An IGBT aging cycle test method and system for a typical mission profile

Abstract

An IGBT aging cycle test method and system for a typical mission profile provided by the present invention include the following steps: obtaining the theoretical energy distribution of a target new energy power station on a long time scale, where the theoretical energy distribution includes a long-time wind speed distribution statistical curve and a wind turbine power curve; calculating to obtain the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT; using the obtained curves as the typical mission profile of the IGBT operation, and carrying out aging cycle tests according to different conditions; the present invention solves the above-mentioned deficiencies existing in the prior art and provides an experimental method basis for subsequent research on the change law of thermal parameters and IGBT life research.

Description

Technical Field

[0001] The present invention belongs to the field of new energy (wind power generation, photovoltaic power generation), and particularly relates to an IGBT aging cycle test method and system for a typical mission profile. Background Art

[0002] With the rapid development of new energy power generation industries represented by wind power generation and photovoltaic power generation in China, the life management of IGBTs, which are the core devices for energy conversion, has become a hot topic. The prerequisite for realizing life management is to deeply explore the operating characteristics of IGBTs through aging tests. However, in actual working conditions, IGBT modules operate in a random high-frequency switching state, the working current is unstable, and there are uncertain switching losses. In traditional DC power cycle experiments, IGBT modules only have on-state losses. Although PWM power cycle realizes high-frequency on-off, there is still a gap between the experimental conditions and guiding actual industrial control. There is an urgent need for an aging theory experimental method that can fit the actual working conditions. Summary of the Invention

[0003] The purpose of the present invention is to provide an IGBT aging cycle test method for a typical mission profile, which solves the defects of unstable working current and uncertain switching losses existing in the existing IGBT aging cycle test methods.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0005] An IGBT aging cycle test method for a typical mission profile provided by the present invention includes the following steps:

[0006] Step 1, obtaining the theoretical energy distribution at a long time scale of a target new energy power station, where the theoretical energy distribution includes a long-time wind speed distribution statistical curve and a wind turbine power curve;

[0007] Step 2, extracting a single-day wind speed change curve at a long time scale by using the long-time wind speed distribution statistical curve;

[0008] Step 3, calculating the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT by using the wind turbine power curve obtained in Step 1 and combining with the single-day wind speed change curve at a long time scale;

[0009] Step 4, respectively smoothing the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT obtained in Step 3 to obtain the corresponding smoothed curves;

[0010] Step 5: Use the obtained corresponding smoothed curve as the typical task profile for the IGBT operation, and conduct aging cycle tests under different conditions.

[0011] Preferably, in Step 3, calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT. The specific method is as follows:

[0012] Calculate the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve based on the single-day wind speed change curve on a long time scale;

[0013] Using the wind turbine power curve obtained in Step 1, and combining with the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve obtained in Step 3, calculate the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve;

[0014] Based on the monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve, calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT.

[0015] Preferably, the method for calculating the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve based on the single-day wind speed change curve on a long time scale is as follows:

[0016] Using the obtained single-day wind speed change curve on a long time scale, conduct probability density function analysis on the long-term wind measurement data, and calculate the most probable wind speed distribution at different times within a single day;

[0017] Respectively fit the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve based on the obtained most probable wind speed distribution at different times within a single day.

[0018] Preferably, use the cubic spline interpolation method to calculate the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve.

[0019] Preferably, use the cubic spline interpolation method to calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT.

[0020] Preferably, in Step 4, obtain the corresponding smoothed curves respectively. The specific method is as follows:

[0021] Take the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT as the typical mission profiles of the IGBT, and smooth the three types of curves respectively based on the smoothing filtering method to obtain the corresponding smoothed curves.

[0022] An IGBT aging cycle test system for typical mission profiles, comprising:

[0023] An acquisition module for acquiring the theoretical energy distribution of a target new energy power station on a long time scale, where the theoretical energy distribution includes a long-term wind speed distribution statistical curve and a wind turbine power curve;

[0024] An extraction module for extracting the single-day wind speed change curve on a long time scale by using the long-term wind speed distribution statistical curve;

[0025] A working current change curve calculation module for calculating the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT by using the wind turbine power curve obtained in combination with the single-day wind speed change curve on a long time scale

[0026] A smoothing processing module for smoothing the obtained monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT respectively to obtain the corresponding smoothed curves;

[0027] A test module for taking the obtained corresponding smoothed curves as the typical mission profiles of the IGBT work and conducting aging cycle tests according to different conditions.

[0028] Preferably, the working current change curve calculation module includes:

[0029] A wind speed change curve calculation unit for calculating the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve according to the single-day wind speed change curve on a long time scale;

[0030] A power change curve calculation unit for calculating the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve by using the obtained wind turbine power curve in combination with the obtained monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve;

[0031] The working current variation curve calculation unit is used to calculate the monthly average daily working current variation curve, quarterly average daily working current variation curve, and annual average daily working current variation curve of the IGBT based on the obtained monthly average daily power variation curve, quarterly average daily power variation curve, and annual average daily power variation curve.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] Aiming at the problems that the IGBT module operates in a random high-frequency switching state, the working current is unstable, and there are uncertain switching losses, according to the energy distribution in different seasons of the target new energy power station, a conditional available resource screening method based on reference points and a construction method for the typical mission profile (Typical Mission Profile) of the new energy power station are proposed; the correlation between the resource distribution and the IGBT working current under the typical profile is established; a PWM carrier modulation method for the module working current sequence is proposed, and a cyclic aging test method under the condition of the IGBT working current sequence corresponding to the typical mission profile is obtained, solving the above-mentioned deficiencies in the prior art and providing an experimental method basis for the subsequent research on the variation law of thermal parameters and the IGBT life research. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic flow chart of the present invention;

[0035] Figure 2 is the wind speed and wind energy histogram and wind speed distribution map of a certain wind farm;

[0036] Figure 3 is the power curve of a certain wind power generator;

[0037] Figure 4 is the available sunlight for photovoltaic power generation in a certain place;

[0038] Figure 5 is the comparison curve of photovoltaic power generation power and solar radiation intensity;

[0039] Figure 6 is the wind speed variation curve within a single day. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following further details the present invention in conjunction with the drawings.

[0041] An IGBT aging cycle test method for a typical mission profile provided by the present invention, the specific process of the embodiment is as shown in the appendix Figure 1 and the specific steps are as follows:

[0042] Step 1, collect the theoretical energy distribution of the target new energy power station on a long time scale. The theoretical energy distribution of wind power generation includes the available wind speed ( Figure 2)), Photovoltaic power generation utilizes available sunlight ( Figure 3 )), Wind turbine power curve ( Figure 4 )), Photovoltaic power generation power and solar radiation intensity comparison curve ( Figure 5 ).

[0043] Step 2, Taking a wind farm as an example, according to the available wind speed statistical results in Step 1 and the long-term wind speed distribution statistical curve, extract the daily wind speed change curve of the target wind farm within a long time scale (within 1 year) (for example, the wind speed on a certain day is as Figure 6 shown), and obtain the daily wind speed change curve on a long time scale. If taking a photovoltaic power station as an example, then obtain the daily sunlight change curve on a long time scale in the same way.

[0044] Step 3, Using the daily wind speed change curve within 1 year obtained in Step 2, conduct a probability density function (PDF) analysis on the long-term wind measurement data, calculate the most probable wind speed distribution at different times within a day, and respectively fit the monthly average daily wind speed change curve WS m , seasonal average daily wind speed change curve WS s and annual average daily wind speed change curve WS a . Similarly, if taking a photovoltaic power station as an example, then respectively fit the monthly average daily sunlight change curve SL m , seasonal average daily sunlight change curve SL s and annual average daily sunlight change curve SL a in the same way.

[0045] Step 4, According to the unit power curve collected in Step 1 and the monthly average daily wind speed change curve WS m , seasonal average daily wind speed change curve WS s , annual average daily wind speed change curve WS a , use the cubic spline interpolation method to calculate the corresponding monthly average daily power change curve P m , seasonal average daily power change curve P s , annual average daily power change curve P a . Similarly, if taking a photovoltaic power station as an example, also use the cubic spline interpolation method to calculate the corresponding monthly average daily power change curve P m , seasonal average daily power change curve P s , annual average daily power change curve P a .

[0046] Step 5, According to the monthly average daily power change curve P m calculated in Step 4 and the seasonal average daily power change curve Ps 1. The annual average daily power variation curve P a , combined with the working current-power curve of the wind turbine generator, uses the cubic spline interpolation method to calculate the monthly average daily working current variation curve I of the IGBT m , the quarterly average daily working current variation curve I s , the annual average daily working current variation curve I a . Similarly, taking a photovoltaic power station as an example, combined with the curve of the photovoltaic power generation power versus the solar radiation intensity, uses the cubic spline interpolation method to calculate the monthly average daily working current variation curve I of the IGBT m , the quarterly average daily working current variation curve I s , the annual average daily working current variation curve I a .

[0047] Step 6. Take the monthly average daily working current variation curve I of the IGBT obtained in Step 5 m , the quarterly average daily working current variation curve I s , the annual average daily working current variation curve I a as the typical mission profile of the IGBT, and smooth the three types of curves respectively based on the Savitzky-Golay smoothing filtering method

[0048] Taking the monthly average daily working current variation curve I m as an example, the specific method of smoothing is as follows

[0049] I m as a time series contains N points. To reduce the highest degree of the fitting polynomial, divide the time series I m into l subsequences, then there is

[0050] I m =[I m-1 , I m-2 ,..., I m-i ,...I m-l

[0051] wherein, I m-1 , I m-2 ,..., I m-i ,...I m-l sequence contains 2q + 1 points. Taking I m-i as an example, assume

[0052] I m-i = I(x);

[0053] x ∈ [-q, q]

[0054] Then there is

[0055] I​m-i = I(x) = [i -q , i -q+1 ,..., i0,..., i q-1 , i q

[0056] In the above formula, i -q , i -q+1 ,..., i0,..., i q-1 , i q are 2q + 1 data points of the subsequence I m-i .

[0057] Step 7, construct an nth-order polynomial f(x) to fit the subsequence I obtained in Step 6 m-i , and this polynomial can be expressed as:

[0058]

[0059] where f(x) is the finally fitted nth-order polynomial, and a k is the kth-order leading coefficient.

[0060] Step 8, calculate the least squares fitting residual E(q) between the polynomial f(x) and the original sequence I(x), and the specific calculation method is as follows:

[0061]

[0062] Step 9, to minimize the least squares residual E(q) obtained in Step 8, let the partial derivatives of the residual E(q) with respect to each leading coefficient be 0, that is:

[0063]

[0064] Simplify the above formula to obtain:

[0065]

[0066] Step 10, for the equation obtained in Step 9, introduce the vector A and the square matrix B, and the calculation methods of the vector A and the square matrix B are as follows:

[0067] A = {x r}; r = 0, 1, 2,..., n - 1, n

[0068] B = A T × A

[0069] Then, according to the above formula, it can be calculated that:

[0070]

[0071] ​Step 11, based on the vector A and the square matrix B obtained in Step 10, further calculate:

[0072] Ba = (A T × A)a = A T x;

[0073] a = (A T × A) -1 · A T x = Hx;

[0074] Wherein,

[0075] a = [a0, a1, a2,..., a n

[0076] x = [1, x, x 2 ,..., x r ,..., x n

[0077] Then H is the convolution coefficient sought.

[0078] Step 12, repeat Steps 7 to 11 to calculate the polynomial fitting result f m of the time series I m to complete curve smoothing.

[0079] Step 13, use the same method to complete the curve smoothing of the time series I s and the time series I a to obtain the polynomial fitting results f s and f a .

[0080] Step 14, use the smoothed curves f m , f s and f a obtained in Step 13 as the typical mission profiles of IGBT operation, and conduct aging cycle tests according to different conditions. Specifically:

[0081] Condition 1, using the traditional method, apply a PWM signal with an amplitude of 10V, a frequency of 10kHz, and a duty cycle of 40% to the gate. Make the IGBT device continuously turn on and off until failure, and collect parameters such as the saturation voltage drop V cesat , current, module junction temperature, module case temperature, and ambient temperature during the test.

[0082] Condition 2, select the smoothed IGBT operating current time series f m , f s and f a corresponding to the monthly, quarterly, and annual average wind speed (illumination) change curves.As the input, a high-frequency stepped square wave is used for accelerated aging. The period T is set to 60 s. The IGBT device is continuously turned on and off until it fails, and parameters such as the saturation voltage drop V cesat , current, module junction temperature, module case temperature, and ambient temperature are collected. By conducting aging tests under different conditions, an experimental data basis is provided for the subsequent research on the variation law of thermal parameters and the research on equivalent calculation methods.

Claims

1. A method for IGBT aging cycle test oriented to a typical mission profile, characterized in that, It includes the following steps: Step 1: Obtain the theoretical energy distribution of the target new energy power station on a long time scale, where the theoretical energy distribution includes the long-term wind speed distribution statistical curve and the wind turbine power curve; Step 2: Extract the single-day wind speed change curve on a long time scale using the long-term wind speed distribution statistical curve; Step 3: Using the wind turbine power curve obtained in Step 1 and combining it with the single-day wind speed change curve on a long time scale, calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT; Step 4: Smooth the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT obtained in Step 3 respectively to obtain the corresponding smoothed curves; Step 5: Use the obtained corresponding smoothed curves as the typical task profiles of the IGBT work, and carry out aging cycle tests according to different conditions. Specifically: Condition 1: Adopt the traditional method, apply a PWM signal with an amplitude of 10V, a frequency of 10kHz, and a duty cycle of 40% to the gate to make the IGBT device continuously turn on and off until it fails, and collect the saturation voltage drop, current, module junction temperature, module case temperature, and ambient temperature during the test; Condition 2: Select the smoothed IGBT operating current time series corresponding to the monthly, quarterly, and annual average wind speed change curves and as inputs, and use a high-frequency stepped square wave for accelerated aging. The period T is set to 60 s, causing the IGBT device to continuously turn on and off until failure. During the test, collect the saturation voltage drop, current, module junction temperature, module case temperature, and ambient temperature In Step 3, calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT. The specific method is: Calculate the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve according to the single-day wind speed change curve on a long time scale; Using the wind turbine power curve obtained in Step 1 and combining it with the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve obtained in Step 3, calculate the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve; Calculate the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT according to the monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve; 2. The IGBT aging cycle test method for a typical mission profile according to claim 1, wherein Calculate the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve according to the single-day wind speed change curve on a long time scale. The specific method is: Use the obtained single-day wind speed change curve on a long time scale to perform probability density function analysis on the long-term wind measurement data, and calculate the most probable wind speed distribution at different times within a single day; Respectively fit the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve according to the obtained most probable wind speed distribution at different times within a single day; 3. The IGBT aging cycle test method for a typical mission profile according to claim 1, characterized in that, Use the cubic spline interpolation method to calculate the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve.

4. A method for IGBT aging cycle test oriented to a typical mission profile according to claim 1, characterized in that The monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT are calculated using the cubic spline interpolation method.

5. The IGBT aging cycle test method for a typical mission profile according to claim 1, characterized in that In step 4, the corresponding smoothed curves are obtained respectively. The specific method is as follows: Taking the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT as the typical mission profiles of the IGBT, the three types of curves are smoothed respectively based on the smoothing filter method to obtain the corresponding smoothed curves.

6. An IGBT aging cycle test system for a typical mission profile, characterized in that The method according to claim 1 includes: An acquisition module for acquiring the theoretical energy distribution of the target new energy power station on a long time scale, where the theoretical energy distribution includes the long-term wind speed distribution statistical curve and the wind turbine power curve; An extraction module for extracting the single-day wind speed change curve on a long time scale using the long-term wind speed distribution statistical curve; A working current change curve calculation module for calculating the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT by combining the wind turbine power curve obtained in with the single-day wind speed change curve on a long time scale; A smoothing processing module for smoothing the obtained monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT respectively to obtain the corresponding smoothed curves; A testing module for using the obtained corresponding smoothed curves as the typical mission profiles of the IGBT to carry out aging cycle tests under different conditions.

7. The IGBT aging cycle test system for a typical mission profile according to claim 6, characterized in that, The working current change curve calculation module includes: A wind speed change curve calculation unit for calculating the monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve according to the single-day wind speed change curve on a long time scale; A power change curve calculation unit for calculating the corresponding monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve by combining the obtained wind turbine power curve with the obtained monthly average daily wind speed change curve, quarterly average daily wind speed change curve, and annual average daily wind speed change curve; A working current change curve calculation unit for calculating the monthly average daily working current change curve, quarterly average daily working current change curve, and annual average daily working current change curve of the IGBT according to the obtained monthly average daily power change curve, quarterly average daily power change curve, and annual average daily power change curve.

Images