A data center machine room power supply test system and a test method thereof

Abstract

The application relates to the technical field of fault testing, in particular to a data center machine room power supply testing system and a testing method thereof. The method periodically acquires electric signals and rated electric signals; determines current characteristic parameters and voltage characteristic parameters according to the difference relationship between the fundamental wave electric signals and the rated electric signals; determines the fault probability value in each cycle according to the phase difference relationship between the extreme value data of the electric signals in each cycle and the extreme value data of the fundamental wave electric signals, the correlation relationship between the current characteristic parameters and the voltage characteristic parameters; and determines the total fault probability value according to the correlation relationship between the fault probability value and the current change characteristic parameters. Therefore, the possibility of the UPS existing fault problems can be accurately tested by considering whether the waveform distortion occurs in the electric signals and whether the change degree of the electric signals in each cycle has synchronization in time.

Description

Technical Field

[0001] This invention relates to the field of fault testing technology, and specifically to a data center power supply testing system and its testing method. Background Technology

[0002] Data center server rooms are used to store massive amounts of data and are a reliable part of modern information technology infrastructure and an important node in network communication. The load devices connected to data center server rooms have extremely high requirements for the reliability of power supply. In actual working scenarios, if the mains power fails, in order to avoid the loss of data from load devices and the resulting business interruption, uninterruptible power supply (UPS) is an indispensable device for data center server rooms.

[0003] Currently, to ensure the stability of power supply to data center computer rooms, UPS fault tests are conducted. In related technologies, UPS fault testing mainly relies on the presence of ripple or harmonics in the UPS's electrical signals to determine whether the UPS is faulty. However, this method often overlooks the numerous electronic components in the UPS and testing equipment. These electronic components can also cause harmonic or ripple electrical signals. Furthermore, during the testing process, electrical signals are usually not continuously obtained. Therefore, even if a single electrical signal does not show harmonics or ripples, if the changes between different single signals are not synchronized, it can still indicate that the UPS is faulty, affecting the stability of power supply to the data center computer room. Summary of the Invention

[0004] To address the technical problems in existing technologies that fail to eliminate interference from harmonic or ripple electrical signals caused by electronic components, and neglect the synchronicity of changes between discontinuous electrical signals, leading to distortion of defined harmonic or ripple characteristics and making it difficult to accurately test for UPS malfunctions, the present invention aims to provide a data center computer room power supply testing system and its testing method. The specific technical solution adopted is as follows:

[0005] This invention proposes a power supply testing method for data center computer rooms, the method comprising:

[0006] Periodically acquire electrical signals from the UPS in the computer room and rated electrical signals from the power supply testing equipment. The UPS is connected to the preset power supply testing equipment. The electrical signals include fundamental current signals and fundamental voltage signals, and the rated electrical signals include rated current signals and rated voltage signals.

[0007] Based on the difference between the fundamental current signal and the rated current signal in each cycle, the current characteristic parameters are determined, and based on the difference between the fundamental voltage signal and the rated voltage signal in each cycle, the voltage characteristic parameters are determined.

[0008] Based on the correlation between current characteristic parameters, voltage characteristic parameters, and phase characteristic parameters in each cycle, the fault characteristic parameters in each cycle are determined. Among them, the phase characteristic parameters are determined based on the phase difference relationship between the extreme value data of the electrical signal and the extreme value data of the fundamental signal. Based on the correlation between current characteristic parameters, voltage characteristic parameters, and fault characteristic parameters in each cycle, the fault probability value in each cycle is determined.

[0009] The total failure probability of the UPS is determined based on the correlation between the failure probability value and the current change characteristic parameter in all cycles. The current change characteristic parameter is determined based on the correlation between the amplitude of the current signal and the current characteristic parameter in each cycle.

[0010] Furthermore, the process of determining the fault probability value includes:

[0011] The stability index of the UPS is determined based on the time-varying trends of current characteristic parameters, voltage characteristic parameters, and fault characteristic parameters over all cycles.

[0012] Based on the correlation between the stability index and the fault characteristic parameters in each cycle, the failure probability value of the UPS in each cycle is determined, wherein the stability index and the fault characteristic parameters are both positively correlated with the failure probability value.

[0013] Furthermore, the extreme value data contains two types of extreme values: one type is a maxima, and the other type is a minima. The process for determining the phase characteristic parameters includes:

[0014] According to the time sequence of each cycle, the proportion of the same type of current extreme values ​​in the current signal within the corresponding cycle is arranged to determine the current extreme value sequence of the UPS. The proportion of the same type of current extreme values ​​is determined based on the quotient of the extreme value data of the same type of current signal and the maximum amplitude of the current signal.

[0015] According to the time sequence of each cycle, the proportion of the same type of voltage extreme values ​​in the corresponding cycle is arranged to determine the voltage extreme value sequence of the UPS. The proportion of the same type of voltage extreme values ​​is determined based on the quotient of the extreme value data of the same type of voltage signal and the maximum amplitude of the voltage signal.

[0016] Based on the similarity between the current extreme value sequence and the voltage extreme value sequence under the same type of extreme value, a first target similarity value between the UPS current signal and the voltage signal is determined, wherein the voltage extreme value sequence has a first minimum translation scalar in the time dimension;

[0017] Based on the similarity between the fundamental current extreme value sequence and the fundamental voltage extreme value sequence under the same type of extreme value, a second target similarity value between the UPS fundamental current signal and the fundamental voltage signal is determined. The fundamental current extreme value sequence is determined based on the extreme value data of the fundamental current signal in each period, and the fundamental voltage extreme value sequence is determined based on the extreme value data of the fundamental voltage signal in each period. The fundamental voltage extreme value sequence has a second minimum translation scalar in the time dimension.

[0018] Based on the absolute difference between the first minimum translation scalar and the second minimum translation scalar, the correlation between the first target similarity value and the second target similarity value, the phase characteristic parameters in each period are determined. The first target similarity value and the second target similarity value are both positively correlated with the phase characteristic parameters, and the absolute difference between the first minimum translation scalar and the second minimum translation scalar is negatively correlated with the phase characteristic parameters.

[0019] Furthermore, the process of determining the stability index includes:

[0020] The current trend value of the UPS is determined based on the time-varying trend of the current characteristic parameters over at least two adjacent cycles.

[0021] The voltage trend value of the UPS is determined based on the time-varying trend of voltage characteristic parameters over at least two adjacent periods.

[0022] Based on the temporal variation trend of fault characteristic parameters over at least two adjacent periods, the fault trend value of the UPS is determined; based on the correlation between the current trend value, voltage trend value and fault trend value, the stability index of the UPS is determined, wherein the current trend value, voltage trend value and fault trend value are all positively correlated with the stability index.

[0023] Furthermore, the process of obtaining the current trend value, the voltage trend value, and the fault trend value includes:

[0024] Calculate the difference between the current characteristic parameters in at least two adjacent cycles to obtain the current change value; perform mean calculation on the current change value to determine the current trend value of the UPS;

[0025] Calculate the difference between voltage characteristic parameters in at least two adjacent cycles to obtain the voltage change value; perform mean calculation on the voltage change value to determine the voltage trend value of the UPS;

[0026] Calculate the difference between fault characteristic parameters in at least two adjacent cycles to obtain the fault change value; perform mean calculation on the fault change value to determine the fault trend value of the UPS.

[0027] Furthermore, the process of determining the current characteristic parameters includes:

[0028] For each period, Fourier decomposition is performed on the electrical signal within the period to determine multiple electrical component signals, including current component signals and voltage component signals.

[0029] Based on the ratio between the frequency of each current component signal and the frequency of the fundamental current signal, a first ratio of each current component signal is determined, wherein the frequency of the fundamental current signal is constrained to be no less than a preset third threshold, the frequency of the current component signal is positively correlated with the first ratio, and the frequency of the fundamental current signal is negatively correlated with the first ratio.

[0030] Perform a summation operation on the product between the first ratio of each current component signal and the amplitude of the current component signal to determine the first summation value;

[0031] The first difference is determined based on the absolute difference between the first array composed of the amplitude and frequency of the fundamental current signal and the second array composed of the amplitude and frequency of the rated current signal;

[0032] Based on the correlation between the first summation value and the first difference value, the current characteristic parameters within the period are determined, wherein both the first summation value and the first difference value are positively correlated with the current characteristic parameters.

[0033] Furthermore, the power supply testing methods for data center computer rooms also include:

[0034] In response to a fault test task for a UPS in a data center, the fault test task includes multiple cycles, with the initial rated current set on the power supply test equipment in the first cycle.

[0035] According to the time sequence of each cycle, the rated current of the preset power supply test equipment is increased one by one in the corresponding cycle.

[0036] If there is a target period in which the rated current increases to the preset current threshold, the target period will be used as the last period of the fault test task, and the fault test task for the data center UPS will be terminated.

[0037] Furthermore, the process for determining the current change characteristic parameters includes:

[0038] Linear fitting is performed on the linear relationship between the amplitude of the current signal and the corresponding current characteristic parameters in each cycle to determine the fitted current characteristic parameters under different current signal amplitudes in each cycle.

[0039] The linear correlation between the amplitude of the current signal and the corresponding current characteristic parameters in each cycle is used to determine the correlation coefficient in each cycle.

[0040] The second difference is determined based on the absolute difference between the current characteristic parameters in the first cycle and the corresponding fitted current characteristic parameters.

[0041] The third difference is determined based on the absolute difference between the current characteristic parameters in the last cycle and the corresponding fitted current characteristic parameters.

[0042] Based on the correlation coefficient, the second difference, and the third difference, the current change characteristic parameters of the UPS are determined, wherein the second difference and the correlation coefficient are both positively correlated with the current change characteristic parameters, and the third difference is negatively correlated with the current change characteristic parameters.

[0043] Furthermore, the process for determining the total failure probability value includes:

[0044] Calculate the average of the failure probability values ​​over all periods to determine the first average value;

[0045] Based on the correlation between the first average value and the current change characteristic parameter, the total failure probability value of the UPS is determined, wherein the first average value is positively correlated with the total failure probability value, and the current change characteristic parameter is negatively correlated with the total failure probability value.

[0046] This invention proposes a power supply testing system for data center computer rooms. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements a power supply testing method for data center computer rooms disclosed in an embodiment of this invention.

[0047] The present invention has the following beneficial effects:

[0048] This invention first periodically acquires the electrical signals of the UPS in the computer room and the rated electrical signals of the power supply testing equipment. By dividing the fault testing process into multiple suitable cycles, it facilitates subsequent monitoring of the synchronicity of changes in electrical signals within different cycles. This makes it easier to discover harmonic or ripple characteristics that are difficult to analyze from a single electrical signal. Then, based on the difference between the fundamental electrical signal and the rated electrical signal in each cycle, current characteristic parameters are determined. Using these current characteristic parameters, the effective value changes of the current in each cycle are initially analyzed, allowing for the assessment of the degree of distortion in the UPS current waveform within each cycle. Next, by analyzing the correlation between the current characteristic parameters, voltage characteristic parameters, and fault characteristic parameters in each cycle, the phase relationship between current and voltage is further analyzed; that is, the changing trend between the waveforms of the fundamental current and fundamental voltage is... Whether or not the signal frequency changes over time is maintained synchronously is crucial to avoid the impact of signal frequency changes at different times. Secondly, due to the increase in current, the effective value of the current increases. It is known that the increase in the effective value under different currents is a normal phenomenon. The change in the effective value of the current within each cycle cannot completely and accurately test whether noise characteristics are present. Therefore, based on the characteristic parameters of current change, we can analyze whether the degree of current distortion and the change between currents are synchronous in time. Thus, by using the total failure probability value, we can determine whether the UPS has no waveform distortion in each cycle, eliminate the interference of harmonic or ripple electrical signals generated by electronic components, and determine whether the degree of change between electrical signals in different cycles of the UPS is synchronous in time. This improves the authenticity and accuracy of UPS fault test results and accurately tests whether the UPS has a fault. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a data center power supply testing method according to an embodiment of the present invention.

[0051] Figure 2 An example diagram of a power supply testing device provided in one embodiment of the present invention;

[0052] Figure 3 This is a schematic diagram of the current characteristic parameter determination process provided in one embodiment of the present invention;

[0053] Figure 4This is a schematic diagram illustrating the process of determining the fault probability value according to an embodiment of the present invention;

[0054] Figure 5 This is a schematic diagram illustrating the process of determining current change characteristic parameters according to an embodiment of the present invention. Detailed Implementation

[0055] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a data center power supply testing system and its testing method according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, the nominal features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0057] The following description, in conjunction with the accompanying drawings, details the specific scheme of a data center power supply testing system and its testing method provided by the present invention.

[0058] Please see Figure 1 The diagram illustrates a flowchart of a data center power supply testing method according to an embodiment of the present invention, the method comprising:

[0059] S101: Periodically acquires the electrical signals of the UPS in the computer room and the rated electrical signals of the power supply test equipment.

[0060] It should be noted that the input terminal of the UPS is connected to the preset power supply test equipment, and the electronic load section of the power supply test equipment is connected to the output terminal of the UPS.

[0061] It should be noted that the specific interval of the cycle can be determined according to actual needs. This embodiment does not impose a specific limitation. For example, 1 minute can be used as the interval of the cycle.

[0062] It is important to understand that due to the special nature of data center computer rooms, uninterrupted and stable power supply is required. In order to avoid affecting normal business processes, the test time and scope can be planned in advance before performing fault testing on the power system of the data center computer room. The specific test time and scope can be determined by those skilled in the art based on the actual situation. Although the mains power supply equipment (usually the battery power supply) of the data center computer room is turned off during the fault testing, professional power testing equipment can be used to facilitate the collection of UPS electrical signals and to ensure the safety of the data center computer room.

[0063] For example, an example diagram of a power supply test device is shown below. Figure 2 As shown, the power supply test equipment provides a programmable AC power supply, which allows adjustment of electrical signal parameters such as voltage, current, and frequency. The electronic load section of the power supply test equipment can simulate different load conditions.

[0064] In this embodiment, to further understand the differences between the test content and electrical signal parameters included in each cycle, in response to the fault test task for the UPS of the data center computer room, wherein the fault test task includes multiple cycles, in the first cycle, the initial rated current of the preset power supply test equipment is set; according to the time sequence of each cycle, the rated current of the preset power supply test equipment is increased in the corresponding cycle; if there is a target cycle in which the set rated current increases to a preset current threshold, the target cycle is taken as the last cycle of the fault test task, and the fault test task for the UPS of the data center computer room ends.

[0065] It should be noted that the specific value of the preset current threshold is determined according to actual needs, and this embodiment does not impose a specific limitation. For example, the preset current threshold can be 120% of the rated current of the UPS when it is working, i.e., 264A.

[0066] S102: Determine the current characteristic parameters based on the difference between the fundamental current signal and the rated current signal in each cycle, and determine the voltage characteristic parameters based on the difference between the fundamental voltage signal and the rated voltage signal in each cycle.

[0067] It should be noted that electrical signals include fundamental current signals and fundamental voltage signals, and rated electrical signals include rated current signals and rated voltage signals.

[0068] It should be noted that the electrical signal carries waveform information, which includes at least amplitude and frequency. The rated electrical signal also carries waveform information to determine its accuracy. The methods for obtaining the waveform information carried by the electrical signal and the waveform information carried by the rated electrical signal are well known to those skilled in the art, and will not be elaborated upon in this embodiment.

[0069] It is important to understand that the main part of an electrical signal is the fundamental frequency signal. The fundamental frequency refers to the lowest frequency sine wave component in a complex periodic oscillation. The amplitude and frequency of the fundamental frequency waveform are fixed. If the fundamental frequency current signal and the rated current signal are not significantly different in amplitude and frequency, it can be concluded that the waveform of the fundamental frequency current signal matches the waveform of the rated current signal well. This indicates that the UPS is in a relatively normal state. Therefore, by analyzing the differences between the fundamental frequency current signal and the rated current signal, the state of the UPS within a single cycle can be preliminarily determined.

[0070] The process of determining current characteristic parameters is as follows: Figure 3 As shown, it includes:

[0071] S102-1: For each period, perform Fourier decomposition on the electrical signal within the period to determine multiple electrical component signals, including current component signals and voltage component signals.

[0072] It should be noted that the technique of performing Fourier decomposition on an electrical signal to determine multiple electrical component signals is a well-known technique in the art, and will not be described in detail in this embodiment.

[0073] S102-2: Determine the first ratio of each current component signal based on the ratio between the frequency of each current component signal and the frequency of the fundamental current signal.

[0074] It should be noted that a non-sinusoidal periodic current signal is expanded into a linear superposition of multiple sinusoidal components according to the Fourier series. The lowest frequency component among these sinusoidal components has the same frequency as the non-sinusoidal periodic current signal and is called the fundamental current. It can be seen that during the test, the power supply test equipment continuously supplies power to the UPS, so the UPS will have an electrical signal. The main part of the electrical signal is the fundamental signal, so the frequency of the fundamental current signal will not be 0.

[0075] S102-3: Perform a summation operation on the product between the first ratio of each current component signal and the amplitude of the current component signal to determine the first summation value.

[0076] Since current components other than the fundamental current also affect the UPS current signal, such as the current signal generated by electronic components, if the amplitude and frequency of a certain current component signal are small, it means that the current component signal has less influence on the fundamental current. Therefore, the fundamental current signal can better represent the UPS current signal, and the degree of UPS waveform distortion is not high. However, if the frequency of the current component signal other than the fundamental current is higher, the influence on the distortion of the UPS current signal waveform will be more severe (obvious multi-peaks will appear). Moreover, if the current component signal also has a large amplitude, the current signal waveform can show obvious concave and convex changes, thus the degree of waveform distortion will be more severe. Therefore, the first summation value can be expressed by the following formula:

[0077]

[0078] Among them, f i f1 represents the frequency of the i-th current component signal; f1 represents the frequency of the fundamental current signal. A represents the first ratio of the i-th current component signal; i This represents the amplitude of the i-th current component signal; This represents the first summation value.

[0079] S102-4: Determine the first difference based on the absolute difference between the first array composed of the amplitude and frequency of the fundamental current signal and the second array composed of the amplitude and frequency of the rated current signal.

[0080] S102-5: Determine the current characteristic parameters within the period based on the correlation between the first summation value and the first difference value, wherein both the first summation value and the first difference value are positively correlated with the current characteristic parameters.

[0081] Since the main component of the electrical signal is the fundamental frequency signal, and the rated current signal comes from the power supply testing equipment and reflects the UPS current signal, a significant difference between the fundamental frequency current signal and the rated current signal indicates a large difference between the UPS current waveform and the fundamental frequency current waveform. Furthermore, since current component signals are used to indicate currents of different frequencies, if each current component signal has a higher frequency and larger amplitude relative to the fundamental frequency signal, these situations will indicate distortion phenomena such as spikes and dips in the UPS current signal waveform. This prevents the UPS current output from being restored to a sine wave, resulting in waveform distortion. In other words, the larger the current characteristic parameter, the greater the current characteristic parameter can be expressed by the following formula:

[0082]

[0083] Where a(I) represents the current characteristic parameter; || represents taking the absolute value of the array; ([A1,f1]-[Ao ,f o [A1,f1] represents the first difference; [A1,f1] represents the first array; A1 represents the amplitude of the fundamental current signal; f1 represents the frequency of the fundamental current signal; [A1,f1] represents the amplitude of the fundamental current signal; [A1,f1] ... amplitude of the fundamental current signal; [A1,f1] represents the amplitude of the fundamental current signal; [ o ,f o ] represents the second array; A o Indicates the amplitude of the rated current signal; f o The frequency of the rated current signal; f i This represents the frequency of the i-th current component signal; A represents the first ratio of the i-th current component signal; i This represents the amplitude of the i-th current component signal; This represents the first summation value.

[0084] It should be noted that the process of determining voltage characteristic parameters is similar to that of determining current characteristic parameters. Therefore, the process of determining voltage characteristic parameters can refer to the process of determining current characteristic parameters, and will not be repeated in this embodiment.

[0085] Since the main component of the electrical signal is the fundamental frequency signal, and the rated voltage signal comes from the power supply testing equipment, reflecting the UPS voltage signal, a significant difference between the fundamental and rated voltage signals indicates a large difference between the UPS voltage waveform and the fundamental voltage waveform. Furthermore, voltage component signals are used to indicate voltages at different frequencies. If each voltage component signal has a higher frequency and larger amplitude relative to the fundamental voltage signal, these conditions will indicate distortion phenomena such as spikes and dips in the UPS voltage signal waveform. This prevents the UPS voltage output from being restored to a sine wave, resulting in waveform distortion. In other words, a larger voltage characteristic parameter indicates a larger voltage characteristic parameter. Therefore, the voltage characteristic parameter can be expressed by the following formula:

[0086]

[0087] Where a(U) represents the voltage characteristic parameter; || represents taking the absolute value of the array; A2 represents the amplitude of the fundamental voltage signal; f2 represents the frequency of the fundamental voltage signal; A u Indicates the amplitude of the rated voltage signal; f u The frequency of the rated voltage signal; f j A represents the frequency of the j-th voltage component signal; j This represents the amplitude of the j-th voltage component signal.

[0088] S103: Based on the correlation between the current characteristic parameters, voltage characteristic parameters and phase characteristic parameters in each cycle, determine the fault characteristic parameters in each cycle. The phase characteristic parameters are determined based on the phase difference between the extreme value data of the electrical signal and the extreme value data of the fundamental signal. Based on the correlation between the current characteristic parameters, voltage characteristic parameters and fault characteristic parameters in each cycle, determine the fault probability value in each cycle.

[0089] It is important to understand that current and voltage characteristic parameters only reveal the waveform characteristics of different frequencies in an electrical signal. However, they cannot directly provide information about the phase relationship between current and voltage. In actual working scenarios, even if the current signal itself does not have obvious harmonic distortion (the Fourier decomposition results are normal), changes in the phase relationship between current and voltage may indicate a potential UPS fault. Therefore, simply using the current and voltage characteristic parameters determined within each cycle is insufficient to accurately reflect the degree of waveform distortion within the cycle, and thus cannot fully test whether the UPS has a fault.

[0090] It is important to understand that since data center server rooms need to receive a stable AC power supply, the mismatch between current and voltage phases can cause abnormal current or voltage in the UPS. Therefore, in order to ensure the stability of the UPS current and voltage output at every moment and to fully test all possible fault conditions of the UPS, it is also necessary to pay attention to the phase matching between current and voltage during fault testing.

[0091] It should be noted that there are two types of extreme values: one is the maximum value and the other is the minimum value.

[0092] It should be noted that the method for obtaining extreme value data of electrical signals is a well-known technique in the art, and will not be described in detail in this embodiment.

[0093] To accurately determine the phase characteristic parameters, one possible implementation is to determine the phase characteristic parameters based on the phase difference relationship between the extreme value data of the electrical signal and the extreme value data of the fundamental signal.

[0094] As an example, the current extreme value sequence of the UPS is determined by arranging the proportion of similar current extreme values ​​within the corresponding period according to the time sequence of each cycle; the voltage extreme value sequence of the UPS is determined by arranging the proportion of similar voltage extreme values ​​within the corresponding period according to the time sequence of each cycle; the first target similarity value between the UPS current signal and voltage signal is determined based on the similarity between the current extreme value sequence and the voltage extreme value sequence under the same type of extreme value, wherein the voltage extreme value sequence has a first minimum shift scalar in the time dimension; the fundamental current signal and fundamental voltage extreme value sequence of the UPS are determined based on the similarity between the fundamental current extreme value sequence and the fundamental voltage extreme value sequence under the same type of extreme value. The second target similarity value between voltage signals is determined by the fundamental current extreme value sequence based on the extreme value data of the fundamental current signal in each period, and the fundamental voltage extreme value sequence based on the extreme value data of the fundamental voltage signal in each period. The fundamental voltage extreme value sequence has a second minimum translation scalar in the time dimension. The phase characteristic parameters in each period are determined according to the absolute difference between the first minimum translation scalar and the second minimum translation scalar, the correlation between the first target similarity value and the second target similarity value. The first target similarity value and the second target similarity value are both positively correlated with the phase characteristic parameters, and the absolute difference between the first minimum translation scalar and the second minimum translation scalar is negatively correlated with the phase characteristic parameters.

[0095] Since extreme value data contains two types of extreme values, and the differences between these different types are significant, focusing on the changes in different types of extreme values ​​is not meaningful. Therefore, we can analyze the similarity between current and voltage under the same type of extreme value. That is, current extreme value sequences can be divided into current maximum and current minimum sequences, voltage extreme value sequences into voltage maximum and voltage minimum sequences, fundamental current extreme value sequences into fundamental current maximum and fundamental current minimum sequences, and fundamental voltage extreme value sequences into fundamental voltage maximum and fundamental voltage minimum sequences. Analyzing the similarity between current and voltage under the same type of extreme values ​​can be divided into the following two cases: one case is analyzing the similarity between the current maximum value sequence and the voltage maximum value sequence; the other case is analyzing the similarity between the current minimum value sequence and the voltage minimum value sequence. Similarly, analyzing the similarity between the fundamental current extreme value sequence and the fundamental voltage extreme value sequence under the same type of extreme values ​​can be divided into the following two cases: one case is analyzing the similarity between the fundamental current maximum value sequence and the fundamental voltage maximum value sequence; the other case is analyzing the similarity between the fundamental current minimum value sequence and the fundamental voltage minimum value sequence.

[0096] It should be noted that the proportion of extreme values ​​of the same type of current is determined based on the ratio of the extreme value data of the same type of current signal to the maximum amplitude of the current signal. It can be seen that whether the electrical signal is negative is closely related to the direction of the electrical signal. If the UPS supplies power to the power supply test equipment, the electrical signal may be negative. However, in the test process, the power supply test equipment supplies power to the UPS. The direction of the current signal indicates that the maximum amplitude of the current signal will not be negative. Therefore, even if it is the current minimum value data, the maximum amplitude of the current signal will not affect the trend of the current minimum value sequence. Thus, the current maximum value sequence can be represented as MI1, and the current minimum value sequence can be represented as MI2.

[0097] It should be noted that the proportion of voltage extreme values ​​of the same type is determined based on the ratio of the extreme value data of the same type of voltage signal to the maximum amplitude of the voltage signal. During the test, the power supply test equipment supplies power to the UPS. The direction of the voltage signal indicates that the maximum amplitude of the voltage signal will not be negative. Therefore, even if it is the voltage minimum value data, the maximum amplitude of the voltage signal will not affect the changing trend of the voltage minimum value sequence. Thus, the voltage maximum value sequence can be represented as MU1, and the voltage minimum value sequence can be represented as MU2.

[0098] Since the current extreme value sequence and the voltage extreme value sequence are different time series, there may be inconsistencies between the two sequences in time. However, only sequences that are consistent in time can better reflect the time delay of the real current and voltage signals. Therefore, the voltage extreme value sequence can be shifted in the time dimension to make the current extreme value sequence and the voltage extreme value sequence consistent in time.

[0099] It should be noted that there is a preset shift time threshold for shifting the voltage extreme value sequence in the time dimension. The specific value of the preset shift time threshold is determined according to the actual situation, and this embodiment does not impose a specific limitation. For example, the preset shift time threshold can be set to a value of Among them, f o Indicates the frequency of the rated current signal.

[0100] The first objective similarity value is used to represent the similarity between current extreme value sequences and voltage extreme value sequences under the same type of extreme values.

[0101] The first minimum shift scalar is used to represent the minimum shift time interval of the voltage extremum sequence in the time dimension.

[0102] The second objective similarity value is used to represent the similarity between the fundamental current extreme value sequence and the fundamental voltage extreme value sequence under the same type of extreme value.

[0103] The second minimum translation scalar is used to represent the minimum translation time interval of the fundamental voltage extremum sequence in the time dimension.

[0104] It should be noted that the current extreme value sequence and the voltage extreme value sequence are different time series. In order to keep the two sequences consistent in time, the voltage extreme value sequence can be shifted by the first minimum shift scalar in the time dimension. Similarly, the fundamental current extreme value sequence and the fundamental voltage extreme value sequence are different time series. In order to keep the two sequences consistent in time, the fundamental voltage extreme value sequence can be shifted by the second minimum shift scalar in the time dimension.

[0105] Since the similarity between current and voltage extreme value sequences can be analyzed using Dynamic Time Warping (dtw), a larger distance indicates a greater difference between the two sequences, suggesting a higher first target similarity value between the current and voltage signals. Furthermore, the more consistent the voltage and current extreme value sequences are in time before time shifting, the shorter the subsequent time shift period for the voltage extreme value sequence will be. Therefore, the first target similarity value and the first minimum shift scalar can be expressed by the following formula:

[0106]

[0107] Where s1 represents the first target similarity value; norm represents the normalization function; δd1(δt1)=dtw(MU1(δt1),MI1); δd1(δt1) represents the similarity between the current maximum sequence MI1 and the voltage maximum sequence MU1; δd2(δt1)=dtw(MU2(δt1),MI2); δd2(δt1) represents the similarity between the current minimum sequence MI2 and the voltage maximum sequence MU2; dtw represents the dynamic time warping algorithm; δt1 represents the minimum translation time interval of the voltage extreme sequence in the time dimension, i.e., the first minimum translation scalar; δt1=argmin|δd1(t)-δd2(t)|; argmin represents returning the minimum value.

[0108] It should be noted that the fundamental current extreme value sequence is determined based on the extreme value data of the fundamental current signal in each period, and the fundamental voltage extreme value sequence is determined based on the extreme value data of the fundamental voltage signal in each period. Since the method of obtaining the fundamental current extreme value sequence is the same as that of obtaining the current extreme value sequence, and the method of obtaining the fundamental voltage extreme value sequence is the same as that of obtaining the voltage extreme value sequence, the subsequent methods of determining the target similarity value and the minimum translation scalar are no different. Therefore, this embodiment only describes in detail the determination method of the first target similarity value and the first minimum translation scalar. The determination method of the second target similarity value and the second minimum translation scalar will not be described in this embodiment. The second target similarity value can be represented as s2, and the second minimum translation scalar can be represented as δt2.

[0109] Since the voltage extreme value sequence and the current extreme value sequence are highly similar, it indicates that the waveforms of the current and voltage should be synchronized in time, that is, the phases of the current and voltage should be relatively consistent. Therefore, the phase characteristic parameter can be expressed by the following formula:

[0110]

[0111] Where p represents the phase characteristic parameter; |δt1-δt2| represents the absolute difference between the first minimum translation scalar δt1 and the second minimum translation scalar δt2; || represents the absolute value of the difference between δt1 and δt2; s1 represents the first target similarity value; s2 represents the second target similarity value; and ε is the preset minimum positive offset.

[0112] It should be noted that by adding a very small positive offset ε to the denominator |δt1-δt2|, mathematically it is ensured that the denominator |δt1-δt2|+ε is strictly greater than zero, thereby completely eliminating the possibility of the denominator being zero in the division operation, ensuring the robustness and executability of the calculation formula. Moreover, ε is preset to a sufficiently small positive value. The principle for selecting its specific value is that under the normal working condition where |δt1-δt2|+ε is significantly greater than zero, the influence of ε on the calculation result p can be ignored. For example, the specific value of ε is determined according to actual needs, and ε can be 0.01. This embodiment does not impose specific limitations.

[0113] Since the phase changes between the UPS current and voltage are relatively consistent, and the waveforms of the current and voltage signals do not show obvious distortion, it indicates that the UPS does not have any significant fault problems within the cycle. Therefore, the fault characteristic parameters can be expressed by the following formula:

[0114]

[0115] Where b represents the fault characteristic parameter; norm represents the normalization function; a(U) represents the voltage characteristic parameter; and a(I) represents the current characteristic parameter.

[0116] In this embodiment, the fault characteristic parameters in each cycle are determined based on the correlation between the current characteristic parameters, voltage characteristic parameters, and phase characteristic parameters in each cycle. The current characteristic parameters and voltage characteristic parameters are positively correlated with the fault characteristic parameters, while the phase characteristic parameters are negatively correlated with the fault characteristic parameters.

[0117] The process of determining the failure probability value is as follows: Figure 4 As shown, it includes:

[0118] S103-1: Determine the UPS stability index based on the time-varying trends of current characteristic parameters, voltage characteristic parameters, and fault characteristic parameters over all cycles.

[0119] It is important to understand that after analyzing whether the current waveform and voltage waveform are distorted in each cycle, and whether the phase changes between current and voltage are consistent, since the fault test process is divided into multiple cycles, it is also possible to focus on whether the current waveform, voltage waveform, and phase changes between current and voltage are stable at different times, that is, to analyze whether the degree of distortion of the current waveform and the degree of distortion of the voltage waveform gradually worsens over time.

[0120] To accurately determine the stability index, in this embodiment, the current trend value of the UPS is determined based on the time-varying trend of the current characteristic parameters over at least two adjacent periods; the voltage trend value of the UPS is determined based on the time-varying trend of the voltage characteristic parameters over at least two adjacent periods; the fault trend value of the UPS is determined based on the time-varying trend of the fault characteristic parameters over at least two adjacent periods; and the stability index of the UPS is determined according to the correlation between the current trend value, voltage trend value, and fault trend value, wherein the current trend value, voltage trend value, and fault trend value are all positively correlated with the stability index.

[0121] In this embodiment, the difference between current characteristic parameters within at least two adjacent cycles is calculated to obtain the current change value; the mean value of the current change value is calculated to determine the current trend value of the UPS; the difference between voltage characteristic parameters within at least two adjacent cycles is calculated to obtain the voltage change value; the mean value of the voltage change value is calculated to determine the voltage trend value of the UPS; the difference between fault characteristic parameters within at least two adjacent cycles is calculated to obtain the fault change value; the mean value of the fault change value is calculated to determine the fault trend value of the UPS.

[0122] Since a greater difference in current characteristic parameters between adjacent cycles indicates a greater trend in current characteristic parameter changes over time, and a greater difference in voltage characteristic parameters between adjacent cycles also indicates a greater trend in voltage characteristic parameter changes over time, and similarly, a greater difference in fault characteristic parameters between adjacent cycles indicates a greater trend in fault characteristic parameter changes over time, it should be noted that the specific determination methods for current trend values, voltage trend values, and fault trend values ​​all adopt the same time-series data trend determination method. Therefore, the current trend values, voltage trend values, and fault trend values ​​of a UPS can be expressed by the following general formula:

[0123] θ(x)=E(x j -x j-1 )

[0124] Where θ(x) represents the trend value; x j Let x represent the parameter in the j-th period; x j-1 The parameter x represents the period adjacent to the j-th period, i.e., the (j-1)-th period; E indicates that the mean is returned.

[0125] It can be understood that parameter x can be divided into three cases: current characteristic parameter, voltage characteristic parameter and fault characteristic parameter. Then θ(a(U)) can represent the voltage trend value; θ(a(I)) can represent the current trend value; and θ(b) can represent the fault trend value.

[0126] If the current characteristic parameters differ significantly between the two most recent adjacent cycles, it indicates that the distortion of the current waveform gradually worsens over time. Similarly, if the voltage characteristic parameters differ significantly between the two most recent adjacent cycles, it indicates that the distortion of the voltage waveform gradually worsens over time. Furthermore, if the fault characteristic parameters differ significantly between the two most recent adjacent cycles, it indicates that the phase relationship between current and voltage is severely out of balance, and the phase difference is increasing over time. All these situations demonstrate that the electrical signal is unstable at different points in time. Therefore, the stability index of a UPS can be expressed by the following formula:

[0127] d=norm([θ(a(U))+θ(a(I))]×θ(b))

[0128] Where d represents the stability index of the UPS; norm represents the normalization function; θ(a(U)) represents the voltage trend value; θ(a(I)) represents the current trend value; θ(b) represents the fault trend value; a(U) represents the voltage characteristic parameter; a(I) represents the current characteristic parameter; and b represents the fault characteristic parameter.

[0129] S103-2: Based on the correlation between stability indicators and fault characteristic parameters in each cycle, determine the failure probability value of the UPS in each cycle. Both stability indicators and fault characteristic parameters are positively correlated with the failure probability value.

[0130] Since a UPS should perform well, the waveforms of the current and voltage, as well as the phase changes between the current and voltage, should be synchronized in time. In other words, the smaller the stability index, the better. Conversely, if the UPS has a significant fault, there will be a noticeable time difference between the fundamental and rated electrical signal waveforms, and the current and voltage signals will exhibit significant distortion. Therefore, the fault probability value can be expressed by the following formula:

[0131] p = d × b

[0132] Where p represents the failure probability value; d represents the stability index; and b represents the failure characteristic parameter.

[0133] S104: Determine the total failure probability value of the UPS based on the correlation between the failure probability value and the current change characteristic parameter in all cycles. The current change characteristic parameter is determined based on the correlation between the amplitude of the current signal and the current characteristic parameter in each cycle.

[0134] It is important to understand that the fault testing process is divided into multiple cycles. In each different cycle, the rated current of the power supply testing equipment can be adjusted, and the fault probability value under different rated currents can be recorded in different cycles. However, it is known that in actual scenarios, the input current is not constant. Therefore, it is also necessary to pay attention to whether the UPS can maintain high power supply stability under different special currents.

[0135] The process of determining the characteristic parameters of current change is as follows: Figure 5 As shown, it includes:

[0136] S104-1: Perform linear fitting on the linear relationship between the amplitude of the current signal and the corresponding current characteristic parameters in each cycle to determine the fitted current characteristic parameters under different current signal amplitudes in each cycle.

[0137] It is important to understand that, theoretically, if a UPS is operating normally, the distortion of the current waveform should be synchronized with the change in current. In this case, there is a linear relationship between the amplitude of the current signal and the corresponding current characteristic parameters. However, if the UPS experiences a loss of synchronization under high current, it indicates a UPS malfunction. To address this loss of synchronization, we can analyze the degree to which the actual current characteristic parameters deviate from the fitted current characteristic parameters.

[0138] S104-2: Determine the correlation coefficient for each period based on the linear correlation between the amplitude of the current signal and the corresponding current characteristic parameters within each period.

[0139] It should be noted that the specific method for calculating the correlation coefficient between the amplitude of the current signal and the corresponding current characteristic parameter is a well-known technique in the art, and will not be described in detail in this embodiment. For example, the Pearson correlation coefficient between the amplitude of the current signal and the corresponding current characteristic parameter is calculated, and the calculated Pearson correlation coefficient is used as the correlation coefficient in each period.

[0140] S104-3: Determine the second difference based on the absolute difference between the current characteristic parameters in the first cycle and the corresponding fitted current characteristic parameters.

[0141] S104-4: Determine the third difference based on the absolute difference between the current characteristic parameters in the last cycle and the corresponding fitted current characteristic parameters.

[0142] It is understandable that the initial rated current is set on the power supply test equipment in the first cycle, and the initial rated current is the minimum rated current in all cycles. It is known that the rated current of the power supply test equipment is increased in the corresponding cycle according to the time sequence of each cycle. Therefore, the rated current in the last cycle is the maximum rated current in all cycles.

[0143] S104-5: Based on the correlation coefficient, the second difference, and the third difference, determine the current change characteristic parameters of the UPS. The second difference and the correlation coefficient are positively correlated with the current change characteristic parameters, while the third difference is negatively correlated with the current change characteristic parameters.

[0144] Since the deviation of the actual current characteristic parameters from the fitted current characteristic parameters is small at both the minimum and maximum rated current, it indicates that the distortion of the current waveform is synchronized with the change in current. The distortion is less affected by an increase in current amplitude. If the correlation coefficient between current amplitude and current waveform distortion is larger at different rated currents, it indicates a positive correlation between the two, meaning the distortion is synchronized with the change in current. This synchronization is a characteristic of a normal UPS. Therefore, the current change characteristic parameters can be expressed by the following formula:

[0145]

[0146] Among them, e beginThis represents the second difference between the current characteristic parameter and the corresponding fitted current characteristic parameter at the minimum rated current; e end This represents the third difference between the current characteristic parameter and the corresponding fitted current characteristic parameter at the maximum rated current; R represents the correlation coefficient between the amplitude of the current signal and the corresponding current characteristic parameter; g represents the current change characteristic parameter; and norm represents the normalization function.

[0147] To intuitively understand whether a UPS is faulty, the total failure probability value of the UPS can be determined. In this embodiment, the failure probability value in all periods is averaged to determine the first average value. Based on the correlation between the first average value and the current change characteristic parameter, the total failure probability value of the UPS is determined. The first average value is positively correlated with the total failure probability value, and the current change characteristic parameter is negatively correlated with the total failure probability value.

[0148] Since, in a normal UPS, the distortion of the current waveform is synchronized with the change in current, and the probability of failure is small under different rated currents, the synchronization of the current waveform, voltage waveform, and phase change between current and voltage in time should be consistent. Therefore, the total probability of failure can be expressed by the following formula:

[0149]

[0150] Where P represents the total failure probability value; p represents the failure probability value in each cycle; g represents the current change characteristic parameter; and E(p) represents the mean of the failure probability values ​​in all cycles, i.e., the first average value.

[0151] In this embodiment, the presence of a UPS fault is determined based on the total fault probability value; if the total fault probability value exceeds a preset fault threshold, the UPS is determined to be faulty.

[0152] It should be noted that the specific value of the preset fault threshold is determined according to actual needs, and this embodiment does not impose a specific limitation. For example, if it is found that a UPS that is working normally and a UPS that is faulty can be well distinguished when the preset fault threshold is 0.8, then the preset fault threshold can be set to 0.8.

[0153] An embodiment of the present invention provides a data center power supply testing system, which includes: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement a data center power supply testing method disclosed in the embodiment of the present invention.

[0154] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the shown nominal or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0155] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for testing power supply of a data center room, characterized in that, The method comprises: periodically acquiring an electrical signal of a machine room UPS and a rated electrical signal of a power supply test device, wherein the UPS is connected with the preset power supply test device, wherein the electrical signal comprises a fundamental current signal and a fundamental voltage signal, and the rated electrical signal comprises a rated current signal and a rated voltage signal; determining a current characteristic parameter according to a difference relationship between the fundamental current signal and the rated current signal in each cycle, and determining a voltage characteristic parameter according to a difference relationship between the fundamental voltage signal and the rated voltage signal in each cycle; determining a fault characteristic parameter in each cycle according to a correlation relationship among the current characteristic parameter, the voltage characteristic parameter and a phase characteristic parameter in each cycle, wherein the phase characteristic parameter is determined based on a phase difference relationship between extreme value data of the electrical signal and extreme value data of the fundamental signal; determining a fault probability value in each cycle according to a correlation relationship among the current characteristic parameter, the voltage characteristic parameter and the fault characteristic parameter in each cycle; determining a total fault probability value of the UPS according to a correlation relationship between the fault probability values in all cycles and a current change characteristic parameter, wherein the current change characteristic parameter is determined based on a correlation relationship between the amplitude of the current signal and the current characteristic parameter in each cycle; The fault probability value determination process comprises: determining a stability index of the UPS according to a change trend of the current characteristic parameter, the voltage characteristic parameter and the fault characteristic parameter in all cycles over time; determining a fault probability value of the UPS in each cycle according to a correlation relationship between the stability index and the fault characteristic parameter in each cycle, wherein the stability index and the fault characteristic parameter are positively correlated with the fault probability value; The stability index determination process comprises: determining a current trend value of the UPS based on a change trend of the current characteristic parameter over time in at least two adjacent cycles; determining a voltage trend value of the UPS based on a change trend of the voltage characteristic parameter over time in at least two adjacent cycles; determining a fault trend value of the UPS based on a change trend of the fault characteristic parameter over time in at least two adjacent cycles; determining a stability index of the UPS according to a correlation relationship among the current trend value, the voltage trend value and the fault trend value, wherein the current trend value, the voltage trend value and the fault trend value are positively correlated with the stability index; The current trend value, the voltage trend value and the fault trend value obtaining process comprises: calculating a difference value between the current characteristic parameters in at least two adjacent cycles to obtain a current change value; performing mean value calculation on the current change value to obtain the current trend value of the UPS; calculating a difference value between the voltage characteristic parameters in at least two adjacent cycles to obtain a voltage change value; performing mean value calculation on the voltage change value to obtain the voltage trend value of the UPS; calculating a difference value between the fault characteristic parameters in at least two adjacent cycles to obtain a fault change value; performing mean value calculation on the fault change value to obtain the fault trend value of the UPS; The current characteristic parameter determination process comprises: For each cycle, the electrical signal in the cycle is subjected to Fourier decomposition to determine a plurality of electrical component signals, wherein the electrical component signals include current component signals and voltage component signals; A ratio between the frequency of each current component signal and the frequency of the fundamental current signal is calculated to obtain a first ratio of each current component signal; A sum operation is performed on the product between the first ratio of each current component signal and the amplitude of the current component signal to determine a first sum value; A first difference value is determined according to the absolute difference between a first array composed of the amplitude and frequency of the fundamental current signal and a second array composed of the amplitude and frequency of the rated current signal; A current characteristic parameter in the cycle is determined according to the correlation between the first sum value and the first difference value, wherein the first sum value and the first difference value are both positively correlated with the current characteristic parameter.

2. The data center power test method of claim 1, wherein, The extreme value data has two types of extreme values, one type is maximum value, and the other type is minimum value, and the phase characteristic parameter determination process comprises: According to the time sequence of each cycle, the same type current extreme value proportion in the corresponding cycle is arranged to determine the current extreme value sequence of the UPS, wherein the same type current extreme value proportion is determined based on the quotient of the same type extreme value data of the current signal and the maximum amplitude of the current signal; According to the time sequence of each cycle, the same type voltage extreme value proportion in the corresponding cycle is arranged to determine the voltage extreme value sequence of the UPS, wherein the same type voltage extreme value proportion is determined based on the quotient of the same type extreme value data of the voltage signal and the maximum amplitude of the voltage signal; A first target similarity value between the current signal and the voltage signal of the UPS is determined according to the similarity between the current extreme value sequence and the voltage extreme value sequence under the same type extreme value, wherein the voltage extreme value sequence has a first minimum translation scalar in the time dimension; A second target similarity value between the fundamental current signal and the fundamental voltage signal of the UPS is determined according to the similarity between the fundamental current extreme value sequence and the fundamental voltage extreme value sequence under the same type extreme value, wherein the fundamental current extreme value sequence is determined based on the extreme value data of the fundamental current signal in each cycle, the fundamental voltage extreme value sequence is determined based on the extreme value data of the fundamental voltage signal in each cycle, and the fundamental voltage extreme value sequence has a second minimum translation scalar in the time dimension; A phase characteristic parameter in each cycle is determined according to the correlation between the absolute difference between the first minimum translation scalar and the second minimum translation scalar, the first target similarity value and the second target similarity value, wherein the first target similarity value and the second target similarity value are both positively correlated with the phase characteristic parameter, and the absolute difference between the first minimum translation scalar and the second minimum translation scalar is negatively correlated with the phase characteristic parameter.

3. The method of claim 1, wherein, The method further comprises: In response to a fault test task for the data center machine room UPS, wherein the fault test task includes a plurality of cycles, an initial rated current is set on the preset power supply test device in the first cycle; According to the time sequence of each cycle, the rated current of the preset power supply test device is increased in the corresponding cycle one by one; If there is a set rated current increase to the target period of the preset current threshold, the target period is taken as the last period of the fault test task, and the fault test task for the data center machine room UPS is ended.

4. The data center power testing method of claim 3, wherein, The current change characteristic parameter determination process comprises: linear fitting the linear relationship between the amplitudes of the current signals in each period and the corresponding current characteristic parameters to determine the fitted current characteristic parameters at different amplitudes of the current signals in each period; determining the correlation coefficients in each period based on the linear correlation relationship between the amplitudes of the current signals and the corresponding current characteristic parameters; determining a second difference value based on the absolute difference value between the current characteristic parameter in the first period and the corresponding fitted current characteristic parameter; determining a third difference value based on the absolute difference value between the current characteristic parameter in the last period and the corresponding fitted current characteristic parameter; determining the current change characteristic parameter of the UPS according to the correlation relationship between the correlation coefficient, the second difference value and the third difference value, wherein the second difference value and the correlation coefficient are positively correlated with the current change characteristic parameter, and the third difference value is negatively correlated with the current change characteristic parameter.

5. The method of claim 1, wherein, The total failure probability value determination process comprises: performing mean value calculation on the failure probability values in all periods to determine a first average value; determining the total failure probability value of the UPS based on the correlation relationship between the first average value and the current change characteristic parameter, wherein the first average value is positively correlated with the total failure probability value, and the current change characteristic parameter is negatively correlated with the total failure probability value.

6. A data center room power testing system, characterized by, The system comprises a memory, a processor and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of the method in any one of claims 1-5 are implemented.

Images