a vector of associated variables with all semiconductor devices in the p-th phase bridge arm as source devices to target devices; is the associated variable of the kth semiconductor device in the pth phase bridge arm as the source device to the target device; p belongs to the phase bridge arm set, that is, p∈W, W={1,2,L,Z}; k belongs to the semiconductor set, that is, k∈W z , W z ={1,2,L,N}. When p≠z and k≠n, it indicates that the source device and the target device are not the same device; when p=z and k=n, it indicates that the source device and the target device are the same device.
[0060] Associated variables Initialized to 0. When the active device degrades, the associated variable Updated to:
[0061]
[0062] Among them, Z p,k,jc and P loss,p,k are the thermal resistance and power loss from the chip to the substrate of the kth semiconductor device in the pth phase bridge arm respectively; ε Ploss,p.k and are the power loss and thermal resistance increased due to the degradation of the kth semiconductor device in the p-phase bridge arm, respectively, expressed as:
[0063]
[0064] in, and is the temperature T p.k and the conduction current i p The on-state voltage drop increment and switching loss increment of the kth semiconductor device in the p-th phase bridge arm under the condition are obtained according to the device data sheet; Γ z,n and χ z,n are the average duty cycle and conduction state of the nth semiconductor device in the zth phase bridge arm working mode, respectively; is the damage degree of the kth semiconductor device in the pth phase bridge arm at time t; k is the incremental coefficient of thermal resistance of the kth semiconductor device degradation; is the nonlinear relationship between the thermal resistance and degradation of the kth semiconductor device in the pth phase bridge arm, usually an exponential function, expressed as:
[0065]
[0066] Formulas (6) to (10) constitute the correlation variable relationship matrix.
[0067] Construct the dependency coefficient matrix:
[0068] When the nth semiconductor device in the zth phase bridge arm is used as the target device, the dependence coefficient matrix C z,n It is expressed as:
[0069]
[0070] in, is the dependence coefficient of all semiconductor devices in the p-th phase bridge arm as source devices on the target device; is the dependence coefficient of the kth semiconductor device in the pth phase bridge arm as a source device on the target device, expressed as:
[0071]
[0072] in, is the random variable of the dependence of the kth semiconductor device in the pth phase bridge arm as the source device on the target device, which is related to the strain rate sensitivity of the unit material; It is the thermal diffusion coefficient of the kth semiconductor device in the p-phase bridge arm as the source device to the target device.
[0073] Formulas (11) to (12) form a dependency coefficient matrix.
[0074] Constructing a mutual dependence degradation relationship model:
[0075] The interdependent degradation relationship model of damage increment and semiconductor device degradation factor, associated variable relationship matrix and dependency coefficient matrix is constructed and expressed as:
[0076]
[0077] Among them, z,n is the damage increment of the nth semiconductor device in the zth phase bridge arm caused by source device degradation.
[0078] Formulas (5) to (13) constitute the interdependent degradation relationship model.
[0079] Based on the thermal model, degradation model and interdependent degradation relationship model, a probabilistic damage model is constructed:
[0080] The probabilistic damage of the traction converter is denoted as a finite set is the damage degree of the nth semiconductor device in the zth phase bridge arm at time t, expressed as:
[0081]
[0082] in, is the initial state of the nth semiconductor device in the zth phase bridge arm at t = 0, η x,z,n is the estimated error of the damage degree of the nth semiconductor device in the zth phase bridge arm, calculated as:
[0083] η x,z,n =(θ ε,z,n) b θ L,z,n (15)
[0084] Λ(t,T z,n ) is a nonlinear relationship related to the physics of unit material failure and is expressed as:
[0085] Λ(t,T z,n )=Λ(t-Δt,T z,n )+L z,n (16)
[0086] Where Δt is the time corresponding to the number of cycles ΔN under the same cumulative plastic strain.
[0087] Formulas (14) to (16) constitute the probabilistic damage model of the traction converter.
[0088] Based on the constructed probabilistic damage model, the limit state equation is constructed to predict the life of the traction converter:
[0089] According to the semiconductor device model used in the traction converter, the crack length failure standard of the nth semiconductor device in the zth phase bridge arm is defined as L fail,z,n , construct the limit state equation, expressed as:
[0090]
[0091] Among them, M z,n (U, t) represents the safety margin of the nth semiconductor device in the zth phase bridge arm at time t, M z,n When (U,t)≤0, the nth semiconductor device in the zth phase bridge arm fails, and U is the uncertainty set, that is, θ D represents the uncertainty of the solder failure threshold;
[0092] Traction converter life prediction:
[0093] According to the limit state equation, the failure probability of the nth semiconductor device in the zth phase bridge arm at time t is P F,z,n , the formula is:
[0094]
[0095] Among them, f U,z,n (u,t) is the failure probability density function; u is an element of the set U; the set W F is the failure set, which satisfies M z,n (U,t)≤0.
[0096] The average life of the traction converter is MTTF s, the average life of the z-phase bridge arm is MTTF z , the average life of the nth semiconductor device in the zth phase bridge arm is MTTF z,n , the calculation formulas are:
[0097]
[0098] Among them, min{} means taking the minimum value of the elements in the brackets.
[0099] The reliability of the traction converter at time t is R s (t), the reliability of the z-th phase bridge arm at time t is R z (t), the reliability of the nth semiconductor device in the zth phase bridge arm at time t is R z,n (t), the calculation formulas are:
[0100]
[0101] Formulas (1) to (24) constitute a traction converter life prediction method based on a probabilistic damage model.
[0102] The traction converter of the above embodiment refers to the traction converter in the CRH2 EMU, with a speed range of 150-300 km / h and operating conditions such as Figure 5 The selected power device modules are IGBT module 5SNA1200E330100 and diode module 5SDD52N4000. The specific parameters of the experiment are shown in Table 1 and Table 2.
[0103] Table 1 Experimental parameters
[0104]
[0105] Table 2 Experimental parameters 2
[0106]
[0107]
[0108] Among them, the average temperature of each unit of each component during the operation of the traction converter is as follows: Figure 6 As shown in the figure, the operating temperature of different units under different converter properties, such as rectification or inversion, is described. It can be seen that the average temperature of units 2, 3, 6, and 7 of components 1 and 2 reaches more than 70 degrees. However, the life of units 6 and 7 is significantly lower than that of units 2 and 3. This is related to the failure definition. Units 6 and 7 are diode modules with longer failure crack definition lengths. The safety margin changes during the operation of the traction converter are shown in Figure 2. Figure 7 As shown in Figure 2, it can be seen that the traction converter will fail at the earliest in 4.5 years. The results of the traction converter life prediction are shown in Figure 2. Figure 8 As shown, the remaining service life of different units under different converter properties is described.
[0109] The embodiment of the present application also provides a traction converter life prediction system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the steps of the method. The traction converter life prediction system can implement various embodiments of the traction converter life prediction method described above, and can achieve the same beneficial effects, which will not be described in detail here.
[0110] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A method for predicting the life of traction converter based on probabilistic damage model. It is characterized in that include: S1: constructing a power loss model of a semiconductor device in a bridge arm of a traction converter, and constructing a thermal model of the semiconductor device based on the power loss model; S2: Based on the physics of solder cracking failure, a degradation model of the semiconductor device is constructed using temperature as an input of the model; S3: According to the heat conduction characteristics, the degradation factor, the correlation variable relationship matrix and the dependence coefficient matrix are constructed; and the interdependent degradation relationship model is constructed; S4: constructing a probabilistic damage model based on the thermal model, the degradation model and the interdependent degradation relationship model; S5: constructing a limit state equation based on the probabilistic damage model; and predicting the life of the traction converter based on the limit state equation.
2. The method for predicting the life of a traction converter based on a probabilistic damage model according to claim 1, It is characterized in that The S1 includes: S11: Assume that the power conversion of the traction converter includes a rectifier part and an inverter part. The phase bridge arms of the rectifier part and the inverter part are recorded as a finite set W = {1, 2, Lz, ..., Z}, where z is the phase bridge arm sequence number, Z is the total number of phase bridge arms, and the value of z is 1 to Z, which represent the phase bridge arms of the rectifier part and the phase bridge arms of the inverter part respectively. The phase bridge arms are composed of semiconductor devices such as power devices and diodes. The semiconductor devices in the zth phase bridge arm are recorded as a finite set W z ={1,2,Ln,…,N}, n is the serial number of the semiconductor device, N is the total number of semiconductor devices, and the value of n is 1 to N, which represents the power device, diode and clamping diode from top to bottom of the phase bridge arm in turn; S12: Construct a power loss model for semiconductor devices in the bridge arm of a traction converter; Calculate the power loss of the nth semiconductor device in the zth phase bridge arm to satisfy the following relationship: in, is the power loss of the nth semiconductor device in the zth phase bridge arm, i z is the conduction current at the AC end of the z-phase bridge arm; f sw is the switching frequency; and They are respectively the conduction current i of the nth semiconductor device in the zth phase bridge arm z The conduction voltage drop and switching loss under Γ z,n and χ z,n are the average duty cycle and conduction state of the nth semiconductor device in the zth phase bridge arm working mode, respectively; S13: constructing a thermal model of semiconductor devices in the bridge arm of the traction converter based on the power loss of the nth semiconductor device in the bridge arm of the zth phase; Considering the junction temperature of the semiconductor device as the temperature of the solder under the chip, a thermal model of the nth semiconductor device in the zth phase bridge arm is constructed to satisfy the following relationship: Among them, T z,n , P loss,z,n , Z z,n,jc and Z z,n,ch are the temperature, power loss, thermal resistance from chip to substrate, and thermal resistance from substrate to heat sink of the nth semiconductor device in the zth phase bridge arm, respectively. is the temperature at the heat sink of the z-th phase bridge arm.
3. The method for predicting the life of a traction converter based on a probabilistic damage model according to claim 1, It is characterized in that The S2 includes: S21: Taking the temperature of semiconductor devices as input, the Manson-Coffin life equation is used to describe the degradation process of semiconductor devices, satisfying the following relationship: Among them, L z,n is the crack length of the nth semiconductor device in the zth phase bridge arm after degradation, ΔT z,n , and ΔN z,n They are the temperature of the nth semiconductor device in the zth phase bridge arm at T z,i The temperature fluctuation range, average temperature and crack length under the condition of L z,i The number of cycles required for is the estimated error of the crack length of the nth semiconductor device in the zth phase bridge arm; a and b are the estimated coefficients of the crack length respectively; Δε p,z,n is the solder of the nth semiconductor device in the zth phase bridge arm at ΔT z,n , The accumulated plastic strain under , satisfies the following relationship: Where A, B, Q and R are the estimated coefficients of accumulated plastic strain; Δα is the difference in thermal expansion coefficient between solder and chip; is the estimated error of the accumulated plastic strain of the nth semiconductor device in the zth phase bridge arm; S22: The degradation model of semiconductor devices in the bridge arm of the traction converter is constructed by formulas (3) to (4).
4. The method for predicting the life of a traction converter based on a probabilistic damage model according to claim 1, It is characterized in that The S3 includes: S31: Construct the semiconductor device degradation factor to satisfy the following relationship: Among them, Θ z.n represents the degradation factor of the nth semiconductor device in the zth phase bridge arm; It means to take the derivative of the variable in the previous bracket with respect to the variable in the next bracket. a and b are the estimated coefficients of crack length, A, B, Q and R are the estimated coefficients of accumulated plastic strain. and ΔN z,n They are the temperature of the nth semiconductor device in the zth phase bridge arm at T z,i The average temperature and crack length under z,i The number of cycles required for S32: construct the correlation variable relationship matrix; The degraded semiconductor device is used as the source device, and the semiconductor device affected by the degradation is used as the target device. When the nth semiconductor device in the zth phase bridge arm is used as the target device, the associated variable relationship matrix is constructed. Satisfies the following relationship: in, is a vector of associated variables with all semiconductor devices in the p-th phase bridge arm as source devices to target devices; is the associated variable of the kth semiconductor device in the pth phase bridge arm as the source device to the target device; p belongs to the phase bridge arm set, that is, p∈W, W={1,2,L,Z}; k belongs to the semiconductor set, that is, k∈W z , W z ={1,2,L,N}, when p≠z and k≠n, it means that the source device and the target device are not the same device; when p=z and k=n, it means that the source device and the target device are the same device; Associated variables Initialized to 0, when the active device degrades, the associated variable Updated to: Among them, Z p,k,jc and P loss,p,k are the chip-to-substrate thermal resistance and power loss of the kth semiconductor device in the pth phase bridge arm, respectively; and are the power loss and thermal resistance increased due to the degradation of the kth semiconductor device in the pth phase bridge arm, respectively, satisfying the following relationship: in, and is the temperature T p.k and the conduction current i p The on-state voltage drop increment and switching loss increment of the kth semiconductor device in the p-th phase bridge arm under the condition are obtained according to the device data sheet; Γ z,n and χ z,n are the average duty cycle and conduction state of the nth semiconductor device in the zth phase bridge arm working mode, respectively; is the damage degree of the kth semiconductor device in the pth phase bridge arm at time t; k is the incremental coefficient of thermal resistance of the kth semiconductor device degradation; is the nonlinear relationship between the thermal resistance and degradation of the kth semiconductor device in the pth phase bridge arm, usually an exponential function, expressed as: Formulas (6) to (10) constitute the correlation variable relationship matrix; S33: construct dependency coefficient matrix; When the nth semiconductor device in the zth phase bridge arm is used as the target device, the dependency coefficient matrix Satisfies the following relationship: in, is the dependence coefficient of all semiconductor devices in the p-th phase bridge arm as source devices on the target device; is the dependence coefficient of the kth semiconductor device in the pth phase bridge arm as a source device on the target device, satisfying the following relationship: in, is the random variable of the dependence of the kth semiconductor device in the pth phase bridge arm as the source device on the target device, which is related to the strain rate sensitivity of the unit material; is the thermal diffusion coefficient of the kth semiconductor device in the p-phase bridge arm as the source device to the target device; The dependence coefficient matrix is formed by formulas (11) to (12); S34: Constructing a model of interdependent degradation relationships; The interdependent degradation relationship model of damage increment and semiconductor device degradation factor, associated variable relationship matrix and dependency coefficient matrix is constructed to satisfy the following relationship: Among them, z,n is the damage increment of the nth semiconductor device in the zth phase bridge arm caused by source device degradation; The interdependent degradation relationship model is constructed by formulas (5) to (13).
5. The method for predicting the life of a traction converter based on a probabilistic damage model according to claim 1, It is characterized in that The S4 includes: The probabilistic damage of the traction converter is denoted as a finite set is the damage degree of the nth semiconductor device in the zth phase bridge arm at time t, satisfying the following relationship: in, is the initial state of the nth semiconductor device in the zth phase bridge arm at t = 0, η x,z,n is the estimated error of the damage degree of the nth semiconductor device in the zth phase bridge arm, calculated as: Among them, Λ(t,T z,n ) is a nonlinear relationship related to the physics of unit material failure and is expressed as: Λ(t,T z,n )=Λ(t-Δt,T z,n )+L z,n (16) Where Δt is the time corresponding to the number of cycles ΔN under the same cumulative plastic strain; The probabilistic damage model of traction converter is constructed by formulas (14) to (16).
6. The method for predicting the life of a traction converter based on a probabilistic damage model according to claim 1, It is characterized in that The S5 includes: S51: According to the semiconductor device model used in the traction converter, the crack length failure standard of the nth semiconductor device in the zth phase bridge arm is defined as L fail,z,n , construct the limit state equation and satisfy the following relationship: Among them, M z,n (U, t) represents the safety margin of the nth semiconductor device in the zth phase bridge arm at time t, M z,n When (U,t)≤0, the nth semiconductor device in the zth phase bridge arm fails, and U is the uncertainty set, that is, represents the uncertainty of the solder failure threshold; S52: performing life prediction of traction converter; According to the limit state equation, the failure probability of the nth semiconductor device in the zth phase bridge arm at time t is P F,z,n , satisfying the following relationship: Among them, f U,z,n (u,t) is the failure probability density function; u is an element of the set U; the set W F is the failure set, which satisfies M z,n (U,t)≤0 is composed of all elements u; The average life of the traction converter is MTTF s , the average life of the z-phase bridge arm is MTTF z , the average life of the nth semiconductor device in the zth phase bridge arm is MTTF z,n , the calculation formulas are: Among them, min{} means taking the minimum value of the elements in the brackets; The reliability of the traction converter at time t is R s (t), the reliability of the z-th phase bridge arm at time t is R z (t), the reliability of the nth semiconductor device in the zth phase bridge arm at time t is R z,n (t), the calculation formulas are: Life prediction is performed based on equations (1) to (24).
7. A traction converter life prediction system based on a probabilistic damage model, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method described in any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Method and system for predicting service life of IGBT module of power supply under passenger car
CN113591336A
Power semiconductor device service life prediction method based on multi-fatigue mode coupling
CN113987783A
Method for estimating service life of semiconductor device with multiple degradation effects
CN116401875A
Semiconductor device service life estimation method based on multi-degradation effect model
CN116401876A
Ammonia arc thruster service life evaluation method and system
CN118468734A
Cited By
Building outer wall damage detection method based on infrared thermal imaging
CN120629260A
