A thermal management high-voltage device variable-condition endurance test method

Abstract

The present application relates to the technical field of device endurance test, and particularly relates to a thermal management high-voltage device variable working condition endurance test method. Power thermal response delay formed by electric load change and cooling side temperature response delay formed by cooling medium temperature state change are respectively identified, and switching time of variable working condition cycle is determined in combination with target and interval of two kinds of temperature responses; cooling medium temperature state switching time of subsequent cycle is corrected according to deviation of actual and interval of each cycle and target interval. Same diagnostic thermal excitation is applied before endurance test and at predetermined cycle node, and endurance state is determined according to index change of current thermal response relative to initial thermal response.

Description

Technical Field

[0001] This invention relates to the field of device durability testing technology, and in particular to a method for testing the durability of thermal management high-voltage devices under varying operating conditions. Background Technology

[0002] During operation, high-voltage devices with liquid-cooled heat exchange structures experience changes in their internal heat loss due to variations in current, voltage, and operating status. Simultaneously, changes in the cooling medium temperature and thermal management status alter the external thermal boundary of the liquid-cooled heat exchange interface. The transfer of heat generated inside the device to the liquid-cooled heat exchange interface via the encapsulation structure takes time, and the temperature change of the cooling medium, through fluid transport, the heat capacity of the heat exchange components, and interfacial heat transfer, also requires a response time to reach the device. These two response processes are not identical.

[0003] Existing endurance tests typically change the electrical load and circulating medium state according to a pre-set time sequence. Even if there is a definite time relationship between the control commands, the actual temperature response on both sides of the liquid-cooled heat exchange interface may still show different advances or lags. As the endurance cycle progresses and the heat transfer characteristics of the device change, the fixed control sequence determined at the beginning of the test may gradually deviate from the original test state. Summary of the Invention

[0004] The present invention aims to solve the above-mentioned technical problems by providing a method for durability testing of thermal management high-voltage devices under varying operating conditions.

[0005] The technical solution of this invention is a method for endurance testing of thermal management high-voltage devices under varying operating conditions, comprising: The high-voltage device under test with a liquid-cooled heat exchange interface is connected to a durability test bench. The durability test bench includes an electrical load circuit and a thermal boundary circuit that can switch between a first temperature state and a second temperature state while keeping the mass flow rate and inlet pressure of the cooling medium within their respective preset ranges. The first temperature data and the second temperature data are obtained by temperature detection points set on the device side and the cooling side of the liquid-cooled heat exchange interface, respectively. While maintaining a stable thermal boundary state, the electrical load is changed, and the power thermal response delay is determined based on the characteristic moment of the change in the first temperature data relative to the electrical load. While maintaining a stable electrical load, the cooling medium temperature state is switched, and the cooling side temperature response delay is determined based on the characteristic moment of the change in the second temperature data relative to the cooling medium inlet temperature. Based on the power thermal response delay, the cooling side temperature response delay, and the target sequence interval between the first temperature response and the second temperature response, the electrical load switching time and the cooling medium temperature state switching time in the variable operating condition cycle are determined. The variable operating condition cycle is executed, and the timing of the cooling medium temperature state switching in subsequent cycles is corrected based on the deviation between the actual and target intervals between the first and second temperature responses in each cycle. Before the endurance test and at predetermined cycle points, the same diagnostic thermal excitation is applied under the same baseline operating conditions to obtain the initial thermal response and the current thermal response. The endurance status is determined based on the change in the preset thermal response index of the two. When the change reaches the preset failure limit, the high voltage device under test is determined to have failed and the corresponding cycle number is recorded. When the target number of cycles is completed and the change does not reach the preset failure limit, the endurance test is determined to have passed.

[0006] In one implementation, two adjacent variable operating condition cycles are respectively designated as the first stress cycle and the second stress cycle. The switching time of the cooling medium temperature state is adjusted according to the sign of the temperature difference between the first temperature data and the second temperature data, so that the temperature difference remains positive during the preset observation period of the first stress cycle and negative during the preset observation period of the second stress cycle, and the extreme values ​​of the device-side temperature in the two stress cycles are between the same preset upper and lower temperature limits.

[0007] In one embodiment, the thermal boundary loop includes a first temperature-regulating branch and a second temperature-regulating branch that respectively maintain the cooling medium at a first temperature state and a second temperature state. The two temperature-regulating branches are selectively connected to the high-voltage device under test at the cooling medium inlet of the device under test through a switching component, and the mass flow rate and inlet pressure of the cooling medium of the two temperature-regulating branches are adjusted respectively so that the changes in mass flow rate and inlet pressure before and after the branch switching are within their respective preset ranges.

[0008] As one implementation method, the characteristic moment is determined based on a preset proportion of the total response amplitude from the steady-state value before the change to the steady-state value after the change in the corresponding temperature response.

[0009] As one implementation, the device-side temperature is monitored during the execution of the variable operating condition cycle. When the device-side temperature enters a warning range adjacent to the preset upper or lower temperature limit, the duration or amplitude of the electrical load in the subsequent cycle is adjusted, and the switching time of the cooling medium temperature state is re-determined based on the adjusted electrical load, so that the extreme value of the device-side temperature is maintained between the preset upper and lower temperature limits, and the first temperature response and the second temperature response maintain the target sequential interval.

[0010] As one implementation, the diagnostic thermal excitation includes applying a preset electrical load pulse to the high-voltage device under test under the condition that the same cooling medium inlet temperature, mass flow rate and inlet pressure are used in each diagnosis, and continuously collecting temperature data on the device side before, during and after the application of the electrical load pulse and during the temperature recovery process after the pulse ends, to obtain the transient thermal response.

[0011] As one implementation method, the change in the preset thermal response index includes at least one of the following: the difference between the initial thermal response and the current thermal response in time to reach the preset response ratio, the temperature rise difference at a specified time, the slope difference in the heating stage, and the characteristic time difference in the cooling stage.

[0012] As one implementation method, based on the actual operating data of the high-voltage device under test, the typical sequence and time interval of changes in electrical load state and thermal management state are statistically analyzed, and the target sequence interval is determined accordingly. Multiple variable operating condition cycles with different target sequence intervals are arranged into a durability test sequence according to the typical sequence.

[0013] As one implementation method, when the mass flow rate, inlet pressure, or inlet temperature of the cooling medium is detected to exceed their respective preset ranges, the current cycle is not counted in the effective cycle count, so that the electrical load and thermal boundary return to the preset reference state. After the mass flow rate, inlet pressure, and inlet temperature of the cooling medium stabilize within their respective preset ranges, the variable operating condition cycle is restarted.

[0014] As one implementation method, the difference between the first temperature data and the second temperature data is changed from a positive value to a negative value or from a negative value to a positive value, and the corresponding sign is continuously maintained for a preset confirmation time after the conversion, which serves as a confirmation condition for the conversion of the thermal action direction. When the thermal action direction conversion confirmation condition is met, it is confirmed that the thermal action direction of the liquid-cooled heat exchange interface has been converted. If the condition is not met, the switching time of the cooling medium temperature state or the set temperature of the first temperature state or the second temperature state is adjusted until the thermal action direction conversion confirmation condition is met.

[0015] The advantages of this invention compared to existing technologies are that it obtains the response delays corresponding to the electrical load and liquid cooling thermal boundary effects under relatively independent conditions, and then determines the switching time of the electrical load and cooling medium temperature states based on the two response delays. This ensures that the time relationship defined by the variable operating condition cycle corresponds to the sequential relationship of the actual temperature responses on both sides of the liquid cooling heat exchange interface. During the durability process, the deviation between the actual sequential interval and the target sequential interval is used to correct the subsequent switching time, allowing the test sequence to be corrected according to changes in the actual thermal response. Simultaneously, the thermal response is compared and diagnosed under the same baseline operating conditions, forming a durability state determination result corresponding to the durability cycle. Attached Figure Description

[0016] Figure 1 A schematic diagram of the durability testing bench system provided in an embodiment of the present invention; Figure 2 A schematic diagram illustrating the power thermal response delay and cooling-side temperature response delay identification process provided for embodiments of the present invention; Figure 3A schematic diagram showing the timing of electrical load switching, cooling medium temperature state switching, and temperature response intervals in a variable operating condition cycle provided for embodiments of the present invention. Figure 4 A flowchart for correcting the switching time of the temperature state of the subsequent circulating cooling medium according to the actual sequential interval, provided for an embodiment of the present invention. Figure 5 A flowchart for obtaining diagnostic thermal response and determining durability status provided for embodiments of the present invention. Detailed Implementation

[0017] The above and other embodiments and advantages of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0018] like Figures 1 to 4 As shown, the high-voltage thermal management device described in this embodiment is a high-voltage device or high-voltage device assembly with a liquid-cooled heat exchange interface that generates electrothermal losses during operation. It can be applied to scenarios such as new energy vehicles and high-voltage power conversion equipment that use liquid-cooled high-voltage power devices. The high-voltage device under test can be a liquid-cooled high-voltage power unit in a power module, inverter, on-board charger, or DC-DC converter.

[0019] During durability testing, the high-voltage device under test (VDT) is connected to the durability test bench. The durability test bench forms at least an electrical load path and a liquid-cooled thermal boundary path. The electrical load path applies an electrical load to the VDT, altering the VDT's own heat loss by changing the electrical load. The thermal boundary path provides cooling medium to the liquid-cooled heat exchange interface of the VDT, changing the external thermal boundary of the liquid-cooled heat exchange interface by changing the temperature state of the cooling medium.

[0020] The thermal boundary loop can switch between a first temperature state and a second temperature state. During temperature state switching, the mass flow rate and inlet pressure of the cooling medium are maintained within their respective preset ranges, ensuring that changes in mass flow rate and pressure exceeding the test conditions are not introduced simultaneously during the temperature state change. The first temperature state and the second temperature state represent the two different cooling medium temperature conditions used in this durability test, and their specific temperatures are determined according to the durability test conditions corresponding to the high-voltage device under test.

[0021] Temperature detection points are set on the device side of the liquid-cooled heat exchange interface to obtain the first temperature data; temperature detection points are set on the cooling side of the liquid-cooled heat exchange interface to obtain the second temperature data. The first temperature data reflects the temperature change on the side of the liquid-cooled heat exchange interface closer to the high-voltage device under test, and the second temperature data reflects the temperature change on the side of the liquid-cooled heat exchange interface closer to the cooling medium. The changes in electrical load, the changes in the inlet temperature of the cooling medium, the first temperature data, and the second temperature data are recorded according to the same time reference so that each characteristic moment can be compared on the same time axis.

[0022] When the heat loss of the high-voltage device under test changes, the transfer of heat from the device's interior to the liquid-cooled heat exchange interface requires a corresponding heat transfer process. Similarly, when the inlet temperature of the cooling medium changes, the new liquid-cooled thermal boundary acting on the liquid-cooled heat exchange interface also requires a fluid transport and heat exchange process. Therefore, even if the changes in electrical load and the changes in the temperature state of the cooling medium have a definite temporal relationship at the control time, their corresponding first and second temperature responses do not necessarily have to occur simultaneously.

[0023] To obtain the response delays for the two thermal action paths, the electrical load is first changed while maintaining a stable thermal boundary state. During this process, the thermal boundary loop does not actively change the temperature state of the cooling medium, ensuring that the change in the first temperature data primarily corresponds to the temperature response resulting from the change in the heat generation state of the high-voltage device under test after the change in electrical load. The power thermal response delay is determined based on the characteristic moments of the first temperature data relative to the change in electrical load.

[0024] Subsequently, the cooling medium temperature state is switched while maintaining a stable electrical load. During this process, the electrical load of the high-voltage device under test is not actively changed, so that the change in the second temperature data primarily corresponds to the temperature response formed by the change in the liquid-cooled thermal boundary after the change in the cooling medium inlet temperature. The cooling-side temperature response delay is determined based on the characteristic moments of the second temperature data relative to the change in the cooling medium inlet temperature.

[0025] Therefore, the power thermal response delay and the cooling-side temperature response delay are obtained under the condition that the other thermal action path remains stable, avoiding the problem of not being able to distinguish the main source of the corresponding temperature response when both action paths change simultaneously. The power thermal response delay is used to characterize the time relationship between a change in electrical load and the appearance of the corresponding feature in the first temperature response; the cooling-side temperature response delay is used to characterize the time relationship between a change in the inlet temperature of the cooling medium and the appearance of the corresponding feature in the second temperature response.

[0026] In one specific implementation, such as Figure 2As shown, the characteristic moment of the temperature response is determined based on the total response amplitude from the steady-state value before the change to the steady-state value after the change. For the first temperature data, the steady-state value before the change is obtained before the change in electrical load, and the corresponding steady-state value after the change is obtained after the change in electrical load. The total response amplitude of the first temperature response is determined by these two values. When the first temperature response reaches a preset proportion of the total response amplitude along the corresponding change direction, the corresponding moment is taken as the characteristic moment of the first temperature response.

[0027] For the second temperature data, the steady-state value before the temperature change of the cooling medium is obtained, and the steady-state value after the corresponding change is obtained after the temperature change. The total response amplitude of the second temperature response is determined by both. When the second temperature response reaches a preset proportion of this total response amplitude, the corresponding time is taken as the characteristic time of the second temperature response. In this method, the characteristic time of each response used for comparison is determined according to the same response proportion, so that the obtained time relationship corresponds to the same temperature response position. The preset proportion is determined before the test starts according to the same response delay identification rule and remains consistent during the mutual comparison of responses.

[0028] After obtaining the power thermal response delay and the cooling-side temperature response delay, a target sequence interval between the first and second temperature responses is determined. The target sequence interval includes the order of the two temperature responses and the corresponding time interval, used to determine the desired actual thermal response timing at the liquid-cooled heat exchange interface during durability cycling.

[0029] Based on the power thermal response delay, the time relationship of the first temperature response's characteristic moment after the electrical load switches at the corresponding time can be determined; based on the cooling-side temperature response delay, the time relationship of the second temperature response's characteristic moment after the cooling medium temperature state switches at the corresponding time can be determined. Combining the target sequence interval between the first and second temperature responses, the timing of the electrical load switching and the timing of the cooling medium temperature state switching are determined.

[0030] Therefore, the endurance test does not directly take the time difference between the electrical load control command and the cooling medium temperature state control command as the actual test state. Instead, it uses the response delay of each of the two thermal action paths to determine the control time in reverse, so that after the two inputs have gone through their respective thermal response processes, the first temperature response and the second temperature response at the liquid cooling heat exchange interface are formed at the target interval.

[0031] After determining the initial switching time, the variable operating condition cycle begins execution. In each variable operating condition cycle, the actual characteristic time of the first temperature response is determined based on the first temperature data, the actual characteristic time of the second temperature response is determined based on the second temperature data, and the actual interval between the first and second temperature responses in this cycle is determined based on the two actual characteristic times.

[0032] The actual sequence interval obtained in the current cycle is compared with the target sequence interval. When there is a deviation, the deviation reflects whether the actual response timing of the liquid-cooled heat exchange interface occurs earlier or later than the predetermined response timing during the current test. Based on this deviation, the switching time of the cooling medium temperature state in subsequent cycles is corrected, so that the second temperature response changes relative to the actual occurrence time of the first temperature response to the time position corresponding to the target sequence interval.

[0033] For example, if the timing of the first temperature response is used as the current comparison basis, and the second temperature response occurs earlier than the target interval, the switching time of the cooling medium temperature state in subsequent cycles will be adjusted accordingly to shift the actual occurrence time of the second temperature response later. If the second temperature response lags behind the target interval, the switching time of the cooling medium temperature state will be adjusted accordingly to shift the actual occurrence time of the second temperature response earlier. The adjusted switching time is used in subsequent variable operating condition cycles, and further corrections will be made based on the actual intervals obtained in that cycle.

[0034] During durability cycling, the state of the liquid-cooled heat exchange interface, package connection, or other heat transfer paths of the high-voltage device under test may change with each cycle, causing its thermal response time to change relative to the initial stage of the test. Using the above-described cycle correction method, it is not required that all durability cycles consistently use the fixed cooling medium temperature state switching time determined at the beginning of the test. Instead, the actual temperature response timing formed in the current cycle is used to correct subsequent cycles.

[0035] In one implementation, such as Figure 3 As shown, two adjacent variable operating condition cycles are designated as the first stress cycle and the second stress cycle, respectively. The temperature difference between the first and second temperature data represents the temperature relationship on both sides of the liquid-cooled heat exchange interface. During the preset observation period of the first stress cycle, the switching time of the cooling medium temperature state is adjusted according to the sign of the temperature difference to keep the temperature difference between the first and second temperature data positive; during the preset observation period of the second stress cycle, the temperature difference is kept negative. The extreme values ​​of the device-side temperature in both the first and second stress cycles are limited to the same preset upper and lower temperature limits. Therefore, two adjacent stress cycles can form different temperature relationships on both sides of the liquid-cooled heat exchange interface without changing the range of extreme values ​​of the device-side temperature. Thus, the liquid-cooled heat exchange interface can alternately experience two opposite temperature gradient relationships under the same device-side temperature boundary, which can be used to simulate the different interfacial thermal states formed by the misalignment of the device's intrinsic thermal response and the liquid-cooled thermal boundary response during actual variable load processes.

[0036] Furthermore, during the thermal action direction conversion between the first and second stress cycles, the difference between the first and second temperature data is changed from positive to negative or vice versa, and the sign of the converted temperature difference continues to be monitored. When the temperature difference maintains the corresponding sign for a preset confirmation time after the conversion, the thermal action direction of the liquid-cooled heat exchange interface is confirmed to have been converted. If the converted temperature difference does not maintain the corresponding sign for a preset confirmation time, the thermal action direction conversion is not confirmed, and the timing of the cooling medium temperature state switching, or the set temperature of the first or second temperature state, is adjusted until the thermal action direction conversion confirmation condition is met. The preset confirmation time is predetermined according to test judgment conditions that can exclude instantaneous zero crossings or short-term fluctuations in the temperature difference, and the same confirmation conditions are used in the stress cycles being compared. Thus, the change in the thermal action direction between the first and second stress cycles is not determined solely by a single instantaneous temperature difference sign, but is jointly confirmed by the temperature difference sign conversion and its duration. The confirmation result further determines whether to continue adjusting the liquid-cooled thermal boundary. This confirmation method can avoid misjudgment of the direction of thermal action caused by instantaneous zero crossing or short-term fluctuations in temperature difference, so that subsequent cycles will only enter the next test state after the interface temperature relationship in the predetermined direction has actually been formed.

[0037] In one implementation of a thermal boundary loop, such as Figure 1 As shown, the thermal boundary loop includes a first temperature-regulating branch and a second temperature-regulating branch. The first temperature-regulating branch maintains the cooling medium in it at a first temperature, and the second temperature-regulating branch maintains the cooling medium in it at a second temperature. The two temperature-regulating branches are selectively connected to the high-voltage device under test (VTD) upstream of the VTD via a switching component.

[0038] When it is necessary to change the temperature state of the cooling medium, the switching component changes the temperature-regulating branch connected to the high-voltage device under test, allowing the cooling medium from the other temperature-regulating branch to enter the high-voltage device under test. The mass flow rate and inlet pressure of the cooling medium in the two temperature-regulating branches are adjusted respectively, so that the changes in mass flow rate and inlet pressure before and after the branch switching are kept within their respective preset ranges, thereby limiting the accompanying changes in flow rate and pressure while changing the temperature state of the cooling medium.

[0039] In one implementation, such as Figure 3 and Figure 4As shown, the device-side temperature is continuously monitored during the variable operating condition cycle. Preset upper and lower temperature limits are used to define the permissible temperature range of the device-side during the durability cycle. Corresponding warning intervals are set near the preset upper or lower temperature limits. The upper and lower temperature limits, and warning intervals are determined based on the permissible temperature range for the durability test of the high-voltage device under test. When the device-side temperature enters the corresponding warning interval, it indicates that continued operation under the current electrical load may cause the extreme device-side temperature in subsequent cycles to approach or exceed the stated temperature range.

[0040] At this point, the duration or amplitude of the electrical load in subsequent cycles is adjusted to alter the heat generation process of the high-voltage device under test. Since the time relationship of the first temperature response is also affected after the electrical load conditions change, the switching time of the cooling medium temperature state is re-determined based on the adjusted electrical load. This ensures that the extreme temperature value on the device side is maintained between the preset upper and lower temperature limits, while the first and second temperature responses maintain the target sequential interval.

[0041] In one implementation, such as Figure 3 As shown, the target sequence interval can be determined based on the actual operating data of the high-voltage device under test. The actual operating data includes the changes in electrical load state and thermal management state corresponding to the high-voltage device under test. By statistically analyzing the sequence and time interval of these two state changes, a representative sequence relationship in the actual operation is obtained, and the target sequence interval used for durability testing is determined accordingly.

[0042] In actual operation, various electrical load state changes and thermal management state changes can occur. For variable operating conditions with different target sequence intervals, corresponding variable operating condition cycles can be formed, and multiple variable operating condition cycles can be arranged according to the typical sequence reflected in the actual operating data, thus forming a durability test sequence. The specific switching time of each variable operating condition cycle is determined based on its corresponding target sequence interval, power thermal response delay, and cooling-side temperature response delay.

[0043] In one implementation, such as Figure 1 and Figure 4 As shown, during the variable operating condition cycle execution, the mass flow rate of the cooling medium, the inlet pressure, and the inlet temperature are also detected. When any of these parameters is detected to exceed its preset range, it indicates that the liquid cooling thermal boundary experienced by the current cycle does not meet the preset test conditions, and therefore the current cycle is not counted in the valid cycle count.

[0044] For this invalid cycle, the electrical load and thermal boundary are returned to a preset baseline state, and the mass flow rate, inlet pressure, and inlet temperature of the cooling medium are continuously monitored. Once the mass flow rate, inlet pressure, and inlet temperature have stabilized back within their respective preset ranges, the variable operating condition cycle is restarted. Thus, the final recorded number of valid cycles corresponds to the variable operating condition cycle completed under the preset liquid cooling thermal boundary conditions.

[0045] In one implementation, such as Figure 5 As shown, in terms of durability condition determination, diagnostics are performed before the start of the durability test and at predetermined cycle nodes. Each diagnostic uses the same baseline operating condition, subjecting the high-voltage device under test to the same diagnostic thermal excitation. The thermal response obtained before the start of the durability test is used as the initial thermal response, and the thermal response obtained at the predetermined cycle node is used as the current thermal response.

[0046] In one embodiment, the diagnostic thermal excitation is a preset electrical load pulse. During each diagnostic test, the inlet temperature, mass flow rate, and inlet pressure of the cooling medium are maintained under the same conditions, and the same preset electrical load pulse is applied to the high-voltage device under test. Device-side temperature data before the application of the electrical load pulse, device-side temperature data during the application of the electrical load pulse, and temperature data during the device-side temperature recovery process after the pulse ends are continuously collected, thereby obtaining a complete transient thermal response.

[0047] Before the endurance test begins, the initial thermal response is obtained according to the diagnostic conditions described above. After completing the predetermined number of cycles, the high-voltage device under test is placed under the same reference operating conditions, and the current thermal response is obtained using the same cooling medium inlet temperature, mass flow rate, inlet pressure, and the same preset electrical load pulse. Since the two diagnostic tests use the same reference operating conditions and thermal excitation, the two transient thermal responses can be compared according to the same thermal response index.

[0048] The change in the preset thermal response index can be calculated as the difference between the time required for the initial thermal response and the current thermal response to reach the preset response ratio. In this case, the same preset response ratio is applied to both the initial and current thermal responses, the corresponding response times are determined, and the difference between the two is obtained.

[0049] The change in the preset thermal response index can also be expressed as the temperature rise difference at a specified time. For both the initial and current thermal responses, select the same specified time corresponding to the diagnostic thermal excitation, determine the corresponding temperature rise, and compare the temperature rise differences.

[0050] The change in the preset thermal response index can also be expressed as the slope difference during the heating phase, by obtaining the corresponding slopes based on the heating phases of the initial thermal response and the current thermal response, and comparing the differences between the two; or by expressing the characteristic time difference during the cooling phase, by obtaining the corresponding characteristic time of the cooling phase during the temperature recovery process after the end of the pulses of the two thermal responses, and then comparing the differences between the two. At least one of the above preset thermal response indices can be used.

[0051] After obtaining the change in the preset thermal response index at each predetermined cycle node, the change is compared with a preset failure limit. The preset failure limit is determined before the test starts based on the durability judgment requirements corresponding to the high-voltage device under test, and the same index extraction and judgment conditions are used for the initial thermal response and each current thermal response. When the change reaches the preset failure limit, the high-voltage device under test is determined to have failed, and the cycle number corresponding to reaching the failure condition is recorded.

[0052] When the durability test has completed the target number of cycles, and the change in the current thermal response relative to the initial thermal response has not yet reached the preset failure limit, the high-voltage device under test is deemed to have passed the durability test. Therefore, the entire test uses the target number of cycles or reaching the preset failure limit as the conditions for forming the corresponding durability test results.

[0053] In a complete implementation process, such as Figures 1 to 5 As shown, the high-voltage device under test is first connected to the durability test bench, and the electrical load path and the liquid-cooled thermal boundary path are established respectively. First and second temperature data are obtained at the liquid-cooled heat exchange interface on the device side and the cooling side. Subsequently, the electrical load is changed under the condition of stable thermal boundary to determine the power thermal response delay; the temperature state of the cooling medium is changed under the condition of stable electrical load to determine the temperature response delay on the cooling side.

[0054] After obtaining two response delays, the electrical load switching time and the cooling medium temperature state switching time are determined according to the target sequence interval used in the durability test, and the variable operating condition cycle is started. After each cycle is completed, the actual sequence interval is determined according to the actual first and second temperature responses, and the cooling medium temperature state switching time of the subsequent cycle is corrected by using the deviation of the actual sequence interval from the target sequence interval.

[0055] During the endurance cycle, when the device-side temperature enters the warning range, the electrical load is adjusted according to the device-side temperature limit, and the corresponding cooling medium temperature state switching time is re-determined; when the cooling medium mass flow rate, inlet pressure, or inlet temperature exceeds the preset range, it is not included in the current cycle, and the cycle is re-executed after the preset reference state is restored.

[0056] After reaching the predetermined diagnostic cycle node, the high-voltage device under test is placed under the same baseline operating conditions as the initial diagnostic test, and the same diagnostic thermal excitation is applied to obtain the current thermal response, which is then compared with the initial thermal response obtained before the endurance test begins. If the change in the preset thermal response index reaches the preset failure limit, the high-voltage device under test is determined to have failed, and the corresponding cycle number is recorded; if the change still does not reach the preset failure limit after completing the target number of cycles, the endurance test is determined to have passed.

[0057] The specific embodiments described above further illustrate the inventive purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, or improvements made by those skilled in the art within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for endurance testing of thermally managed high-voltage devices under varying operating conditions, characterized in that, include: The high-voltage device under test with a liquid-cooled heat exchange interface is connected to a durability test bench. The durability test bench includes an electrical load circuit and a thermal boundary circuit that can switch between a first temperature state and a second temperature state while keeping the mass flow rate and inlet pressure of the cooling medium within their respective preset ranges. The first temperature data and the second temperature data are obtained by temperature detection points set on the device side and the cooling side of the liquid-cooled heat exchange interface, respectively. While maintaining a stable thermal boundary state, the electrical load is changed, and the power thermal response delay is determined based on the characteristic moment of the change in the first temperature data relative to the electrical load. While maintaining a stable electrical load, the cooling medium temperature state is switched, and the cooling side temperature response delay is determined based on the characteristic moment of the change in the second temperature data relative to the cooling medium inlet temperature. Based on the power thermal response delay, the cooling side temperature response delay, and the target sequence interval between the first temperature response and the second temperature response, the electrical load switching time and the cooling medium temperature state switching time in the variable operating condition cycle are determined. The variable operating condition cycle is executed, and the timing of the cooling medium temperature state switching in subsequent cycles is corrected based on the deviation between the actual and target intervals between the first and second temperature responses in each cycle. Before the endurance test and at predetermined cycle nodes, the same diagnostic thermal excitation is applied under the same baseline operating conditions to obtain the initial thermal response and the current thermal response. The endurance status is determined based on the change in the preset thermal response index of the two. When the change reaches the preset failure limit, the high voltage device under test is determined to have failed and the corresponding cycle number is recorded. The durability test is considered passed when the target number of cycles is completed and the change does not reach the preset failure limit.

2. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, Two adjacent variable operating condition cycles are respectively designated as the first stress cycle and the second stress cycle. The switching time of the cooling medium temperature state is adjusted according to the sign of the temperature difference between the first temperature data and the second temperature data, so that the temperature difference is kept positive during the preset observation period of the first stress cycle and the temperature difference is kept negative during the preset observation period of the second stress cycle, and the extreme values ​​of the device side temperature in the two stress cycles are between the same preset upper and lower temperature limits.

3. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, The thermal boundary loop includes a first temperature-regulating branch and a second temperature-regulating branch that maintain the cooling medium at a first temperature state and a second temperature state, respectively. The two temperature-regulating branches are selectively connected to the high-voltage device under test at the cooling medium inlet of the device under test through a switching component, and the mass flow rate and inlet pressure of the cooling medium of the two temperature-regulating branches are adjusted respectively so that the changes in mass flow rate and inlet pressure before and after the branch switching are within their respective preset ranges.

4. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, The characteristic moment is determined based on a preset proportion of the total response amplitude from the pre-change steady-state value to the post-change steady-state value in relation to the corresponding temperature response.

5. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, During the execution of variable operating condition cycles, the device-side temperature is monitored. When the device-side temperature enters the warning range adjacent to the preset upper or lower temperature limit, the duration or amplitude of the electrical load in the subsequent cycle is adjusted, and the switching time of the cooling medium temperature state is re-determined based on the adjusted electrical load, so that the extreme value of the device-side temperature is kept between the preset upper and lower temperature limits, and the first temperature response and the second temperature response are kept at the target interval.

6. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, The diagnostic thermal excitation involves applying a preset electrical load pulse to the high-voltage device under test under the same conditions of inlet temperature, mass flow rate and inlet pressure of the cooling medium in each diagnosis, and continuously collecting temperature data on the device side before, during and after the application of the electrical load pulse, and during the temperature recovery process after the pulse ends, to obtain the transient thermal response.

7. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 6, characterized in that, The change in the preset thermal response index includes at least one of the following: the difference in time required for the initial thermal response and the current thermal response to reach the preset response ratio, the temperature rise difference at a specified time, the slope difference during the heating phase, and the characteristic time difference during the cooling phase.

8. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, Based on the actual operating data of the high-voltage device under test, the typical sequence and time interval of changes in electrical load state and thermal management state are statistically analyzed, and the target sequence interval is determined accordingly. Multiple variable operating condition cycles with different target sequence intervals are arranged into a durability test sequence according to the typical sequence.

9. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 1, characterized in that, When the mass flow rate, inlet pressure, or inlet temperature of the cooling medium is detected to exceed their respective preset ranges, the current cycle is not counted in the effective cycle count, so that the electrical load and thermal boundary return to the preset reference state. After the mass flow rate, inlet pressure, and inlet temperature of the cooling medium stabilize within their respective preset ranges, the variable operating condition cycle is restarted.

10. The method for variable operating condition durability testing of thermal management high-voltage devices according to claim 2, characterized in that, The difference between the first temperature data and the second temperature data is changed from a positive value to a negative value or from a negative value to a positive value, and the corresponding sign is maintained continuously for a preset confirmation time after the conversion, which serves as the confirmation condition for the conversion of the thermal action direction. When the condition is met, the conversion of the thermal action direction of the liquid cooling heat exchange interface is confirmed. If the condition is not met, the switching time of the cooling medium temperature state or the set temperature of the first temperature state or the second temperature state is adjusted until the thermal action direction conversion confirmation condition is met.

Images