An experimental system for automatic power-on / off testing

Abstract

This invention discloses an automatic power-on / off testing experimental system, belonging to the field of automatic power-on / off testing technology. It is used to automatically measure the holding time of a single power-off phase in automatic power-on / off tests employing adjustable slope power supply, relay switching, and short-term power-on capacitor charging. After meeting the pre-test conditions, the system extracts the reference voltage content immediately adjacent to the lowest operating voltage boundary and uses a unified standard to determine the start and end points of the window around the power-off transition process, generating the power-off holding time. Through this design, it can suppress boundary mis-selection under continuous transition conditions, reduce the transmission influence of start-point error to end-point judgment, and improve the comparability of power-off holding results from different rounds.

Description

Technical Field

[0001] This invention relates to the field of automatic power-on / off testing technology, and more specifically, to an experimental system for implementing automatic power-on / off testing. Background Technology

[0002] Automatic power-on / off testing technology is typically used to repeatedly power on, power off, and briefly interrupt the power supply to the device under test (DUT) under automatic cyclic conditions to examine its stability and durability under different power supply conditions. In this type of testing, existing technologies usually first establish a single-cycle power-off test condition, then continuously sample the voltage at the DUT during the power-off phase, and determine the boundary based on a preset duration, termination voltage, or local voltage change characteristics, thereby calculating the hold time. For situations where the power-off boundary is relatively clear or the transition process is short, the above system can usually complete the basic hold time measurement.

[0003] The existing technology has the following shortcomings: On the one hand, in test scenarios employing adjustable slope power supply, relay switching, and short-term power-up via a power-up capacitor, the power-down phase typically involves a continuous process of controlled descent, relay release transition, and power-up takeover maintenance, which can easily lead to confusion between voltage changes in different stages. If the window start point is still determined based on a fixed duration or a single local change characteristic, it can easily result in incorrect starting point boundary determination. The challenge lies in the lack of a unified boundary determination criterion for continuous transition processes in existing systems. On the other hand, the relay release length can drift with historical cycles and device states, making the position and length of the transition process in different cycles inconsistent. If the window end point is still determined solely based on the termination voltage or cumulative drop threshold, the starting point error can easily propagate to the end point determination, making it difficult to establish a unified standard for power-down hold-up time between different cycles. The challenge lies in the lack of a determination criterion that can maintain boundary consistency under cycle drift conditions. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an automatic power-on and power-off test experimental system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: An automatic power-on / off testing experimental system, comprising: The operating condition establishment unit is used to establish a controlled slow-descent operating condition of the voltage at the measured terminal when the preconditions are met, and to trigger the cancellation relay coil drive signal to make the measured terminal enter the power-down holding window search process. The voltage sampling unit is used to extract the front-end reference segment and form the candidate sampling point content for the start-point search after the relay coil drive signal is canceled, and form the candidate sampling point content for the end-point search after the start-point of the power-down retention window is determined. The window start point determination unit is used to form a comparison benchmark for the descent exit condition based on the previous reference segment. For each candidate sampling point in the candidate sampling point content of the start point search, it first determines whether the candidate sampling point meets the descent exit condition. Then, it takes the candidate sampling point as the start point and extracts the start point confirmation segment according to the modified release reference duration, and determines whether the power takeover condition is met within the start point confirmation segment. The candidate sampling point that meets both the descent exit condition and the power takeover condition for the first time is determined as the start point of the power failure retention window. The window endpoint determination unit is used to determine whether the cumulative voltage drop from the start of the power-down retention window to the candidate sampling point meets the allowable voltage drop depletion condition for each candidate sampling point formed with the start of the power-down retention window as the starting point, and then determine whether the candidate sampling point meets the power-up retention exit condition; the candidate sampling point that meets both the allowable voltage drop depletion condition and the power-up retention exit condition for the first time is determined as the power-down retention window endpoint. The hold-up time generation unit is used to generate the power-down hold-up time based on the time interval between the start and end of the power-down hold-up window under a uniform sampling timing.

[0006] In a preferred embodiment, the operating condition establishment unit generates a relay status detection result by reading back a signal corresponding to the relay contact status, and generates a power supply capability detection result by performing a power supply capability detection on the power supply capacitor after precharging. The detection quantity for power supply capability detection includes at least the voltage across the power supply capacitor or a power supply support capability characterization quantity based on the short-term discharge response after precharging. When the relay status detection result meets the current round of test switching requirements and the detection quantity falls within a preset available range, the precondition is determined to be met.

[0007] In a preferred embodiment, the operating condition establishment unit establishes a controlled descent condition for the measured terminal voltage at a fixed descent slope or a segmented descent slope; the target extraction interval is a predetermined interval reserved before the lowest operating voltage boundary and used to cover the last complete descent control cycle; when the measured terminal voltage enters the target extraction interval, the deactivation relay coil drive signal is triggered.

[0008] In a preferred embodiment, the voltage sampling unit extracts the measured terminal voltage content from the last complete descent control cycle immediately adjacent to the lowest operating voltage boundary to form the front reference segment; the complete descent control cycle is a single continuous descent control cycle with complete start and end boundaries during the controlled descent process and which is not cut off by the revocation of the relay coil drive action.

[0009] In a preferred embodiment, during the start-up search phase after the relay coil drive signal is removed, the voltage sampling unit continuously reads the voltage of the measured terminal using the same fixed short sampling period to form candidate sampling points for the start-up search; and during the end-up search phase after the start of the power-down retention window is determined, it continuously reads the voltage of the measured terminal using the same fixed short sampling period to form candidate sampling points for the end-up search. The candidate sampling points for the start-up search are formed from the moment corresponding to the removal of the relay coil drive signal, and the candidate sampling points for the end-up search continue to be formed from the moment corresponding to the start of the power-down retention window.

[0010] In a preferred embodiment, the window start point determination unit forms a natural fluctuation band based on the statistical upper and lower bounds of the voltage drop of each adjacent sampling point in the previous reference segment; when there are obvious abnormal sampling points, the obvious abnormal sampling points are first removed, and then the natural fluctuation band is formed; for each candidate sampling point, the local drop rate characterized by the voltage change between the current candidate sampling point and the previous sampling point is calculated, and the descent exit condition is determined based on whether the local drop rate deviates from the natural fluctuation band.

[0011] In a preferred embodiment, the window start point determination unit extracts a start point confirmation segment from the candidate sampling point that meets the descent exit condition, starting from the candidate sampling point and extending backward according to the corrected release reference duration. The corrected release reference duration is jointly determined by the baseline release reference duration and the release drift correction amount. The release drift correction amount is determined based on the average or median offset of the actual release duration of each relay in a preset historical cycle relative to the baseline release reference duration. When a preset number of consecutive sampling points in the start point confirmation segment continue to descend in the same direction, and the descent amount of adjacent sampling points does not show a reverse jump, and the change in the descent amount between adjacent sampling points does not exceed a preset interruption threshold, the start point confirmation segment is determined to meet the power takeover condition.

[0012] In a preferred embodiment, the window endpoint determination unit uses the voltage corresponding to the start point of the power-down retention window as a unified starting point. For each candidate sampling point, the difference between the unified starting point voltage and the voltage corresponding to the current candidate sampling point is used to form a cumulative voltage drop, and the cumulative voltage drop is used to determine whether the allowable drop depletion condition is met. The preset allowable drop is given by the allowable difference between the minimum operating voltage boundary and the voltage corresponding to the start point of the power-down retention window, or it is preset according to the allowable drop range in the test standard.

[0013] In a preferred embodiment, the window endpoint determination unit reads a preset number of consecutive sampling points after the current candidate sampling point; the window endpoint determination unit adopts a continuous observation length consistent with the power supply takeover determination of the window start point determination unit, and determines a preset number of consecutive sampling points based on the correspondence between the continuous observation length and the fixed short sampling period; when the decrease amount of adjacent sampling points in the preset number of consecutive sampling points after the current candidate sampling point shows a reverse jump, or the change amplitude between the decrease amounts of adjacent sampling points exceeds a preset maintenance threshold, the current candidate sampling point is determined to meet the power supply maintenance exit condition; the candidate sampling point that first simultaneously meets the allowable drop depletion condition and the power supply maintenance exit condition is determined as the power-down maintenance window endpoint.

[0014] In a preferred embodiment, the hold time generation unit reads the sampling time corresponding to the start of the power-down hold window and the sampling time corresponding to the end of the power-down hold window under a unified sampling timing, and generates the power-down hold time by subtracting the sampling time corresponding to the start of the window from the sampling time corresponding to the end of the window.

[0015] The advantages and effects of the automatic power-on / off testing experimental system of the present invention are as follows: This invention provides an automated power-on / off testing system. By applying a unified standard to determine the start and end points of the power-off hold window, the system differentiates the controlled descent tail-end residual, relay release transition, and power-on takeover processes during boundary determination, thereby improving the stability of window boundary determination. Furthermore, the start and end points of the window form a closed loop within the same decision chain and unified sampling timing, controlling the propagation of errors from the start point to the end point, ensuring a consistent basis for determining power-off hold times generated in different rounds. Therefore, the generated power-off hold times can more stably characterize the single-round power-off hold capability and provide directly usable results for subsequent round comparisons, statistical analysis, and test result recording. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 A flowchart for determining the power-down retention window and generating the retention time is provided. Figure 3 A schematic diagram illustrating the power-off retention window determination; Figure 4 This is a schematic diagram of the connection topology between the relay array and the device under test in the test platform; Figure 5 This is a schematic diagram of the short-term power supply branch in the test platform; Figure 6 This is a schematic diagram of the hardware protection circuits for overvoltage, undervoltage, and overcurrent in the test platform. Detailed Implementation

[0017] The technical solutions of the embodiments 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] This invention provides an automatic power-on / off testing experimental system, applicable to the automatic measurement of the hold duration during a single power-off phase in automatic power-on / off tests employing adjustable slope power supply, relay switching, and short-term power-on capacitor charging. The system standardizes the determination of the start and end points of the power-off hold window, forming the basis for subsequent hold time generation and round comparison. Existing systems typically first establish a single-round power-off test condition, then continuously sample the voltage during the power-off phase after removing the relay coil drive signal, and determine the window boundaries based on fixed duration positions, termination voltages, or single local change characteristics. For cases where the power-off boundaries are relatively clear or the transition process is short, the aforementioned system can usually complete the basic hold time measurement.

[0019] However, in test systems employing adjustable slope power supply, relay switching, and short-term power sustaining via a power-down capacitor, the power-down phase is not a direct hard disconnection. Instead, it first undergoes controlled descent, followed by relay release transition, and then enters the power sustaining process. This causes the residual voltage changes at the tail end of the controlled descent, the relay release process, and the initial stage of the power sustaining process to overlap. Furthermore, the relay release length drifts with historical cycles and device states, resulting in inconsistent positions and lengths of transition processes in each cycle. If existing systems still determine window boundaries based on fixed duration positions or single voltage change characteristics, they are prone to prematurely incorporating intervals that have not yet exited the tail end of the descent phase into the power-down sustaining window at the starting point, and to prematurely truncate the window at the end point before the power sustaining process has truly exited. Alternatively, the error may be propagated to the end-point judgment after the starting point has drifted, leading to unstable positioning of the power-down sustaining time.

[0020] Therefore, under the conditions of controlled descent to continuous transition to power-on takeover and the relay release length drifting with historical cycles and device states, this invention stably determines the starting and ending boundaries of the power-down retention window around the voltage transition process of a single power-down phase, and generates comparable power-down retention times under a unified boundary caliber.

[0021] Based on the above design, this invention constructs a test processing flow corresponding to an automatic power-on / off test experimental system. (Refer to...) Figure 1 , Figure 1 The following is a schematic diagram of the system structure of the present invention.

[0022] The operating condition establishment unit is used to pre-determine whether the current round meets the conditions for entering the starting point search process before the start of the power-down retention window search begins. If the preconditions for the starting point search are met, it establishes the controlled descent pre-state of the measured terminal voltage, forming the starting point search permission result R101. This unit reads the relay status detection result X101, the power supply capability detection result X102, and the measured terminal voltage sampling result X103, and completes the precondition determination, controlled descent process establishment, and target extraction interval trigger cancellation based on X101, X102, and X103. R101 includes at least the relay status availability, power supply capability availability, and the established controlled descent pre-state. Among them, the starting point search precondition refers to the entry condition that the current round simultaneously meets the conditions that the relay status is available and the power supply capability is available; the controlled descent prestate refers to the prestate that the voltage at the measured end has formed a continuous descent tail segment according to the predetermined descent diameter before the relay coil drive signal is removed; the target extraction interval refers to the predetermined interval that is reserved before the lowest working voltage boundary and can cover the last complete descent control cycle.

[0023] The voltage sampling unit, upon receiving the start-point search permission result R101, generates a boundary search support result R102 based on the measured terminal voltage sampling result X103, which is directly used for subsequent window boundary determination. This unit reads the start-point search permission result R101 and the measured terminal voltage sampling result X103, and completes the extraction of the front-end reference segment, the formation of candidate content on the start side, and the formation of candidate content on the end side based on R101 and X103. R102 includes at least the front-end reference segment, candidate sampling point content for the start-point search, and candidate sampling point content for the end-point search. The front-end reference segment refers to the continuous voltage content extracted from the last complete descent control cycle immediately adjacent to the lowest operating voltage boundary; the candidate sampling point content for the start-point search refers to the candidate sampling point content formed point by point at a fixed short sampling period after the relay coil drive signal is removed; and the candidate sampling point content for the end-point search refers to the candidate sampling point content formed point by point at the same fixed short sampling period after the window start point is determined.

[0024] The window start point determination unit reads the contents of the preceding reference segment and the candidate sampling points for the start point search from the boundary search support result R102, and completes the dual-condition joint determination of the start point of the power-down retention window, forming the start point R103 of the power-down retention window. This unit reads the boundary search support result R102, and based on R102, completes the formation of the start-side comparison benchmark, the judgment of gradual descent exit, the interception of the start point confirmation segment and the judgment of power takeover, and the locking of the first point that simultaneously meets the conditions. R103 is the candidate sampling point that simultaneously meets the conditions of gradual descent exit and power takeover for the first time when searching from front to back along the candidate sampling point contents for the start point search. The preceding reference segment is used to form a comparison benchmark on the starting side; the starting point confirmation segment refers to the continuous voltage content intercepted from the candidate sampling point that meets the descent exit condition, according to the modified release reference duration, to determine whether the local voltage change after the candidate sampling point has reflected the characteristics of continued power takeover; the modified release reference duration is used to limit the coverage length of the starting point confirmation segment; the preset interruption threshold is used to limit the allowable upper limit of the descent change amplitude of adjacent sampling points in the continued power takeover judgment; the continuous observation length is a preset observation time length used to determine whether the continued power takeover or continued power maintenance state is continuously established, and the preset number of continuous sampling points is determined according to the correspondence between the continuous observation length and the fixed short sampling period; the preset starting point search interval is used to limit the time range of the starting point of the point-by-point search window of this unit.

[0025] The window endpoint determination unit reads the candidate sampling point content for endpoint search from the boundary search support result R102, and combines it with the power-down retention window start point R103 to complete the dual-condition joint determination of the power-down retention window endpoint, forming the power-down retention window endpoint R104. This unit reads the boundary search support result R102 and the power-down retention window start point R103, and based on R102 and R103, completes the establishment of a unified starting benchmark, calculation of cumulative voltage drop and judgment of allowable voltage drop depletion, judgment of continued power maintenance exit, and locking of the first point that simultaneously meets the condition. R104 is the candidate sampling point that first simultaneously meets the allowable voltage drop depletion condition and the continued power maintenance exit condition when searching from front to back along the candidate sampling point content for endpoint search. R103 serves as a unified starting point for calculating the cumulative voltage drop; the preset allowable drop is used to limit the upper limit of the allowable cumulative voltage drop from the start of the window; the preset maintenance threshold is used to limit the allowable upper limit of the change in the voltage drop of adjacent sampling points under the power maintenance state; the continuous observation length is consistent with the window start determination unit, and the preset number of continuous sampling points is determined based on the correspondence between the continuous observation length and the fixed short sampling period; the preset endpoint search interval is used to limit the time range of the point-by-point search window endpoint of this unit.

[0026] The hold time generation unit is used to generate the power-down hold time R105 under a unified sampling timing sequence after the start point R103 and end point R104 of the power-down hold window have been effectively determined. This unit reads the start point R103 and end point R104 of the power-down hold window, and performs boundary validity verification, unified timing sequence reading and time interval calculation, and result generation and output based on R103 and R104. R105 is the effective hold duration represented by the time interval between the sampling time corresponding to the start point of the window and the sampling time corresponding to the end point of the window. Among them, the sampling time corresponding to the start point of the window refers to the sampling time corresponding to R103 under the unified sampling timing sequence; the sampling time corresponding to the end point of the window refers to the sampling time corresponding to R104 under the unified sampling timing sequence; the unified sampling timing sequence is a unified sampling time caliber consistent with the fixed short sampling period; the equivalent sampling point conversion caliber is only used as an equivalent calculation method for the time interval under the unified sampling timing sequence, and does not change the main calculation caliber of subtracting the sampling time corresponding to the start point of the window from the sampling time corresponding to the end point of the window.

[0027] In summary, this invention no longer uses fixed-duration or single-threshold methods to select the power-down hold window boundaries separately. Instead, it uses a unified method to determine the window start and end points under the same decision chain, preventing confusion between the controlled descent tail-end residual, relay release transition, and power-on takeover processes. This design reduces the transmission impact of incorrect start point selection on end point determination, avoids overall window boundary shifting forward or backward, and ensures a consistent decision basis for power-down hold times generated in different rounds. Therefore, the power-down hold time can more stably characterize the single-round power-down hold capability and provides a directly usable closed-loop result for subsequent round comparisons, statistical analysis, and test result recording.

[0028] The following detailed description, with reference to specific embodiments, illustrates the process and operational effects of the collaborative execution of power-down retention window determination and retention time generation by various units in the system. It should be understood that these embodiments are merely illustrative of the technical solutions of the present invention and are not intended to limit it. Without altering the essence of the invention, relevant steps, parameters, and unit divisions can be appropriately adjusted.

[0029] In an optional embodiment, the working condition establishment unit reads X101, X102, and X103, and performs precondition determination, controlled descent process establishment, and target extraction interval trigger cancellation to form a starting point search permission result R101. In the precondition determination, this unit starts from X101 and X102 and jointly judges whether the current round meets the starting point search preconditions. X101 is obtained through the readback signal corresponding to the relay contact state and is used to characterize whether the relay is in the switching state required for the current round of testing. X102 is obtained through the energy-saving capacity detection after precharging the energy-saving capacitor. The detection quantity of the energy-saving capacity detection includes at least the voltage across the energy-saving capacitor, or the energy-saving support capacity characterization quantity formed based on the short-time discharge response after precharging. When the readback signal indicates that the relay is in the switching state required for the current round of testing, and the detection quantity falls within the preset available range, it is determined that the current round meets the starting point search preconditions; when any condition is not met, R101 is not formed, and the current round ends or enters the corresponding state branch.

[0030] During the controlled descent process establishment, this unit starts from X103 and continuously reduces the power supply output according to a predetermined descent range. This causes the voltage at the measured terminal to enter a continuous descent process towards the minimum operating voltage boundary, forming the controlled descent pre-state required for subsequent window start determination. The minimum operating voltage boundary is the lowest voltage boundary allowed for the device under test to maintain normal operation. The predetermined descent range is preferably a fixed descent slope or a segmented descent slope, used to ensure that the voltage at the measured terminal maintains a continuous sampling descent process before approaching the minimum operating voltage boundary.

[0031] In the target extraction interval trigger cancellation, this unit starts from the current sampled value of X103 corresponding to the controlled descent process, and determines whether the voltage of the measured terminal has entered the position close to the minimum working voltage boundary according to the target extraction interval; when the voltage of the measured terminal enters the target extraction interval, this unit ends the controlled descent process establishment stage and triggers the cancellation relay coil drive signal, so that the current round enters the subsequent starting point search process.

[0032] Through the above processing, this unit generates the starting point search permission result R101, and writes the relay status availability, power supply availability, and controlled descent pre-state establishment status into R101. Then, R101 is provided to the voltage sampling unit for reading, serving as the admission basis for the subsequent front-end reference segment extraction and boundary search support result formation.

[0033] In an optional embodiment, the voltage sampling unit reads R101 and X103 and performs front-end reference segment extraction, start-end candidate content formation and end-end candidate content formation to form boundary search support result R102.

[0034] In the extraction of the preceding reference segment, this unit starts from R101 and X103. After confirming that the current round is allowed to enter the subsequent window start-up search process, it extracts the preceding reference segment from the last complete descent control cycle immediately adjacent to the lowest operating voltage boundary according to the target extraction interval, and writes it into R102. A complete descent control cycle is a single continuous descent control cycle with complete start and end boundaries during controlled descent and not interrupted by the revocation relay coil drive action. Its start point is the start time of this continuous descent control, and its end point is the end time of this continuous descent control. The preceding reference segment is used to characterize the natural variation characteristics of the descent tail segment and to provide the window start-up determination unit with a comparison benchmark on the start side.

[0035] During the formation of candidate content on the starting side, this unit starts from X103, reads the voltage of the measured terminal point by point at a fixed short sampling period after the relay coil drive signal is removed, and writes the candidate sampling point content for the starting point search into R102. The fixed short sampling period remains consistent in both the starting point search phase and the ending point search phase, and its length is sufficient to distinguish the voltage transition process after the relay coil drive signal is removed and the voltage changes near the window boundary. The candidate sampling point content for the starting point search is formed from the moment the relay coil drive signal is removed and is continuously updated until the window starting point determination unit determines the candidate sampling point corresponding to the starting point of the power-down retention window.

[0036] In the formation of candidate content on the endpoint side, this unit starts from X103. After the window start-up determination unit determines the start point of the power-down retention window, it continues to read the voltage of the measured terminal point point by point according to the same fixed short sampling period, and writes the candidate sampling point content for endpoint search into R102. The candidate sampling point content for endpoint search continues to be formed from the time corresponding to the start point of the power-down retention window, and is continuously updated until the window endpoint determination unit determines the candidate sampling point corresponding to the endpoint of the power-down retention window. The candidate sampling point corresponding to the start point of the power-down retention window serves as both the termination position of the start-up content and the starting position of the endpoint content.

[0037] Through the above processing, this unit generates the boundary search support result R102, and writes the preceding reference segment, the candidate sampling point content for the starting point search, and the candidate sampling point content for the ending point search into R102. R102 is then provided to the window starting point determination unit and the window ending point determination unit for reading. Specifically, the window starting point determination unit reads the preceding reference segment and the candidate sampling point content for the starting point search from R102, and the window ending point determination unit reads the candidate sampling point content for the ending point search from R102.

[0038] like Figure 2 As shown, in this embodiment, the starting point R103 of the power-down retention window is first determined based on R102, and the ending point R104 of the power-down retention window is determined based on R102 and R103. Then, the power-down retention time R105 is generated based on R103 and R104.

[0039] In an optional embodiment, the window start point determination unit reads R102 and performs start point side comparison benchmark formation, descent exit judgment, start point confirmation segment interception and power takeover judgment, and first simultaneous satisfaction of point locking to form the power failure retention window start point R103.

[0040] In the formation of the reference comparison at the starting point, this unit reads the preceding reference segment from R102 and forms a natural fluctuation band based on the voltage drop of each adjacent sampling point within the preceding reference segment. This natural fluctuation band serves as the reference comparison for subsequent descent exit judgment. The natural fluctuation band is formed by the statistical upper and lower bounds of the voltage drop of each adjacent sampling point within the preceding reference segment after removing obviously abnormal sampling points, and is used to characterize the natural variation range of the controlled descent tail segment. When obviously abnormal sampling points exist, they are first removed, and then the voltage drop of the remaining adjacent sampling points is statistically analyzed to form the natural fluctuation band.

[0041] In the descent exit judgment, this unit reads the candidate sampling points for the starting point search point by point from R102, calculates the corresponding local descent velocity for each candidate sampling point, and then compares the local descent velocity with the natural fluctuation band. The local descent velocity refers to the local descent characteristic equivalent to the voltage change between the current candidate sampling point and its previous sampling point under a fixed short sampling period. Since the fixed short sampling period is consistent, the voltage change between adjacent sampling points can reflect the change in local descent velocity. When the local descent velocity still falls within the natural fluctuation band, the candidate sampling point is determined not to meet the descent exit condition; when the local descent velocity leaves the natural fluctuation band, the candidate sampling point is determined to meet the descent exit condition, and the candidate sampling point is used as the starting candidate point for the subsequent starting point confirmation segment.

[0042] In the starting point confirmation segment interception and power takeover judgment, this unit intercepts the starting point confirmation segment for each candidate sampling point that meets the slow descent exit condition, based on the corrected release reference duration. Within the starting point confirmation segment, it sequentially reads the continuous sampling points following the candidate sampling point to determine whether the power takeover condition is met. The corrected release reference duration is jointly determined by the baseline release reference duration and the release drift correction amount. The release drift correction amount is determined based on the average or median offset of the actual release duration of each relay in a preset historical cycle relative to the baseline release reference duration. The continuous observation length is determined based on the time length used to cover the minimum observation interval for continuous maintenance after release under a fixed short sampling period. The preset number of continuous sampling points is calculated based on the correspondence between the continuous observation length and the fixed short sampling period. When the preset number of continuous sampling points within the starting point confirmation segment maintain a downward trend, and the descent amount of adjacent sampling points does not show a reverse jump, and the change in the descent amount between adjacent sampling points does not exceed the preset interruption threshold, the starting point confirmation segment is determined to meet the power takeover condition; otherwise, the starting point confirmation segment is determined not to meet the power takeover condition.

[0043] In the initial simultaneous locking of the starting point, this unit searches for candidate sampling points in chronological order from front to back, identifying those that simultaneously meet both the descent exit condition and the power takeover condition as valid starting point candidates. It then locks the candidate sampling point that first simultaneously meets both conditions and writes this candidate sampling point as the starting point R103 of the power-down hold window. After R103 is formed, it is output to the window end-of-window determination unit and the hold-time generation unit for reading. For the window end-of-window determination unit, R103 serves as the unified starting point for the end-of-window calculation. If no candidate sampling point simultaneously meeting both the descent exit condition and the power takeover condition is found within the preset starting point search interval, this unit does not form R103, but only retains the undetermined starting point state as a boundary branch for subsequent process handling.

[0044] In an optional embodiment, the window end point determination unit reads R102 and R103, and performs unified starting benchmark establishment, cumulative voltage drop calculation and allowable voltage drop depletion judgment, power maintenance exit judgment, and first simultaneous point locking to form the power failure retention window end point R104.

[0045] In establishing a unified starting point, this unit reads R103 and retrieves the candidate sampling point content for endpoint search from R102, using the voltage corresponding to R103 as the unified starting point for cumulative voltage drop. The candidate sampling point content for endpoint search is pre-formed by the voltage sampling unit and written into R102. This unit only reads and judges this content, without performing any formation or appending actions.

[0046] In the calculation of cumulative voltage drop and the determination of allowable voltage drop depletion, this unit reads the corresponding voltage of each candidate sampling point in the candidate sampling point content of the endpoint search, and forms the cumulative voltage drop corresponding to the current candidate sampling point by the difference between the voltage corresponding to R103 and the voltage corresponding to the current candidate sampling point. The cumulative voltage drop is used to characterize the extent to which the sustaining voltage support has been consumed since the start of the window. Subsequently, this unit compares the cumulative voltage drop corresponding to the current candidate sampling point with the preset allowable voltage drop. When the cumulative voltage drop reaches the preset allowable voltage drop, the current candidate sampling point is determined to meet the allowable voltage drop depletion condition; when the cumulative voltage drop does not reach the preset allowable voltage drop, the current candidate sampling point is determined not to meet the allowable voltage drop depletion condition. The preset allowable voltage drop is determined based on the allowable voltage drop range of the device under test, preferably given by the allowable difference between the minimum operating voltage boundary and the voltage corresponding to the start of the window, or preset according to the allowable voltage drop range in the test standard.

[0047] In the power sustaining exit judgment, this unit reads the continuous sampling points after the current candidate sampling point and determines whether the continuous sampling points still maintain the continuous downlink sustaining state corresponding to the power sustaining takeover phase. The continuous downlink sustaining state refers to a continuous descent state where, after the window start-up judgment unit has determined that the power sustaining takeover phase has begun, the descent amount of adjacent sampling points has not shown a reverse jump, and the change in the descent amount between adjacent sampling points does not exceed a preset sustaining threshold. The preset sustaining threshold is set based on the sampling resolution and the upper limit of the allowable local descent amount fluctuation, and can be corrected by combining historical calibration results if necessary. The window end-up judgment unit adopts the same continuous observation length as the power sustaining takeover judgment of the window start-up judgment unit, and determines a preset number of continuous sampling points for the power sustaining exit judgment based on the correspondence between the continuous observation length and the fixed short sampling period, to maintain the continuity between the start-up judgment and the end-up judgment. When a continuous downward instability occurs in a preset number of consecutive sampling points after the current candidate sampling point, or when the change in the descent amount between adjacent sampling points exceeds a preset maintenance threshold, the current candidate sampling point is determined to meet the continued energy maintenance exit condition; when the preset number of consecutive sampling points still maintain a continuous downward maintenance state, the current candidate sampling point is determined not to meet the continued energy maintenance exit condition; the change in the descent amount between adjacent sampling points is the absolute value of the difference between two adjacent descent amounts, and this absolute value is compared with the preset maintenance threshold.

[0048] In the initial simultaneous fulfillment of the point locking condition, this unit searches for candidate sampling points in chronological order along the endpoint, checking both the allowable drop depletion condition and the power sustaining exit condition. The candidate sampling point that first simultaneously meets both conditions is written as the endpoint R104 of the power-down hold window. R104 is then output to the hold time generation unit as the termination boundary for hold time calculation. If no candidate sampling point simultaneously meeting both the allowable drop depletion condition and the power sustaining exit condition is found within the preset endpoint search interval, this unit does not generate R104, but only retains the endpoint undetermined state as a boundary branch for subsequent processing.

[0049] In an optional embodiment, the hold time generation unit reads R103 and R104, and performs boundary validity verification, unified timing reading and time interval calculation, and result generation and output to form the power-down hold time R105.

[0050] During boundary validity verification, this unit reads R103 and R104 and determines whether the window start and end points have been effectively formed. When R103 or R104 is not effectively determined, this unit does not perform hold time value generation, but only retains the ungenerated state of the corresponding round; when both R103 and R104 have been effectively formed, this unit enters the unified timing reading and time interval calculation stage.

[0051] In the unified timing sequence reading and time interval calculation, this unit uses a unified sampling timing sequence as the time comparison caliber, and reads the sampling time of the window start corresponding to R103 and the sampling time of the window end corresponding to R104 accordingly. Subsequently, this unit subtracts the sampling time of the window start corresponding to the sampling time of the window end from the sampling time of the window end to form the time interval, which is used as the direct calculation quantity for generating the power-down hold-up time. The unified sampling timing sequence is consistent with the fixed short sampling period, using the sampling time corresponding to the same sampling sequence as the time caliber, thus ensuring that the window start and window end have a consistent time comparison basis.

[0052] In result generation and output, this unit writes the aforementioned time interval as the power-off hold-up time R105, and outputs R105 as the effective hold-up duration result for the current round. Once R105 is generated, it can be associated with the current round's test identifier and written into the test result record for subsequent statistical analysis, round comparison, or test report generation. When this unit retains the "not generated" state, it does not output the hold-up time value, but only retains the ungenerated result for that round. This unit associates the current round's test identifier with the "start point undetermined" state, "end point undetermined" state, or "no generated result" state and writes it into the test result record for subsequent round comparison, statistical analysis, or test report generation.

[0053] In some implementations, when the sampling times corresponding to the window start and end points are both within the same unified sampling time sequence, this unit can also form an equivalent time interval to the time difference by combining the number of sampling point intervals between the window start and end points with a fixed short sampling period. The conversion of the number of sampling point intervals is only used as a supplementary caliber for time difference calculation, to achieve an equivalent conversion consistent with the main calculation caliber under a unified sampling time sequence, without changing the result generation method that uses the difference between the sampling times corresponding to the window end point and the sampling times corresponding to the window start point as the main caliber.

[0054] In one specific embodiment, taking a single-round power-down test scenario of an industrial control device under test with a rated operating voltage of 24 volts and a full-load operating current of 2 amps on an automatic power-on / off test platform as an example, the process of the aforementioned units collaboratively completing the determination of the power-down hold window and the generation of the hold time is explained. Figures 3 to 6 As shown, the test platform includes an adjustable voltage source, a relay array, an MCU control unit, a sampling unit, and a hardware protection unit. The adjustable voltage source outputs a DC voltage with an adjustable slope. The relay array is connected to the adjustable voltage source and the device under test and has normally closed contacts and a power supply capacitor. The MCU control unit is used to set test parameters and acquire the voltage and current of the device under test. The hardware protection unit is set independently of the MCU control unit and is used to cut off the output in case of overvoltage, undervoltage, or overcurrent. The test platform adopts a voltage slope adjustable power supply, a relay array, and an MCU closed-loop control structure, and uses 16-bit ADC differential sampling as the basis for sampling the voltage of the device under test. The power supply branch consists of a double-pole double-throw relay and a 4700 microfarad low-ESR power supply capacitor.

[0055] In this embodiment, the aforementioned test platform is used to support the collaborative processing of various units within the system. The current round first reads the relay status detection result X101, the power supply capability detection result X102, and the voltage sampling result of the tested terminal X103, and completes the precondition determination, the establishment of the controlled descent pre-state, and the trigger cancellation of the target extraction interval. When the relay status meets the current round's test switching requirements and the power supply capability meets the current round's power supply requirements, a starting point search permission result R101 is formed; when either condition is not met, R101 is not formed, and the current round ends or transitions to the corresponding state branch. R101 includes at least the relay status availability, the power supply capability availability, and the established controlled descent pre-state, serving as the admission basis for the formation of subsequent boundary search support results.

[0056] For ease of combination Figures 4 to 6Understand the connection and protection relationships of the test platform. The markings in the diagram are explained as follows: K1 represents the first relay, K2 represents the second relay, C1 represents the regenerative capacitor, Q2 represents the protective switch connected in series in the power supply output path, R5 represents the sampling resistor, U1 represents the undervoltage comparator, U2 represents the overvoltage comparator, U3 represents the overcurrent comparator, J1-1 represents the output terminal connected to the positive terminal of the device under test (DUT), J1-2 represents the output terminal connected to the negative terminal of the DUT, DUT+ represents the positive terminal of the DUT, and DUT- represents the negative terminal of the DUT. Figure 4 In the diagram, COM1 and COM2 represent the common terminals of the relay, NC1 and NC2 represent the normally closed terminals of the relay, and NO1 and NO2 represent the normally open terminals of the relay. MCU-DO1 and MCU-DO2 in the diagram represent the control output signals used to drive the relay coil.

[0057] like Figure 4 As shown, the adjustable voltage source in the test platform is connected to the device under test (DUT) via a relay array. The normally closed contacts and the recharge capacitor in the relay array constitute a short-time recharge branch. The sampling unit reads the voltage sampling result X103 from the DUT. The relay status detection result X101 and the recharge capability detection result X102 are also formed under this connection topology, thus providing a platform connection basis for the operating condition establishment unit to form the starting point search permission result R101 and the voltage sampling unit to form the boundary search support result R102. After R101 is effectively formed, the voltage sampling unit forms the boundary search support result R102 based on R101 and X103. Specifically, after the DUT voltage enters the target extraction range adjacent to the minimum operating voltage boundary, the previous reference segment is extracted from the last complete descent control cycle adjacent to the minimum operating voltage boundary. After the relay coil drive signal is removed, the candidate sampling point content for the starting point search is formed point by point according to a fixed short sampling period. After the window starting point determination unit determines the starting point of the power-down retention window, the candidate sampling point content for the endpoint search is formed point by point according to the same fixed short sampling period. Therefore, R102 includes at least a front reference segment, candidate sampling point content for the start-point search, and candidate sampling point content for the end-point search, which are respectively provided to the window start-point determination unit and the window end-point determination unit for reading. The candidate sampling point content for both the start-point and end-point searches are candidate sampling point contents formed according to a fixed short sampling period and arranged in the sampling time sequence. Each candidate sampling point content includes at least the corresponding sampling time and the corresponding measured terminal voltage value.

[0058] During the window start-up determination phase, the window start-up determination unit reads the contents of the preceding reference segment and the candidate sampling points for start-up search from R102. First, it forms a start-up comparison benchmark based on the preceding reference segment. Then, it sequentially performs descent exit judgment, start-up confirmation segment truncation, and power recovery takeover judgment on each candidate sampling point. The candidate sampling point that first simultaneously meets both the descent exit condition and the power recovery takeover condition is written as the power-down retention window start-up R103. For example... Figure 3 As shown, although the earlier candidate point P1 has entered the candidate sampling point content for the starting point search, it does not simultaneously meet the descent exit condition and the power takeover condition, and therefore is not written as R103; candidate point P2 is the first candidate sampling point along the starting point search direction to simultaneously meet the descent exit condition and the power takeover condition, and is therefore written as the window starting point R103. Thus, Figure 3 P1 and P2 in the diagram correspond to the unlocked candidate starting point and the window starting point R103, respectively.

[0059] During the window endpoint determination phase, the window endpoint determination unit reads the candidate sampling point content for endpoint search from R102, and uses the voltage corresponding to R103 as the unified starting point for calculating the cumulative voltage drop. It calculates the cumulative voltage drop for each endpoint candidate sampling point, determines whether it meets the allowable voltage drop depletion condition, and simultaneously determines whether it meets the power sustaining exit condition. The candidate sampling point that first simultaneously meets both the allowable voltage drop depletion condition and the power sustaining exit condition is written as the power-down sustaining window endpoint R104. For example... Figure 3 As shown, candidate point Pn is the first candidate sampling point that simultaneously satisfies the allowed fall depletion condition and the continued energy maintenance exit condition along the endpoint search direction, and is therefore written as the window endpoint R104. Figure 3 P2 in the lower right corner corresponds to the window start point R103. The window end point determination unit uses the voltage corresponding to R103 as the unified starting point for calculating the cumulative voltage drop, and locks the window end point R104 on this basis.

[0060] During the hold time generation phase, the hold time generation unit enters the unified timing reading and time interval calculation phase only when R103 and R104 have been effectively formed. It adopts the unified sampling timing consistent with the fixed short sampling period as the time comparison caliber, reads the window start sampling time corresponding to R103 and the window end sampling time corresponding to R104, subtracts the window start sampling time corresponding to the window end sampling time from the window start sampling time to form the time interval, and writes it as the power-down hold time R105. Figure 3 In This corresponds to the power-down hold-up time R105 generated by R103 and R104. The conversion of the number of sampling point intervals under the unified sampling timing is only used as a supplementary measure and does not change the result generation method that uses the above time difference as the main measure.

[0061] In this embodiment, as Figure 5As shown, the short-time sustaining branch in the test platform consists of a normally closed relay contact and a sustaining capacitor. The device parameters and theoretical sustaining capability of the sustaining branch can be verified in engineering according to the platform specifications. The sustaining capacitor capacity can meet the requirements. ,in, This indicates the capacitance of the rechargeable capacitor, measured in farads. This indicates the load current of the device under test in the current test scenario, expressed in amperes. Indicates the short-term sustainment duration of the target, in seconds; This indicates the voltage at the measured terminal when the continued power support begins. This indicates the minimum operating voltage boundary allowed for the tested device to maintain normal operation. For the theoretical sustainment duration, it can be calculated as follows: Calculate, where, Indicates the allowable drop amount, and ; This represents the load current during the power-up maintenance phase. In this embodiment, when the operating voltage of the device under test is 24 volts, the full-load operating current is 2 amps, the allowable voltage drop is 1.2 volts, and the power-up capacitor is 4700 microfarads, the theoretical power-up maintenance time is approximately 2.8 milliseconds. Combined with the measured time of approximately 3 milliseconds from the de-energization of the relay coil to the closure of the normally closed contact, it can be shown that the platform has the hardware foundation to provide short-term power-up support on the order of less than 10 milliseconds. It should be noted that the above formulas and parameter calculations are only used to illustrate the hardware capabilities of the power-up branch and the basis for parameter selection. The power-down retention time R105 is still generated by subtracting the sampling time corresponding to the start of the window from the sampling time corresponding to the end of the window under the unified sampling timing.

[0062] In some implementations, to facilitate engineering implementation, example settings may be provided for fixed short sampling period, reference release duration, release drift correction amount, corrected release reference duration, preset interruption threshold, preset maintenance threshold, preset number of continuous sampling points, preset allowable drop amount, and start search interval and end search interval.

[0063] In this embodiment, the fixed short sampling period is set to 20 microseconds. The sampling unit continuously samples the voltage at the measured terminal at 50kHz. Its function is to distinguish the voltage transition process after the relay release and the short-term changes near the window boundary. The reference release duration can be set to 3 milliseconds, corresponding to the measured release time from the de-energization of the relay coil to the closure of the normally closed contact. Its function is to provide a basic coverage length for the start-point confirmation segment. The release drift correction can be set to 0.2 milliseconds, which is used to cover small release time offsets caused by historical cycles, temperature rise, and device aging, without excessively widening the start-point confirmation segment. Accordingly, the corrected release reference duration can be set to 3.2 milliseconds.

[0064] The preset allowable voltage drop can be set to 1.2 volts for example. This value corresponds to the engineering specification of a 24-volt device under test with a 5% allowable voltage drop, and is used to limit the upper limit of the cumulative voltage drop allowed from the start of the window. The preset number of continuous sampling points can be set to 20 for example, corresponding to a continuous observation length of approximately 0.4 milliseconds, to avoid triggering the power supply takeover judgment or power supply maintenance exit judgment by a single abnormal sampling point.

[0065] In this embodiment, both the preset interruption threshold and the preset sustaining threshold can be set to 0.02 volts to limit the allowable upper limit of the voltage drop variation between adjacent sampling points. This value is chosen because, under the example conditions of 24 volts, 2 amps, 4700 microfarads, and a 20 microsecond sampling period, the average voltage drop per sampling interval, calculated based on the theoretical sustaining process, is approximately 8.6 millivolts. Setting the threshold to 20 millivolts allows for the suppression of interference from single-point noise and short-term glitches on the judgment results while preserving normal sustaining fluctuations. The preset starting search interval can be set to 0 to 6 milliseconds after the relay coil drive signal is removed, because this interval covers approximately 3 milliseconds of the relay release process and the subsequent initial stage of sustaining control. The preset ending search interval can be set to 0 to 8 milliseconds after the window start point, because this interval covers approximately 2.8 milliseconds of the theoretical sustaining duration and the subsequent exit observation margin. In the above values, 20 microseconds corresponds to the sampling period setting of the sampling unit in this embodiment, 3 milliseconds corresponds to the measured release time from the de-energization of the relay coil to the closure of the normally closed contact in this embodiment, and 1.2 volts corresponds to the engineering setting of the 24-volt device under test determined according to the 5% allowable drop range; 0.2 milliseconds, 20, 0.02 volts, 0 to 6 milliseconds, and 0 to 8 milliseconds are suggested example values ​​of this embodiment, used to illustrate the setting method of each judgment parameter and search interval. They can be adjusted later according to the platform noise level, historical calibration results, and the allowable drop range of the device under test.

[0066] Through the above processing, this embodiment completes the determination and result generation of the power-down retention window on a given hardware platform according to the object chain of R101, R102, R103, R104, and R105. Among them, Figure 3 P1, P2, Pn and Each of these corresponds to an unlocked candidate starting point, window starting point R103, window ending point R104, and power-off retention time R105, respectively. Figure 4 Used to illustrate the platform connection basis that forms X101, X102 and X103; Figure 5 Used to explain the short-time power supply branch and the basis for its parameter selection; such as Figure 6As shown, the hardware protection unit in the test platform includes overvoltage, undervoltage, and overcurrent comparison branches, and is set independently of the MCU control unit, thus ensuring that the above window determination process is based on a protectable and repeatable test platform. Therefore, the formula explains the hardware's power sustaining capability, and the window boundaries explain the system result generation, ensuring that the generated power-down hold time falls within the effective hold interval after controlled descent exits and before power sustaining exits, thereby improving the statistical consistency and inter-round comparability of multi-round automatic power-up and power-down test results. It should be noted that... Figure 6 The overvoltage comparator, undervoltage comparator, and overcurrent comparator in the test device detect the overvoltage, undervoltage, and overcurrent states on the output path of the test device, respectively. When any comparison result reaches the corresponding trigger condition, the latching branch is triggered and the protection switch is turned off, thereby cutting off the output path of the test device.

[0067] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0068] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different systems to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0069] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0071] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A test system for automatic power-on / off testing, characterized in that, include: The operating condition establishment unit is used to establish a controlled slow-descent operating condition of the voltage at the measured terminal when the preconditions are met, and to trigger the cancellation relay coil drive signal to make the measured terminal enter the power-down holding window search process. The voltage sampling unit is used to extract the front-end reference segment and form the candidate sampling point content for the start-point search after the relay coil drive signal is canceled, and form the candidate sampling point content for the end-point search after the start-point of the power-down retention window is determined. The window start point determination unit is used as a comparison benchmark to form the descent exit condition based on the previous reference segment; For each candidate sampling point in the candidate sampling point content of the starting point search, first determine whether the candidate sampling point meets the slow descent exit condition, then take the candidate sampling point as the starting point and extract the starting point confirmation segment according to the modified release reference duration, and determine whether the start-up takeover condition is met within the starting point confirmation segment. The candidate sampling point that first simultaneously meets the conditions for slow descent exit and continued power takeover is determined as the starting point of the power failure retention window. The window endpoint determination unit is used to determine whether the cumulative voltage drop from the start of the power-down retention window to the candidate sampling point meets the allowable voltage drop depletion condition for each candidate sampling point formed with the start of the power-down retention window as the starting point, and then determine whether the candidate sampling point meets the continued power maintenance exit condition. The candidate sampling point that first simultaneously meets the allowable drop depletion condition and the continued energy maintenance exit condition is determined as the end point of the power failure maintenance window; The hold-up time generation unit is used to generate the power-down hold-up time based on the time interval between the start and end of the power-down hold-up window under a uniform sampling timing.

2. The automatic power-on / off testing experimental system according to claim 1, characterized in that, The operating condition establishment unit generates a relay status detection result by reading back the signal corresponding to the relay contact status, and generates a power supply capability detection result by performing a power supply capability detection on the power supply capacitor after precharging. The detection quantity of the power supply capability detection includes at least the voltage across the power supply capacitor or a power supply support capability characterization quantity based on the short-time discharge response after precharging. When the relay status detection result meets the current round of test switching requirements and the detection quantity falls within the preset available range, the precondition is determined to be met.

3. The automatic power-on / off testing experimental system according to claim 2, characterized in that, The operating condition establishment unit establishes a controlled descent operating condition for the measured terminal voltage at a fixed descent slope or a segmented descent slope; the target extraction interval is a predetermined interval reserved before the lowest operating voltage boundary and used to cover the last complete descent control cycle; when the measured terminal voltage enters the target extraction interval, the deactivation relay coil drive signal is triggered.

4. The automatic power-on / off testing experimental system according to claim 1, characterized in that, The voltage sampling unit extracts the voltage content of the measured terminal from the last complete descent control cycle immediately adjacent to the lowest operating voltage boundary to form the front reference segment; the complete descent control cycle is a single continuous descent control cycle with complete start and end boundaries during the controlled descent process and which is not cut off by the revoked relay coil drive action.

5. The automatic power-on / off testing experimental system according to claim 4, characterized in that, In the starting search phase after the relay coil drive signal is removed, the voltage sampling unit continuously reads the voltage of the measured terminal using the same fixed short sampling period to form candidate sampling points for the starting search; and in the ending search phase after the starting point of the power-down retention window is determined, it continuously reads the voltage of the measured terminal using the same fixed short sampling period to form candidate sampling points for the ending search. The candidate sampling points for the starting search are formed from the moment the relay coil drive signal is removed, and the candidate sampling points for the ending search continue to be formed from the moment the starting point of the power-down retention window is determined.

6. The automatic power-on / off testing experimental system according to claim 1, characterized in that, The window start point determination unit forms a natural fluctuation band based on the statistical upper and lower bounds of the voltage drop of each adjacent sampling point in the previous reference segment; when there are obvious abnormal sampling points, the obvious abnormal sampling points are removed first, and then the natural fluctuation band is formed. For each candidate sampling point, the local descent rate, characterized by the voltage change between the current candidate sampling point and the previous sampling point, is calculated, and the descent exit condition is determined based on whether the local descent rate deviates from the natural fluctuation band.

7. The automatic power-on / off testing experimental system according to claim 6, characterized in that, The window start point determination unit extracts a start point confirmation segment from the candidate sampling point that meets the slow descent exit condition, based on the candidate sampling point and extending backward according to the corrected release reference duration. The corrected release reference duration is jointly determined by the baseline release reference duration and the release drift correction amount. The release drift correction amount is determined based on the average or median offset of the actual release duration of each relay in the preset historical rounds relative to the baseline release reference duration. When a preset number of consecutive sampling points in the start point confirmation segment continue to descend in the same direction, and the descent amount of adjacent sampling points does not show a reverse jump, and the change amplitude between the descent amounts of adjacent sampling points does not exceed the preset interruption threshold, the start point confirmation segment is determined to meet the power takeover condition.

8. The automatic power-on / off testing experimental system according to claim 1, characterized in that, The window endpoint determination unit uses the voltage corresponding to the start point of the power-down retention window as a unified calculation benchmark. For each candidate sampling point, the difference between the unified calculation benchmark voltage and the voltage corresponding to the current candidate sampling point is used to form a cumulative voltage drop, and the cumulative voltage drop is used to determine whether the allowable voltage drop depletion condition is met. The preset allowable drop is given by the allowable difference between the minimum operating voltage boundary and the voltage corresponding to the start of the power-down retention window, or it can be preset according to the allowable drop range in the test standard.

9. The automatic power-on / off testing experimental system according to claim 8, characterized in that, The window endpoint determination unit reads a preset number of consecutive sampling points after the current candidate sampling point; the window endpoint determination unit adopts the same continuous observation length as the window start point determination unit and determines the preset number of consecutive sampling points based on the correspondence between the continuous observation length and the fixed short sampling period. When the decrease in the amount of adjacent sampling points changes in the opposite direction among a preset number of consecutive sampling points after the current candidate sampling point, or when the change in the decrease in the amount of adjacent sampling points exceeds a preset maintenance threshold, the current candidate sampling point is determined to meet the conditions for continued maintenance and exit. The candidate sampling point that first simultaneously meets the allowable drop depletion condition and the continued power maintenance exit condition is determined as the end point of the power-down retention window.

10. The automatic power-on / off testing experimental system according to claim 1, characterized in that, The hold time generation unit reads the sampling time corresponding to the start of the power-down hold window and the sampling time corresponding to the end of the power-down hold window under the unified sampling timing, and generates the power-down hold time by subtracting the sampling time corresponding to the start of the window from the sampling time corresponding to the end of the window.

Images