CN113449461A - Switch device aging equivalent method - Google Patents

Switch device aging equivalent method Download PDF

Info

Publication number
CN113449461A
CN113449461A CN202110639520.7A CN202110639520A CN113449461A CN 113449461 A CN113449461 A CN 113449461A CN 202110639520 A CN202110639520 A CN 202110639520A CN 113449461 A CN113449461 A CN 113449461A
Authority
CN
China
Prior art keywords
switching device
aging
switching
current
total loss
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110639520.7A
Other languages
Chinese (zh)
Inventor
孟苓辉
周振威
陈义强
陈媛
贺致远
何世烈
刘俊斌
时林林
俞鹏飞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Electronic Product Reliability and Environmental Testing Research Institute
Original Assignee
China Electronic Product Reliability and Environmental Testing Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Electronic Product Reliability and Environmental Testing Research Institute filed Critical China Electronic Product Reliability and Environmental Testing Research Institute
Priority to CN202110639520.7A priority Critical patent/CN113449461A/en
Publication of CN113449461A publication Critical patent/CN113449461A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/04—Ageing analysis or optimisation against ageing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Electronic Switches (AREA)

Abstract

The invention relates to the technical field of switch device aging experiments, and discloses a switch device aging equivalent method, which comprises the steps of adopting different equivalent algorithms to carry out iterative optimization on modulation parameters of a switch device; calculating the total loss of the switching device in the current mode aging experiment in each iteration; and searching the optimal solution of the modulation parameter of the switching device by taking the minimum value of the total loss difference value of the switching device under the current type and voltage type aging experiments as a target function. When the aging equivalent method of the switching device provided by the invention is applied to a current type aging experiment, different equivalent algorithms are adopted to determine the optimal control strategy of the current type aging method, so that the current type aging method can achieve the same aging effect as a voltage type aging method. Meanwhile, the current type aging method can reflect the aging process and the residual service life of the IGBT module under the actual load working condition, and the aging experiment aiming at the switching device is simpler, the cost is lower, and the operation reliability is higher.

Description

Switch device aging equivalent method
Technical Field
The invention relates to the technical field of switch device aging experiments, in particular to a switch device aging equivalent method.
Background
The traction converter is a core power device of a train, and an Insulated Gate Bipolar Transistor (IGBT) is a device with high failure rate in the traction converter. The service characteristics of the IGBT device are researched, the performance degradation trend of the IGBT device is known, a reasonable maintenance plan is favorably made, and the problem of unnecessary cost waste caused by excessive maintenance is avoided. The reliability analysis and service life prediction of the IGBT device mostly adopt voltage type space vector modulation IGBT aging experimental device and method. The duty ratio, modulation depth and aging current of a modulation strategy adopted by the voltage type aging method are usually fixed or can only change in small amplitude, and parameters of a device are changed in large amplitude under the actual working condition. Therefore, the voltage type space vector modulation IGBT aging method cannot reflect the aging process and the residual service life of the IGBT module under the actual load working condition.
Disclosure of Invention
Therefore, it is necessary to provide an equivalent aging method for a switching device for the problem that the voltage-type space vector modulation IGBT aging method cannot reflect the aging process and the residual service life of the IGBT module under the actual load working condition.
A switch device aging equivalent method comprises the steps of adopting different equivalent algorithms to carry out iterative optimization on modulation parameters of a switch device; calculating the total loss of the switching device in the current mode aging experiment in each iteration; and searching the optimal solution of the modulation parameter of the switching device by taking the minimum value of the total loss difference value of the switching device under the aging experiments of a current mode and a voltage mode as an objective function.
According to the aging equivalence method of the switching device, different equivalence algorithms are adopted to conduct iteration optimization on modulation parameters of the switching device, in each iteration, the total loss of the switching device in a current type aging experiment under the condition of the modulation parameters obtained by current iteration is calculated, the minimum value of the total loss difference value of the switching device in the current type aging experiment and the voltage type aging experiment is used as a target function, and the optimal solution of the modulation parameters of the switching device is found. When the aging equivalent method of the switching device provided by the invention is applied to a current type aging experiment, different equivalent algorithms are adopted to determine the optimal control strategy of the current type aging method, so that the current type aging method can achieve the same aging effect as a voltage type aging method. Meanwhile, the current type aging method can reflect the aging process and the residual service life of the IGBT module under the actual load working condition, and the aging experiment aiming at the switching device is simpler, the cost is lower, and the operation reliability is higher.
In one embodiment, the iteratively optimizing the modulation parameters of the switching device by using different equivalent algorithms includes iteratively optimizing the modulation parameters of the switching device by using a genetic algorithm on a switching cycle scale of the switching device; on the modulation wave period scale of the switching device, iterative optimization is carried out on the modulation parameters of the switching device by adopting a particle swarm algorithm; and on the operation period scale of the switching device, carrying out iterative optimization on the modulation parameters of the switching device by adopting a simulated annealing algorithm.
In one embodiment, the calculating the total loss of the switching device in the current mode aging experiment in each iteration comprises calculating the total loss of the switching device in one switching period in the current mode aging experiment; and calculating the total loss of the switching device in one modulation wave period in the current-mode aging experiment.
In one embodiment, the modulation parameters include a switching frequency of the switching device, a duty cycle of the switching device, and an aging current output by the aging current source.
In one embodiment, the calculating the total loss of the switching device in one switching period in the current-mode aging experiment includes fitting a static curve of the switching device to obtain a linear expression of collector-emitter voltage and collector current; calculating the on-state loss of the switching device in one switching period; acquiring the waveforms of the voltage and the current of the switching device in the switching-on and switching-off processes, and performing curve fitting in a segmented manner; integrating the loss of each section of curve to obtain the switching loss in one switching period; and acquiring the total loss of the switching device in one switching period according to the on-state loss and the switching loss.
In one embodiment, the calculating the total loss of the switching device in the current-mode aging experiment comprises calculating an on-state loss and a switching loss of the switching device in one modulation wave period respectively; and acquiring the total loss of the switching device in one modulation wave period according to the on-state loss and the switching loss.
In one embodiment, the iteratively optimizing the modulation parameter of the switching device by using a genetic algorithm on the scale of the switching period of the switching device includes initializing the modulation parameter of the switching device in the current-mode aging experiment system; updating the modulation parameters in real time according to the junction temperature change of the switching device, and calculating the total loss of the switching device in the current type aging experiment system; judging whether the error between the total loss and a preset total loss is within a preset error range or not; if the error between the total loss and the preset total loss is within a preset error range, defining the current modulation parameter as an optimal solution and outputting the optimal solution; otherwise, the modulation parameters are subjected to genetic operation of selection and variation, and the loss is calculated again until the error between the new total loss and the preset total loss is within the preset error range.
In one embodiment, the preset total loss is obtained according to an aging experiment result of the voltage type aging experiment system.
In one embodiment, the performing iterative optimization on the modulation parameters of the switching device by using a particle swarm optimization algorithm on the modulation wave period scale of the switching device includes solving the obtained solution set according to the genetic algorithm to initialize a particle swarm; calculating the fitness of the particle swarm; judging whether the fitness is within an error range; if the fitness is within the error range, defining the modulation parameters in the current particle swarm as an optimal solution and outputting the optimal solution; otherwise, updating the speed and the position of the particle swarm to generate a new particle swarm, and calculating the fitness of the new particle swarm again until the new fitness is within the error range.
In one embodiment, the iterative optimization of the modulation parameters of the switching device by using a simulated annealing algorithm on the scale of the operation cycle of the switching device includes initializing the modulation parameters and the temperature, and calculating the total loss difference under two aging experiments of a current mode and a voltage mode in one train cycle; generating a new parameter value combination, calculating a new total loss difference value and the difference between the new total loss difference value and the previous total loss difference value, and acquiring a judgment value; comparing the judgment value with a preset value; if the judgment value is smaller than or equal to a preset value, combining the new parameter values to serve as initialization data of next simulation; if the judgment value is larger than the preset value, calculating the new point receiving probability; selecting a random number and comparing the new point acceptance probability with the random number; if the new point receiving probability is larger than or equal to the random number, combining the new parameter values to serve as initialization data of next simulation; otherwise, abandoning the new parameter value combination and taking the original parameter value combination as the initialization data of the next simulation.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the specification, and other drawings can be obtained by those skilled in the art without inventive labor.
FIG. 1 is a schematic flow chart of a method for equivalent aging method of a switching device according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a method for selecting an equivalent algorithm according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating a method for calculating total loss in a current-mode aging test according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating a method for calculating total loss during a switching cycle according to an embodiment of the present invention;
FIG. 5 is a flowchart illustrating a method for calculating the total loss in a modulation wave period according to an embodiment of the present invention;
FIG. 6 is a flowchart illustrating a method for iterative optimization using a genetic algorithm according to an embodiment of the present invention;
FIG. 7 is a flowchart illustrating a method for iterative optimization using a particle swarm optimization algorithm according to an embodiment of the present invention;
fig. 8 is a flowchart illustrating a method for iterative optimization by using a simulated annealing algorithm according to an embodiment of the present invention.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The IGBT is a composite fully-controlled voltage-driven power Semiconductor device composed of a BJT (Bipolar Junction Transistor) and an MOS (insulated gate Field Effect Transistor), and has the advantages of both high input impedance of an MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) and low on-state voltage drop of a gtr (giant Transistor). The failure rate of the IGBT influences the failure rate of the traction converter, so that the research on the service characteristics and the performance degradation trend of the IGBT is very important. The aging of the IGBT module is mainly caused by the fact that heating power is generated in the working process, and different materials of all layers of the module lead to different thermal expansion coefficients of all layers, so that the thermal stress of all layers is different. A material having a small thermal expansion coefficient is subjected to tensile stress and is thus stretched; a material having a large thermal expansion coefficient is subjected to a compressive stress and is thus compressed. Under the action of cyclic thermal shock, thermal resistance is increased, junction temperature is increased more quickly, cracking of a solder layer and falling and breaking of a bonding wire are caused, vicious circle is formed, and finally, failure of a device is caused. According to the failure mechanism, the aging of the IGBT module is determined by the heating power.
At present, voltage type space vector modulation IGBT aging experimental device and method are mainly adopted for reliability analysis and service life prediction of IGBT devices. The duty cycle, modulation depth and aging current of the modulation strategy adopted by the voltage type aging method are generally fixed or vary within a small amplitude range. However, because the parameters of the device are greatly changed all the time under the actual working condition, the aging process and the remaining service life of the IGBT module under the actual load condition cannot be reflected by the result of the voltage type aging experiment. In addition, the traditional voltage type aging experiment has high requirements on the power supply voltage and the power and expensive experiment cost when simulating the actual working condition.
In order to solve the drawbacks of the conventional aging experiment, the present invention provides an equivalent aging method for a switching device, and fig. 1 is a schematic flow chart of the equivalent aging method for a switching device according to an embodiment of the present invention, where the method includes the following steps S100 to S300.
Step S100: and (4) carrying out iterative optimization on the modulation parameters of the switching device by adopting different equivalent algorithms.
From the perspective of different layers, different algorithms are respectively adopted to carry out iterative optimization on the values of the modulation parameters of the switching device, and the values of the modulation parameters in the control strategy are improved through the nested use of the different algorithms.
Step S200: the total loss of the switching device in the current-mode aging experiment was calculated at each iteration.
When iterative optimization is carried out on the values of the modulation parameters of the switching device by adopting different algorithms, the total loss of the switching device is calculated when a current type aging experiment is carried out under the condition of setting different modulation parameters. And determining an algorithm stopping condition in iterative optimization by taking the total loss of the switching device as a judgment standard. The current type aging system has the advantages of simple structure, low device cost and small occupied space. The current type aging system is adopted to carry out aging experiments on the switching device, and the defects that the traditional aging experiments are high in experiment cost, the modulation strategy is not accordant with the actual working condition and the like can be overcome.
Step S300: and searching the optimal solution of the modulation parameter of the switching device by taking the minimum value of the total loss difference value of the switching device under the current type and voltage type aging experiments as a target function.
In order to ensure that the current type aging method can achieve the same aging effect as the voltage type aging method, the minimum value of the total loss difference value of the switching device under the current type aging experiment and the voltage type aging experiment is used as a target function to find the optimal solution of the modulation parameter of the switching device. When the value of the modulation parameter of the switching device is iterated by using an equivalent algorithm, the value of a group of modulation parameters is obtained every iteration. And taking the value as a control strategy of the current type aging experiment to carry out the aging experiment on the switching device, and acquiring the total loss of the switching device at the moment.
When the difference value between the total loss of the switching device in the two current-type aging experiments and the total loss of the switching device in the two voltage-type aging experiments reaches a preset condition, the current-type aging method can achieve the same aging effect as the voltage-type aging method, and the value at the moment is judged to be the optimal solution of the modulation parameter. When the switching device aging equivalent method provided by the embodiment is applied to a current type aging experiment, different equivalent algorithms are adopted to determine the optimal control strategy of the current type aging method, so that the current type aging method can achieve the same aging effect as a voltage type aging method. Meanwhile, the current type aging method can reflect the aging process and the residual service life of the IGBT module under the actual load working condition, and the aging experiment aiming at the switching device is simpler, the cost is lower, and the operation reliability is higher.
Fig. 2 is a flowchart illustrating a method for selecting an equivalent algorithm according to an embodiment of the present invention, in which the iterative optimization of the modulation parameters of the switching device using different equivalent algorithms includes the following steps S110 to S130.
Step S110: and on the scale of the switching period of the switching device, iterative optimization is carried out on the modulation parameters of the switching device by adopting a genetic algorithm.
During the turn-on and turn-off process of the IGBT, the voltage variation range of the IGBT is related to the voltage of the direct current bus. Meanwhile, the product of the voltage and the current of the IGBT during the switching transition process determines the switching loss. The dc bus voltage of the Voltage Source Inverter (VSI) is therefore dependent on the voltage source, much higher than the dc bus voltage of the load dependent Current Source Inverter (CSI). This results in a different range of voltage variation during switching of the device. And the VSI has 3 switching tubes on at any moment, and the CSI has 2 switching tubes on. Therefore, in the Voltage Source Inverter (VSI), the loss difference of the switching devices is large in the switching period, and it is difficult to achieve real power equivalence.
As can be known from simulation analysis of the built inverter module, the average value and the variation trend of the junction temperature of the switching device depend on the operating condition of the train, and the slight fluctuation of the junction temperature is mainly influenced by the heating condition of the switching device in the modulation wave period and is less influenced by the heating condition in the switching period.
For the above analysis, by combining the characteristics of Genetic Algorithm (GA), a wide approximate optimal solution space of the modulation parameters of the switching device can be obtained on a switching cycle level, the selectivity of the solution is expanded, and a solution set infinitely close to the optimal solution can be explored under the condition that the optimal solution cannot be obtained. Meanwhile, in the aspect of convergence, the GA already has a mature convergence analysis method, and the convergence speed can be estimated, so that a genetic algorithm GA is used for solving a solution set of the optimal solution of the modulation parameters in a switching period.
Step S120: and performing iterative optimization on the modulation parameters of the switching device by adopting a particle swarm algorithm on the modulation wave period scale of the switching device.
The Particle Swarm Optimization (PSO) has memory, all particles with better solution information are stored, and the optimal solution obtained by the genetic algorithm GA in step S120 can be used as the historical optimal solution of the PSO. Since the modulation wave period is composed of several switching periods, the solution range is expanded. The particle swarm optimization PSO is a one-way information sharing mechanism, and the particles share information only through the optimal points searched currently. The whole searching and updating process of the particles is the process of following the current optimal solution. In most cases, the speed of converging the search particles in the particle swarm algorithm to the optimal point is faster than the speed of converging the evolutionary individuals in the genetic algorithm to the optimal solution, that is, the search efficiency is higher, so that the particle swarm algorithm is selected to solve the solution set of the optimal solution of the modulation parameters in one modulation wave period.
Step S130: and on the operation period scale of the switching device, the modulation parameters of the switching device are subjected to iterative optimization by adopting a simulated annealing algorithm.
Simulated Annealing (SAA) algorithms produce more accurate optimized solutions than the two previous algorithms, but the optimization algorithms run at a slower speed. Since the train working period is composed of a plurality of modulation wave periods, the range of the optimal solution on the train working period level is further expanded. The value range of the optimal solution of the modulation parameters can be partially reduced through the screening of the equivalent algorithms in the previous two times, so that the longer operation time of the simulated annealing algorithm can be compensated to a certain extent, and therefore the solution set of the optimal solution of the modulation parameters is solved by using the simulated annealing algorithm in a train working period.
In the embodiment, the optimal values of the modulation parameters are respectively searched from the 3 levels in one switching cycle, one modulation wave cycle and one train working cycle. And (3) judging the difference value of the current total loss and the reference value by calculating the loss equivalence of the IGBT switching device in the current type aging experiment so as to obtain the standard of the algorithm stopping condition. Each layer adopts different algorithms respectively, the value accuracy of the control strategy parameters is improved through the nested use of the different algorithms, and the optimal control strategy of the current type aging method is determined, so that the current type aging method can achieve the same aging effect as the voltage type aging method.
In one embodiment, the modulation parameters include a switching frequency of the switching device, a duty cycle, and an aging current output by the aging current source. The switching frequency, the duty ratio (namely modulation depth) of the switching device and the aging current of the aging current source are constraint conditions of a loss equivalence algorithm in the aging experiment method of the switching device, and the optimal solution of each modulation parameter is searched through iterative calculation. During the aging experiment, the switching frequency and the duty ratio (namely modulation depth) of the switching device and the aging current of the aging current source are adjusted to be the values of the optimal solution, so that the current type aging experiment system can be ensured to approach the practical application scene as much as possible.
Fig. 3 is a flowchart illustrating a method for calculating a total loss in a current-mode aging experiment according to an embodiment of the present invention, wherein calculating the total loss of the switching device in the current-mode aging experiment in each iteration includes the following steps S210 to S220.
Step S210: and calculating the total loss of the switching device in one switching period in the current-mode aging experiment.
The modulation wave period is composed of a plurality of switching periods, so that the total loss of the switching device in one switching period in a current-mode aging experiment needs to be calculated firstly.
Fig. 4 is a flowchart illustrating a method for calculating a total loss in a switching period according to an embodiment of the present invention, in which calculating a total loss of a switching device in a modulation wave period in a current-mode aging experiment includes the following steps S211 to S215.
Step S211: and fitting a static curve of the switching device to obtain a linear expression of the collector-emitter voltage and the collector current.
When the on-state loss of each switching device is calculated, curve fitting needs to be carried out on the static curve of the IGBT switching tube. After the static characteristic curve of the IGBT is fitted, a linear expression of the voltage of the collector and the emitter and the current of the collector of the IGBT can be obtained through derivation, and the expression is as follows:
UCE(sat)＝RTIC+UCEO；
wherein, UCE(sat)Is collector-emitter voltage, RTIs the forward on-resistance of the switching device, ICIs the collector current, UCEOIs the pull-up column voltage of the switching device. Wherein, the forward on-resistance RTHarmony column voltage UCEOAre related to the junction temperature of the switching device.
Step S212: the on-state losses of the switching devices within one switching cycle are calculated.
The on-state loss in one switching cycle can be calculated from a linear expression of the collector-emitter voltage and the collector current. The on-state loss over a switching cycle is calculated as follows:
PTcon＝UCE*IC＝(RTIC+UCEO)*IC＝f(IC,Tuj)；
wherein, PTconIs the on-state loss, U, in one switching cycleCEIs collector-emitter voltage, ICIs the collector current, RTIs the forward on-resistance of the switching device, UCEOIs the pull-up column voltage of the switching device, f (I)C,Tuj) As a function of power, TujIs the junction temperature of the switching device.
Step S213: and acquiring the waveforms of the voltage and the current of the switching device in the switching-on and switching-off processes, and performing curve fitting in a segmented manner.
In this embodiment, a double-pulse test experiment is used to test the switching device to obtain the waveforms of the voltage and the current of the switching device during the on and off processes, and the waveforms of the voltage and the current of the switching device are subjected to piecewise curve fitting respectively.
Step S214: and integrating the loss of each section of curve to obtain the switching loss in one switching period.
The waveforms of the voltage and the current of the IGBT in the switching-on and switching-off processes can be obtained through a double-pulse test experiment. And performing piecewise curve fitting on the waveforms of the voltage and the current of the IGBT according to the transient characteristics of the IGBT switch tube, and integrating the loss of each piecewise curve to calculate the loss energy under one switching state conversion. The calculation formula of the loss energy of the primary switching state is as follows:
Figure BDA0003106707170000111
wherein E isonFor the loss of energy when the switching device is switched on, EoffFor the loss energy at turn-off of the switching device, tonIs the on-time of the switching device, toffFor the turn-off time of the switching device, uCE(t) is a function of collector-emitter voltage, i (t) is a function of aging current, and t is time. In this embodiment, the gate voltage is raised to 10% of the driving voltage to the collectorThe time from the drop of the gate voltage to 90% of the drive voltage to the rise of the collector current to 10% of the test current is defined as the on-time, and the off-time.
Step S215: and acquiring the total loss of the switching device in one switching period according to the on-state loss and the switching loss.
Since the total loss of the switching device is equal to the sum of the on-state loss and the switching loss, the total loss of the switching device in one switching period can be obtained after the on-state loss and the switching loss of the switching device in one switching period are obtained.
Step S220: and calculating the total loss of the switching device in one modulation wave period in the current-mode aging experiment.
After the loss energy of the switching device in the primary switching period is obtained, the switching loss in one modulation wave period can be calculated according to the loss energy of the switching device in the primary switching period.
Fig. 5 is a flowchart illustrating a method for calculating a total loss in a modulation wave period according to an embodiment of the present invention, where in an embodiment, calculating a total loss of a switching device in a modulation wave period in a current-mode aging experiment includes the following steps S221 to S222.
Step S221: the on-state loss and the switching loss of the switching device in one modulation wave period are calculated separately.
Assuming that the output current i (t) is Isin ω t, the power factor is Φ, and the PWM modulation function is
Figure BDA0003106707170000121
Then
Figure BDA0003106707170000122
Where m is the modulation depth, t (t) is a pulse function, and when the switching device is turned on, t (t) is 1; when the switching device is turned off, t (t) is 0.
The on-state loss over one modulation wave period is calculated as:
Figure BDA0003106707170000123
wherein, PTcondIs the on-state loss, T, in one modulation wave period0Power frequency period, tonFor the on-time of the switching device, UCEFor collector-emitter voltage, i (t) is the output current, T (t) is the pulse function, and t is time.
The switching loss over one modulation wave period is calculated as:
Figure BDA0003106707170000131
wherein, PSWIs the switching loss in one modulation wave period, f is the switching frequency of the switching device, T0Is a power frequency period, TSWSwitching time of switching devices in one power frequency cycle, EonEnergy loss for conduction of the switching device, EoffThe loss energy for the switching device to turn off, t is the time. Wherein the energy loss E of one-time conduction of the switching deviceonAnd loss energy E of one turn-offoffAnd the type of the IGBT is determined.
Step S222: and acquiring the total loss of the switching device in one modulation wave period according to the on-state loss and the switching loss.
Since the total loss of the switching device is the sum of the on-state loss and the switching loss, the total loss of the switching device in one modulation wave period can be obtained after the on-state loss and the switching loss of the switching device in one modulation wave period are obtained.
Fig. 6 is a flowchart illustrating a method for iterative optimization by using a genetic algorithm according to an embodiment of the present invention, where in an embodiment, iterative optimization of a modulation parameter of a switching device by using a genetic algorithm on a switching cycle scale of the switching device includes the following steps S111 to S115.
In the embodiment, the minimum value of the total loss difference of the switching device in the current type aging mode and the voltage type aging mode is used as a target function, the switching frequency, the duty ratio (modulation depth) and the aging current are used as constraint conditions, and the genetic algorithm is adopted to iterate the modulation parameters to find the optimal solution.
Step S111: and initializing the modulation parameters of the switching device in the current-mode aging experimental system.
A population of the genetic algorithm is initialized, and in this embodiment, the population is modulation parameters such as a switching frequency, a duty ratio (modulation depth), an aging current, and the like. And (3) randomly selecting the initialization values of parameters such as switching frequency, duty ratio, aging current and the like, and ensuring that the initialization values are within a reasonable range. After initialization is completed, fitness is calculated. In this embodiment, the fitness is the total loss of the current switching device in this embodiment. And measuring the fitness of each individual in the genetic operation process by using a fitness function, namely judging the difference value of the current total loss and a reference value to obtain a criterion of excellence and disadvantage.
Step S112: and updating modulation parameters in real time according to the junction temperature change of the switching device, and calculating the total loss of the switching device in the current type aging experiment system.
Step S113: and judging whether the error between the current total loss and the preset total loss is within a preset error range.
And updating the value of the modulation parameter in real time according to the change of the junction temperature of the switching device in the current type aging experiment, and calculating the total loss of the current switching device. And judging whether the current total loss is the same as the total loss in the voltage type aging mode or within a certain error range. In this embodiment, the total loss condition can be determined by determining whether the heating condition of the switching tube is the same as or within a certain error range from the heat loss generated by the voltage-type aging method, so as to screen a spatial solution of the optimal point.
Step S114: and if the error between the total loss and the preset total loss is within the preset error range, defining the current modulation parameter as the optimal solution and outputting the optimal solution.
If the total loss of the switching device in the current type aging experiment is the same as the total loss in the voltage type aging experiment or the difference value is within a certain error range, the current modulation parameter is judged to be the optimal solution, and the values of the current modulation parameters such as the switching frequency, the duty ratio and the aging current are output as the optimal solution. When the current type aging experiment is carried out on the switching device, the modulation parameter of the switching device adopts the optimal solution to carry out value taking, and the loss of the switching device in the current type aging experiment and the loss of the switching device in the voltage type aging experiment can be equivalent.
Step S115: otherwise, the genetic operation of selection and variation is carried out on the modulation parameters, and the loss is calculated again until the error between the new total loss and the preset total loss is within the preset error range.
If the total loss of the switching device in the current-mode aging experiment is different from the total loss in the voltage-mode aging experiment, and the difference value is not within a certain error range, genetic operations such as selection and variation of a genetic algorithm are utilized to carry out iteration to generate a new population. In this embodiment, the new population refers to updating the switching frequency, the duty ratio and the aging current. And (5) circulating the processes until the optimal solution judgment condition is met, and ending and outputting the optimal solution.
The selection process of genetic algorithms typically employs a relatively fair roulette strategy, with the idea that the probability of each individual being selected is proportional to the fitness level. However, the selection using roulette strategy may result in the discarding of individuals of the previous generation that have a very good fitness. Thus, the roulette strategy plus the selection method to retain the best individual is employed in this embodiment. The best individuals in the parent population are reserved, and then N-1 individuals are selected from the parent and the offspring according to a roulette strategy. According to population generation, if a new number is not generated, and only the arrangement sequence of the numbers in the number series is changed, the feasibility of the number series solution is not changed, so that only mutation operation can be performed, and the complexity of the solution can be greatly simplified.
In one embodiment, the preset total loss is obtained according to an aging experiment result of the voltage type aging experiment system. In this embodiment, the loss equivalence algorithm in the aging test method for the switching device takes the minimum value of the total loss difference of the switching device in the current type aging test method and the voltage type aging test method as a target function, and judges whether the total loss of each switching device in the current type aging test system is the same as the total loss in the voltage type aging test method or within a certain error range, so as to judge whether the loss equivalence of the two different test methods can be realized.
Fig. 7 is a flowchart illustrating a method for performing iterative optimization by using a particle swarm optimization according to an embodiment of the present invention, where in one embodiment, performing iterative optimization on a modulation parameter of a switching device by using the particle swarm optimization on a modulation wave period scale of the switching device includes the following steps S121 to S125.
The core idea of the particle swarm algorithm is that the particles track the individual optimal positions as well as the global optimal positions in each iteration. The individual optimal position refers to an optimal value found after a certain solution passes through n iterations, and the global optimal position refers to an optimal value found after all solutions pass through n iterations. And continuously updating the speed and the position to generate new particles until the operation is stopped after the algorithm termination condition is met.
Step S121: and solving the obtained solution set according to a genetic algorithm to initialize the particle swarm.
Wherein the population of particles can be initialized from a solution set solved by a genetic algorithm.
Step S122: and calculating the fitness of the particle swarm.
The particle swarm optimization is applied to the setting of modulation parameters such as aging current, modulation depth, switching frequency, power frequency cycle switching time and conduction time, and the parameter setting process is essentially a parameter optimization process based on a specific objective function. In this embodiment, in order to suppress the error, the absolute value of the difference between the total loss in the voltage-type aging method and the total loss in the modulation wave period in the current-type aging method is integrated as the target function, that is, the fitness function is:
Figure BDA0003106707170000161
in the formula, J is the fitness, and e (t) is the absolute value of the difference value of the total loss in the modulation wave period under the voltage type aging method and the total loss under the current type aging method. Here, the loss in the next modulation wave period in the current-mode aging method may be calculated according to steps S221 to S222. Considering that infinite running time does not occur in the actual process, the upper integration limit can be made to coincide with the modulation wave period when actually performing calculation.
Step S123: and judging whether the fitness is in the error range.
Step S124: and if the fitness is within the error range, defining the modulation parameters in the current particle swarm as the optimal solution and outputting the optimal solution.
The fitness of the particle group is calculated according to step S122, and it is determined whether the fitness of the particle group is within the error range. If the fitness is in the error range, defining the modulation parameters in the current particle swarm as an optimal solution and outputting the optimal solution, and outputting the values of the modulation parameters such as the current switching frequency, the duty ratio, the aging current and the like as the optimal solution. When the current type aging experiment is carried out on the switching device, the modulation parameter of the switching device adopts the optimal solution to carry out value taking, and the loss of the switching device in the current type aging experiment and the loss of the switching device in the voltage type aging experiment can be equivalent.
Step S125: otherwise, updating the speed and the position of the particle swarm to generate a new particle swarm, and calculating the fitness of the new particle swarm again until the new fitness is within the error range.
During each iteration, the particles track the individual optimal positions as well as the global optimal positions. And continuously updating the speed and the position to generate new particles, and calculating the fitness of the new particles until the fitness of the new particles meets the algorithm termination condition and then stopping running.
The calculation of the update speed and position is as follows:
Figure BDA0003106707170000171
Figure BDA0003106707170000172
in the formula, i is the order of the particles, and the first solution is when i is 1, and so on. d is the dimension of the particle vector, and the total loss in the modulation wave period is mainly related to the aging current, the modulation depth, the switching frequency, the switching time of the power frequency period and the conduction time according to the calculation formula of the IGBT on-state loss and the switching loss in the modulation wave period, so that the total dimension d of the vector is 5.
Figure BDA0003106707170000173
Respectively an initial speed and an updating speed of a d-dimensional vector in the particle i in the k iteration; w is an inertia factor; c. C1Is an individual accelerating factor, c2Is a global acceleration factor; rand1，rand2Is the interval [0,1]A random number within;
Figure BDA0003106707170000174
for the individual best position of the d-th vector in particle i in the kth iteration,
Figure BDA0003106707170000175
the global optimal position of the d-dimension vector in the particle i in the k-th iteration is obtained;
Figure BDA0003106707170000176
for the initial position of the d-th vector in particle i in the k-th iteration,
Figure BDA0003106707170000177
is the updated position of the d-th vector in the particle i in the k-th iteration.
The learning factor gives different accelerations to the individual items and the global item, so that the individual items and the global item are searched along with the individual best position and the global best position, c2>c1Taking an empirical value c1＝1.2，c21.8. The inertia factor w can ensure that the particles move in the original speed direction, and when the value of w is large, the global searching capability of the particle swarm algorithm is strong; whilewhen the value of w is small, the local searching capability of the particle swarm algorithm is strong. w is typically [0.7, 1]]A certain value of (a).
In this embodiment, in order to avoid the possibility that the algorithm is poor in convergence and falls into a local minimum value, an inertial weight reduction method may be adopted, so that the algorithm is kept at a larger value in the initial stage, global search is performed, the search range is expanded, local search is performed in the later stage, and the convergence speed is increased. The inertial weighting reduction method is calculated as follows:
Figure BDA0003106707170000181
wherein w (i) is an inertia factor at the ith iteration; g is the total iteration number; w is amax、wminMaximum and minimum values of the inertia factor, wmax、wminCan be respectively 0.15 and 0.05. i is the order of the particles, i is the first solution when i equals 1, and so on. The total number of iterations G is a constant given before the algorithm starts.
Fig. 8 is a flowchart illustrating a method for iterative optimization by using a simulated annealing algorithm according to an embodiment of the present invention, where in an embodiment, the iterative optimization of the modulation parameter of the switching device by using the simulated annealing algorithm on the scale of the operation cycle of the switching device includes the following steps S131 to S138.
The simulated annealing algorithm can obtain a more accurate optimized solution than the two previous algorithms, but the optimization algorithm has a slower operation speed. The train working period is composed of a plurality of modulation wave periods, so that the range of the optimal solution is further expanded, and the range of the parameter optimal solution is partially reduced through the previous two screenings, so that the running time can be prevented from being overlong to a certain extent. And optimizing in the overall element space to be selected by using a single-target optimization algorithm to determine an objective function, namely the difference value of the total loss of the two aging methods.
And carrying out single parameter variable optimization on modulation parameters such as switching frequency, duty ratio, aging current, modulation depth and the like. Namely, a certain parameter variable is controlled to be fixed and then fine-tuned to other parameter variables in an equal amplitude manner. Wherein, the fixed value of the parameter with invariable numerical value and the fine tuning range of other variables can be selected according to the result of the particle swarm algorithm.
Because the setting of the iteration times is related to the requirement of precision and the running time of the algorithm, the iteration times are reasonably set according to the precision requirement after multiple attempts. And (3) controlling each variable to obtain the number of the variables to be solved according with the precision under the same iteration number, and carrying out weight division on each variable, wherein the weight coefficient is a certain value in (0, 1). After each weight is obtained, the annealing algorithm is reused, all parameters are changed simultaneously, and a final solution is obtained.
Step S131: and initializing the modulation parameters and the temperature, and calculating the total loss difference value of the current mode and the voltage mode in one train period.
And setting a reasonable feasible range for each modulation parameter such as switching frequency, duty ratio, aging current, modulation depth and the like to form a candidate space of the overall elements. For example, the switching frequency is generally set to 1000 to 2000Hz, the duty ratio epsilon (0,1), the aging current and the modulation depth need to be determined according to actual conditions. Setting initial temperature and modulation parameters such as initial switching frequency, duty ratio, aging current, modulation depth and the like, and calculating a target function value at the moment, namely the loss difference f (X) under the two aging methods in the train periodi). Since the train period is constituted by the modulated wave period, f (X)i) And J is a difference value, namely an adaptive value, in the modulation wave period and can be obtained by calculation according to the fitness function.
Step S132: and generating a new parameter value combination, calculating a new total loss difference value and the difference between the new total loss difference value and the previous total loss difference value, and acquiring a judgment value.
In the present embodiment, the perturbation value Δ x of each modulation parameter is randomly generatediRedistributing the parameter disturbance according to the weight of each parameter to obtain a new disturbance value delta xi', thereby generating a new parameter value combination Xi'＝Xi+ Δ X. The objective function value f (X) at this time is calculatedi') and new damageThe difference delta f between the loss difference and the previous loss difference is f (X)i')-f(Xi) And obtaining a judgment value delta f. The total loss is related to parameters such as aging current, modulation depth, switching frequency, power frequency period switching time and conduction time, each parameter such as aging current, modulation depth, switching frequency, power frequency period switching time and conduction time has a disturbance value, and the Delta X is a set of the disturbance values of the parameters.
Step S133: and comparing the judgment value with a preset value.
Step S134: and if the judgment value is smaller than or equal to the preset value, combining the new parameter values to serve as initialization data of the next simulation.
In the present embodiment, the preset value is 0. That is, if Δ f is less than or equal to 0, the newly obtained parameter combination value is used as the initial point of the next simulation.
Step S135: and if the judgment value is larger than the preset value, calculating the new point receiving probability.
If Δ f >0, calculating a new point acceptance probability, wherein the calculation formula of the new point acceptance probability is as follows:
Figure BDA0003106707170000201
in the formula, p (Δ f) is a new point acceptance probability. In the present embodiment, the above calculation formula is Metropolis criterion. The simulated annealing algorithm is based on the principle of simulated solid annealing, Δ E refers to the internal energy of the two states, and T refers to the current temperature. In the parameters corresponding to this embodiment, Δ E is the determination value Δ f, T is a parameter for adjusting the convergence speed of the algorithm, and the value of T can be determined only by debugging for many times.
Step S136: a random number is selected and the new point acceptance probability is compared to the random number.
A random number r is obtained, r being a pseudo random number subject to a uniform distribution over the interval [0,1], r being for [0,1 ].
Step S137: and if the new point receiving probability is larger than or equal to the random number, combining the new parameter values to serve as initialization data of the next simulation.
If p (Δ f) ≧ 0, the new combination of modulation parameters is taken as the initial point for the next simulation.
Step S138: otherwise, the new parameter value combination is abandoned, and the original parameter value combination is still used as the initialization data of the next simulation.
If p (delta f) <0, the new point is abandoned, and the original parameter combination value is still taken as the initial point of the next simulation. And (4) repeating the steps S131 to S137 until the IGBT losses are approximately equal in the voltage mode and the current mode or within a certain precision range, and stopping algorithm iteration.
When the current type space vector control strategy is adopted to carry out the IGBT aging experiment, different algorithms are respectively adopted according to the respective characteristics of the algorithms aiming at the aging periods of 3 layers, so that the parameter setting process is easy to converge, and is more efficient and accurate, thereby realizing power equivalence with the traditional voltage type aging method. The current type aging experiment can better reflect the aging process and the residual service life of the IGBT module under the actual load working condition. Compared with the traditional aging experiment, the method has lower requirements on power supply voltage and power and lower cost. The control strategy of the current type aging method is well determined on the basis of the power equivalent algorithm, and a foundation is laid for realizing the aging experiment.
It should be understood that although the various steps in the flowcharts of fig. 1-8 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-8 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps or stages.
In the description herein, references to the description of "some embodiments," "other embodiments," "desired embodiments," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, a schematic description of the above terminology may not necessarily refer to the same embodiment or example.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A method for equivalence aging of a switching device is characterized by comprising the following steps:
carrying out iterative optimization on the modulation parameters of the switching device by adopting different equivalent algorithms;
calculating the total loss of the switching device in the current mode aging experiment in each iteration;
and searching the optimal solution of the modulation parameter of the switching device by taking the minimum value of the total loss difference value of the switching device under the aging experiments of a current mode and a voltage mode as an objective function.
2. The equivalent method for aging of switching devices as claimed in claim 1, wherein said iteratively optimizing the modulation parameters of the switching devices using different equivalent algorithms comprises:
on the scale of the switching period of the switching device, iterative optimization is carried out on the modulation parameters of the switching device by adopting a genetic algorithm;
on the modulation wave period scale of the switching device, iterative optimization is carried out on the modulation parameters of the switching device by adopting a particle swarm algorithm;
and on the operation period scale of the switching device, carrying out iterative optimization on the modulation parameters of the switching device by adopting a simulated annealing algorithm.
3. The switching device aging equivalence method according to claim 2, wherein the calculating the total loss of the switching device in a current mode aging experiment in each iteration comprises:
calculating the total loss of the switching device in a switching period in the current-mode aging experiment;
and calculating the total loss of the switching device in one modulation wave period in the current-mode aging experiment.
4. The equivalent method for aging of the switching device as claimed in claim 1, wherein the modulation parameters include the switching frequency and duty cycle of the switching device and the aging current outputted by the aging current source.
5. The equivalent method for aging the switching device as claimed in claim 3, wherein said calculating the total loss of the switching device in the current-mode aging experiment during a switching period comprises:
fitting a static curve of the switching device to obtain a linear expression of collector-emitter voltage and collector current;
calculating the on-state loss of the switching device in one switching period;
acquiring the waveforms of the voltage and the current of the switching device in the switching-on and switching-off processes, and performing curve fitting in a segmented manner;
integrating the loss of each section of curve to obtain the switching loss in one switching period;
and acquiring the total loss of the switching device in one switching period according to the on-state loss and the switching loss.
6. The equivalent method for aging the switching device as claimed in claim 3, wherein said calculating the total loss of the switching device in one modulation wave period in the current-mode aging experiment comprises:
respectively calculating the on-state loss and the switching loss of the switching device in one modulation wave period;
and acquiring the total loss of the switching device in one modulation wave period according to the on-state loss and the switching loss.
7. The equivalent method for aging of a switching device according to claim 2, wherein said iteratively optimizing the modulation parameters of the switching device using a genetic algorithm on the scale of the switching period of the switching device comprises:
initializing modulation parameters of the switching device in the current-mode aging experimental system;
updating the modulation parameters in real time according to the junction temperature change of the switching device, and calculating the total loss of the switching device in the current type aging experiment system;
judging whether the error between the total loss and a preset total loss is within a preset error range or not;
if the error between the total loss and the preset total loss is within a preset error range, defining the current modulation parameter as an optimal solution and outputting the optimal solution;
otherwise, the modulation parameters are subjected to genetic operation of selection and variation, and the loss is calculated again until the error between the new total loss and the preset total loss is within the preset error range.
8. The equivalent method for aging of the switching device as claimed in claim 7, wherein the preset total loss is obtained according to an aging experiment result of a voltage type aging experiment system.
9. The equivalent aging method for the switching device as claimed in claim 8, wherein the iterative optimization of the modulation parameters of the switching device by using the particle swarm optimization on the modulation wave period scale of the switching device comprises:
solving the obtained solution set initialization particle swarm according to the genetic algorithm;
calculating the fitness of the particle swarm;
judging whether the fitness is within an error range;
if the fitness is within the error range, defining the modulation parameters in the current particle swarm as an optimal solution and outputting the optimal solution;
otherwise, updating the speed and the position of the particle swarm to generate a new particle swarm, and calculating the fitness of the new particle swarm again until the new fitness is within the error range.
10. The equivalent method for aging of the switching device according to claim 2, wherein the iterative optimization of the modulation parameters of the switching device by using the simulated annealing algorithm on the scale of the operation cycle of the switching device comprises:
initializing modulation parameters and temperature, and calculating the total loss difference value under two aging experiments of a current mode and a voltage mode in one train period;
generating a new parameter value combination, calculating a new total loss difference value and the difference between the new total loss difference value and the previous total loss difference value, and acquiring a judgment value;
comparing the judgment value with a preset value;
if the judgment value is smaller than or equal to the preset value, combining the new parameter values to serve as initialization data of next simulation;
if the judgment value is larger than the preset value, calculating the new point receiving probability;
selecting a random number and comparing the new point acceptance probability with the random number;
if the new point receiving probability is larger than or equal to the random number, combining the new parameter values to serve as initialization data of next simulation;
otherwise, abandoning the new parameter value combination and taking the original parameter value combination as the initialization data of the next simulation.
CN202110639520.7A 2021-06-08 2021-06-08 Switch device aging equivalent method Pending CN113449461A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110639520.7A CN113449461A (en) 2021-06-08 2021-06-08 Switch device aging equivalent method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110639520.7A CN113449461A (en) 2021-06-08 2021-06-08 Switch device aging equivalent method

Publications (1)

Publication Number Publication Date
CN113449461A true CN113449461A (en) 2021-09-28

Family

ID=77810937

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110639520.7A Pending CN113449461A (en) 2021-06-08 2021-06-08 Switch device aging equivalent method

Country Status (1)

Country Link
CN (1) CN113449461A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116562173A (en) * 2023-07-07 2023-08-08 南京邮电大学 Semiconductor device junction terminal geometric parameter design method based on simulated annealing algorithm

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040019208A (en) * 2002-08-27 2004-03-05 엘지.필립스 엘시디 주식회사 Aging Circuit For Organic Electroluminescence Device And Method Of Driving The same
CN101807900A (en) * 2010-03-10 2010-08-18 北京航空航天大学 Particle filter technology based on parallel genetic resampling
CN104620163A (en) * 2012-09-13 2015-05-13 浜松光子学株式会社 Optical modulation control method, control program, control device, and laser light irradiation device
CN109086604A (en) * 2018-07-05 2018-12-25 成都信息工程大学 Android malicious act software identification method and system based on sparse Bayesian model
CN109101738A (en) * 2018-08-24 2018-12-28 河北工业大学 A kind of IGBT module degree of aging appraisal procedure
CN110221189A (en) * 2019-06-05 2019-09-10 合肥工业大学 A kind of method of IGBT module bonding line on-line condition monitoring
CN112526333A (en) * 2020-11-05 2021-03-19 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） Current type aging test system and switch device aging test method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20040019208A (en) * 2002-08-27 2004-03-05 엘지.필립스 엘시디 주식회사 Aging Circuit For Organic Electroluminescence Device And Method Of Driving The same
CN101807900A (en) * 2010-03-10 2010-08-18 北京航空航天大学 Particle filter technology based on parallel genetic resampling
CN104620163A (en) * 2012-09-13 2015-05-13 浜松光子学株式会社 Optical modulation control method, control program, control device, and laser light irradiation device
CN109086604A (en) * 2018-07-05 2018-12-25 成都信息工程大学 Android malicious act software identification method and system based on sparse Bayesian model
CN109101738A (en) * 2018-08-24 2018-12-28 河北工业大学 A kind of IGBT module degree of aging appraisal procedure
CN110221189A (en) * 2019-06-05 2019-09-10 合肥工业大学 A kind of method of IGBT module bonding line on-line condition monitoring
CN112526333A (en) * 2020-11-05 2021-03-19 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） Current type aging test system and switch device aging test method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116562173A (en) * 2023-07-07 2023-08-08 南京邮电大学 Semiconductor device junction terminal geometric parameter design method based on simulated annealing algorithm
CN116562173B (en) * 2023-07-07 2023-09-12 南京邮电大学 Semiconductor device junction terminal geometric parameter design method based on simulated annealing algorithm

Similar Documents

Publication Publication Date Title
CN109101738B (en) IGBT module aging degree evaluation method
US9071150B2 (en) Variable frequency iteration MPPT for resonant power converters
EP1635438A2 (en) Method of controlling photovoltaic power generation system
US11855522B2 (en) Method and device for controlling an electronic converter in a technical system using methods of artificial intelligence
CN110213784B (en) Flow prediction method and device
CN113449461A (en) Switch device aging equivalent method
CN103593287A (en) Genetic-algorithm-based method for automatically generating data stream test cases
CN112152477B (en) Improved flying capacitor MMC topology and modulation strategy thereof
Visairo et al. Use of evolutionary algorithms for design optimization of power converters
CN108594927A (en) A kind of local shades photovoltaic array MPPT controls based on the full search algorithm that leapfrogs at random
CN114204547A (en) Power distribution network multi-measure combination loss reduction optimization method considering source network load storage cooperative optimization
Chabni et al. ANN-based SHEPWM using a harmony search on a new multilevel inverter topology
CN116258165A (en) Multi-target neural architecture searching method integrating convolution and self-attention
US20220368224A1 (en) Device and method for controlling converter
Jegajothi et al. An efficient metaheuristic optimization based fuzzy controller for brushless DC drives lifetime expansion
JP3785029B2 (en) Control apparatus and control method
CN111682792B (en) Converter model prediction control method based on multi-step prediction
CN106685290A (en) Method for self-turning parameters of rotary speed controller of permanent magnet synchronous motor based on chaos molecule kinetic theory algorithm
CN114640293B (en) Linear induction motor control method and system driven by three-level inverter
CN112836894B (en) Converter optimization method, system, computer equipment and storage medium
CN113746109B (en) Sequential model predictive control method for reducing switching loss of grid-connected inverter
CN109217357A (en) A kind of grid-connected photovoltaic system MPPT method based on Markov model
CN106980733A (en) A kind of improved four-dimensional Thinned arrays side lobe suppression method
Sandell et al. Multiobjective optimisation of active gate drivers for fast-switching MOSFETs
CN108983863B (en) A kind of photovoltaic maximum power tracking method based on improvement glowworm swarm algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20210928

RJ01 Rejection of invention patent application after publication