Life prediction method and device of power output circuit, electronic equipment and medium

Abstract

The invention provides a life prediction method and device for a power output circuit, electronic equipment and a medium, and the method comprises the steps: determining a signal collection point according to a design data file of a target power output circuit; acquiring a first signal of the target power supply output circuit through the information acquisition point location; processing the first signal by adopting signal time-frequency transformation to obtain a second signal; performing feature extraction on the second signal to obtain a first feature of the target power supply output circuit; performing feature screening on the first feature to obtain a second feature, the second feature being used for representing the tendency of the first feature; and predicting the service life of the target power supply output circuit according to the second feature to obtain a prediction result. The beneficial effects of the invention are that life prediction of the power supply output circuit is convenient and more accordant with the expected life of the power supply output circuit.

Description

Technical Field

[0001] The present invention relates to the field of computer artificial intelligence technology, and in particular to a life prediction method, device, electronic equipment and medium for a power output circuit. Background Art

[0002] The switching power supply circuit is a key circuit in electronic boards. When key components such as aluminum electrolytic capacitors, MOSFETs, and inductors are degraded, it will generally lead to output voltage fluctuations, increased ripples, and other problems. The current electronic board life extraction generally selects the physical parameters of weak components, which has the disadvantages of complex and lossy measurement, difficult to evaluate on the board, and each different component requires a special measurement tool.

[0003] The existing technology also has the following problems: the collected circuit signals are nonlinear, non-stationary random signals with large data volume and high redundancy. It is difficult to extract effective life characteristic values ​​by directly analyzing the time domain waveform; the extracted multiple signal features are not all degradation life characteristics that change with time, and cannot all be used as life characteristics. As a result, there is a significant deviation in the predicted life. Summary of the invention

[0004] The main purpose of the embodiments of the present invention is to provide a method, device, electronic device and medium for predicting the life of a power output circuit, so that the life prediction of the power output circuit is convenient and more consistent with its expected life.

[0005] One aspect of the present invention provides a method for predicting the life of a power output circuit, characterized by comprising:

[0006] Determine the signal acquisition points according to the design data file of the target power output circuit;

[0007] Collecting a first signal of the target power output circuit through the information collection point, wherein the first signal is used to represent an original electrical signal of the target power output circuit;

[0008] Processing the first signal by using signal time-frequency transformation to obtain the second signal;

[0009] Perform feature extraction on the second signal to obtain a first feature of the target power output circuit, wherein the first feature includes at least one of energy, information entropy, and time-domain statistical indicators;

[0010] Performing feature screening on the first feature to obtain a second feature, where the second feature is used to characterize the trend of the first feature;

[0011] The life of the target power output circuit is predicted according to the second feature to obtain a prediction result.

[0012] According to the life prediction method of the power output circuit, the first signal of the target power output circuit is collected through the information collection point, including:

[0013] Determining sampling settings according to the design data file;

[0014] Acquiring the first signal by using an oscilloscope and the sampling setting using a minimum ground loop method;

[0015] The first signal is preprocessed, and the preprocessing includes deleting residual errors in the first signal using a 3σ criterion of a normal distribution.

[0016] According to the life prediction method of the power output circuit, the first signal is processed by using signal time-frequency transformation to obtain the second signal, including:

[0017] The signal time-frequency includes at least one of wavelet packet decomposition and empirical mode decomposition;

[0018] The wavelet packet decomposition is used to perform Fourier transform processing on the first signal to obtain the main peak frequency of the spectrum diagram; the wavelet packet decomposition layer number is determined according to the main peak frequency, the first signal is decomposed using the wavelet basis function according to the wavelet packet decomposition layer number to obtain a frequency band signal, and the main peak frequency is separated into different frequency band signals to obtain a reconstructed signal;

[0019] The empirical mode decomposition performs empirical mode decomposition on the first signal to obtain an intrinsic mode function signal, wherein the intrinsic mode function signal is used to characterize the frequency distribution of the first signal.

[0020] According to the life prediction method of the power output circuit, performing feature extraction on the second signal to obtain the first feature of the target power output circuit includes:

[0021] Perform segmented energy extraction on the reconstructed signal and the intrinsic mode function signal, and calculate the energy percentage, which is calculated as follows:

[0022]

[0023]

[0024] Among them, E j,n and E n Respectively represent the energy of each frequency band of the reconstructed signal and the energy of each intrinsic mode function signal, Ep j,n and Ep n represents the energy percentage of the frequency band signal and the intrinsic mode function signal, respectively, where j is the number of wavelet packet layers, n is the signal sequence number, Sj,n represents the reconstructed signal, x k represents the discrete point amplitude of the reconstructed signal, k is the discrete point number, N is the total number of discrete points, c n(i) is the nth intrinsic mode function signal component of the signal decomposition, and m is the number of intrinsic mode function signals;

[0025] The information entropy is calculated for the frequency band signal and the intrinsic mode function signal after wavelet packet decomposition respectively. The calculation method is:

[0026] S(X)=-∑p 2 (x)log p 2 (x), L(X) = -∑log p 2 (x),

[0027] Where S(X) is the information entropy of the frequency band signal after wavelet packet decomposition, L(X) is the information entropy of the intrinsic mode function signal, and p(x) is one of the energy of the frequency band signal or the energy of the intrinsic mode function signal;

[0028] The time domain statistical index is calculated for the frequency band signal where the main peak is located and the intrinsic mode function signal after wavelet packet decomposition, wherein the time domain statistical index includes at least one of mean, variance, root mean square value, kurtosis, skewness and kurtosis.

[0029] According to the life prediction method of the power output circuit, the first feature is subjected to feature screening to obtain the second feature, including:

[0030] Calculating the Spearman correlation coefficient of the first feature;

[0031] If the number of samples of the first feature is less than or equal to the preset sample value, the trend of the first feature is determined by comparing the Spearman correlation coefficient with the comparison table;

[0032] If the number of samples of the first feature is greater than the preset sample value, a check value T is generated according to the Spearman correlation coefficient.

[0033]

[0034] where ρxy is the Spearman correlation coefficient, and

[0035]

[0036] q is the number of samples, d is the level difference between the two variables, if the check value T is greater than the critical value of the comparison table, the two variables in the first feature have a trend, otherwise, there is no trend;

[0037] The first feature with trend is retained, and the first feature without trend is deleted to obtain the second feature.

[0038] According to the life prediction method of the power output circuit, the life of the target power output circuit is predicted according to the second feature to obtain a prediction result, including:

[0039] The target power output circuit is predicted based on the second feature and time to obtain a prediction result, and the prediction result is compared with a preset trend to determine the life of the target power output circuit.

[0040] According to the life prediction method of the power output circuit, the method further includes:

[0041] The target power output circuit is continuously detected, the second feature is updated, and the life of the target power output circuit is updated.

[0042] A second aspect of the present invention discloses a life prediction device for a power supply output circuit, comprising:

[0043] The first module is used to determine the signal collection points according to the design data file of the target power output circuit;

[0044] A second module is used to collect a first signal of the target power output circuit through the information collection point, where the first signal is used to represent an original electrical signal of the target power output circuit;

[0045] A third module is used to process the first signal by using signal time-frequency transformation to obtain the second signal;

[0046] A fourth module is used to perform feature extraction on the second signal to obtain a first feature of the target power output circuit, where the first feature includes at least one of energy, information entropy, and time-domain statistical indicators;

[0047] A fifth module is used to perform feature screening on the first feature to obtain a second feature, where the second feature is used to characterize the trend of the first feature;

[0048] The sixth module is used to predict the life of the target power output circuit according to the second feature to obtain a prediction result.

[0049] Another aspect of an embodiment of the present invention provides an electronic device, including a processor and a memory;

[0050] The memory is used to store programs;

[0051] The processor executes the program to implement the method described above.

[0052] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method described above.

[0053] The beneficial effects of the present invention are as follows: the original signal is decomposed by using the time-frequency transformation method, and then the part that can reflect the main information is selected, and the characteristic value is extracted by combining the signal energy, information entropy and time domain statistical indicators. The signal can be decomposed according to the frequency component, reducing the data dimension and redundancy. Extracting features on this basis can effectively mine the deep information of the circuit; Spearman correlation analysis is used for the extracted multiple electrical signal features to screen the electrical signal features with the degradation trend of the power output circuit, and the features without obvious trends can be eliminated, and the features with the degradation trend can be screened out as the life characteristics of the electronic board power output circuit, thereby improving the accuracy of the life of the electronic board power output circuit.

[0054] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0056] Figure 1 Schematic diagram of a life prediction system for a power output circuit according to an embodiment of the present invention.

[0057] Figure 2 It is a flow chart of a method for predicting the life of a power output circuit according to an embodiment of the present invention.

[0058] Figure 3 It is a schematic diagram of a signal acquisition process according to an embodiment of the present invention.

[0059] Figure 4 It is a schematic diagram of the wavelet packet decomposition process of an embodiment of the present invention.

[0060] Figure 5 Schematic diagram of wavelet packet decomposition according to an embodiment of the present invention.

[0061] Figure 6 It is a schematic diagram of the feature screening process of an embodiment of the present invention.

[0062] Figure 7 2 is a diagram of a life prediction and analysis device for a power supply output circuit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. In the subsequent description, the use of suffixes such as "module", "component" or "unit" used to represent elements is only for the purpose of facilitating the description of the present invention, and has no specific meaning in itself. Therefore, "module", "component" or "unit" can be used in a mixed manner. "First", "second" and the like are only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the order of the indicated technical features. In this subsequent description, the continuous numbering of the method steps is for the convenience of review and understanding. In combination with the overall technical solution of the present invention and the logical relationship between the various steps, adjusting the implementation order between the steps will not affect the technical effect achieved by the technical solution of the present invention. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0064] Terminology Note:

[0065] Time-frequency transform: Using time and frequency as variables to describe the signal, it can provide the joint distribution information of the signal in time and frequency, thereby calibrating the frequency and amplitude at any time, and has the ability to process non-stationary nonlinear signals.

[0066] Wavelet packet decomposition: It is improved based on wavelet decomposition and multi-resolution analysis. Its essence is to let the signal pass through a series of filters with different center frequencies but the same bandwidth. It can select the appropriate frequency band according to the characteristics of the signal, thereby improving the time-frequency resolution.

[0067] Empirical Mode Decomposition: Adaptively extract a set of intrinsic mode functions (IMF) with actual physical meaning from the signal based on the local characteristic structure of the signal on the time scale.

[0068] Information entropy: In the field of signal processing, it can characterize the different operating states of the system by the degree of change in the system's information probability distribution. When the system degrades, the information entropy will also change accordingly.

[0069] Spearman correlation analysis: a non-parametric indicator to measure the dependence between two variables, using a monotonic equation to evaluate the correlation between two variables.

[0070] like Figure 1As shown, an embodiment of the present invention provides a schematic diagram of a life prediction system for a power output circuit, which includes a target power output circuit 100, a collection device 200, a server 300 and a client 400, wherein the power output circuit 100, the collection device 200, the server 300 and the client 400 are connected in sequence, wherein the collection device 200 determines a signal collection point according to a design data file of the target power output circuit 100; the collection device 200 collects a first signal of the target power output circuit 100 through the information collection point, the first signal is used to characterize the original electrical signal of the target power output circuit 100, and the collection device 200 sends the original electrical signal to the server 300; the server 300 processes the first signal using a signal time-frequency transformation to obtain a second signal; the server 300 performs feature extraction on the second signal to obtain a first feature of the target power output circuit 100, the first feature including at least one of energy, information entropy and time-domain statistical indicators; the server 300 performs feature screening on the first feature to obtain a second feature, the second feature is used to characterize the trend of the first feature; the life of the target power output circuit 100 is predicted according to the second feature to obtain a prediction result.

[0071] In some embodiments, the first signal is an original signal collected by the collection device 200, and the second signal is a pre-processed signal.

[0072] In some embodiments, the acquisition device 200 is an oscilloscope.

[0073] like Figure 2 As shown, an embodiment of the present invention provides a process of a life prediction method for a power output circuit, and the method specifically includes but is not limited to steps S100-S600:

[0074] S100, determining signal collection points according to a design data file of a target power output circuit.

[0075] In some embodiments, the design data file may be a schematic diagram of a power output circuit, a PCB design file, or a BOM (product data structure) file.

[0076] S200, collecting a first signal of a target power output circuit through an information collection point, where the first signal is used to represent an original electrical signal of the target power output circuit.

[0077] In some embodiments, reference Figure 3 The signal acquisition process diagram shown includes but is not limited to steps S210 to S230:

[0078] S210, determining sampling settings according to the design data file;

[0079] S220, collecting the first signal by using an oscilloscope and a sampling setting and adopting a minimum ground loop method;

[0080] In some embodiments, the test environment and sampling settings of the oscilloscope can be determined according to the target power supply output circuit. Specifically, the oscilloscope adopts the minimum ground loop method for acquisition. For example, taking into account a certain accuracy and subsequent calculation speed, the sampling settings of the oscilloscope are bandwidth limit 20MHz, AC coupling, sampling rate 10MHz, sampling time 0.1s, and amplitude range 10mV / div.

[0081] S230, preprocessing the first signal, where the preprocessing includes deleting residual errors in the first signal using a 3σ criterion of a normal distribution.

[0082] In some embodiments, the probability that the residual error falls outside 3σ is approximately 0.26%. If a measured value of a residual error greater than 3σ is found, the point can be considered to be an outlier point and should be eliminated. The formula is u-3σ≤x≤u+3u, where σ is the standard deviation and u is the mean.

[0083] S300, processing the first signal by using signal time-frequency transformation to obtain a second signal.

[0084] In some embodiments, the time-frequency transform includes at least one of wavelet packet decomposition and empirical mode decomposition;

[0085] refer to Figure 4 The schematic diagram of the wavelet packet decomposition process shown in FIG. 1 includes but is not limited to steps S310 to S320:

[0086] S310, wavelet packet decomposition is used to perform Fourier transform processing on the first signal to obtain a main peak frequency of a frequency spectrum;

[0087] S320, determining the number of wavelet packet decomposition layers according to the main peak frequency, decomposing the first signal using wavelet basis functions according to the number of wavelet packet decomposition layers to obtain frequency band signals, and separating the main peak frequency into different frequency band signals to obtain a reconstructed signal.

[0088] For example, reference Figure 5 The wavelet packet decomposition schematic diagram shown includes: performing Fourier transform on the collected and preprocessed voltage ripple signal to identify the main peak frequency of the amplitude spectrum. Then, the wavelet packet decomposition layer number n and the wavelet basis function are determined, and the signal S is decomposed into 2^n frequency band signals through wavelet packet decomposition, and the main peak is separated into different frequency bands.

[0089] In some embodiments, empirical mode decomposition performs empirical mode decomposition on the first signal through empirical mode decomposition to obtain an intrinsic mode function signal, wherein the intrinsic mode function signal is used to characterize the frequency distribution (Intrinsic Mode Functions, IMF) of the first signal, and IMF is a frequency domain representation to display the frequency distribution of the signal.

[0090] S400, performing feature extraction on the second signal to obtain a first feature of a target power output circuit, wherein the first feature includes at least one of energy, information entropy, and a time-domain statistical index.

[0091] In some embodiments, the energy feature extraction includes segmented energy extraction of the reconstructed signal and the intrinsic mode function signal, and calculating the energy percentage, which is calculated as follows:

[0092]

[0093]

[0094] Among them, E j,n and E n Respectively represent the energy of each frequency band of the reconstructed signal and the energy of each intrinsic mode function signal, Ep j,n and Ep n They represent the energy percentages of the frequency band signal and the intrinsic mode function signal, respectively, where j is the number of wavelet packet layers, n is the signal number, and S j,n represents the reconstructed signal, x k represents the discrete point amplitude of the reconstructed signal, k is the discrete point number, N is the total number of discrete points, c n(i) is the nth intrinsic mode function signal component of the signal decomposition, and m is the number of intrinsic mode function signals;

[0095] The extraction of information entropy includes: calculating the information entropy of the frequency band signal and the intrinsic mode function signal after wavelet packet decomposition respectively, and the calculation method is:

[0096] S(X)=-∑p 2 (x)log p 2 (x), L(X) = -∑log p 2 (x),

[0097] Where S(X) is the information entropy of the frequency band signal after wavelet packet decomposition, L(X) is the information entropy of the intrinsic mode function signal, and p(x) is one of the energy of the frequency band signal or the energy of the intrinsic mode function signal;

[0098] The time domain statistical indicators include: calculating the time domain statistical indicators of the frequency band signal where the main peak is located and the intrinsic mode function signal after wavelet packet decomposition, wherein the time domain statistical indicators include at least one of the mean, variance, root mean square value, kurtosis, skewness and kurtosis.

[0099] For example, the mean, variance, root mean square value, kurtosis, skewness and kurtosis are calculated as follows:

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106] x is the signal amplitude, and n is the signal length. S500, feature screening is performed on the first feature to obtain a second feature, and the second feature is used to characterize the trend of the first feature.

[0107] In some embodiments, reference Figure 6 The feature screening process shown in FIG. 1 includes but is not limited to the steps

[0108] S510, calculating the Spearman correlation coefficient of the first feature;

[0109] S520, if the number of samples of the first feature is less than or equal to the preset value of the samples, the trend of the first feature is determined by comparing the Spearman correlation coefficient with the comparison table; if the number of samples of the first feature is greater than the preset value of the samples, a verification value T is generated according to the Spearman correlation coefficient, and the trend of the first feature is determined by the verification value T.

[0110] S530, retaining the first feature with a trend, deleting the first feature without a trend, and obtaining a second feature.

[0111] In some embodiments,

[0112]

[0113] where ρxy is the Spearman correlation coefficient, and

[0114]

[0115] q is the number of samples, d is the level difference between the two variables, if the check value T is greater than the critical value of the comparison table, the two variables in the first feature have a trend, otherwise, there is no trend;

[0116] For example, when the sample size is small (q≤30), ρ can be directly calculated as xy Value and compare with ρ table, only ρ xy A significant conclusion can be drawn only when the value is greater than or equal to the critical value in the table; when the sample size is large (q>30), it is necessary to construct the hypothesis H0: the two variables are related; H1: the two variables are not related. Construct the test value T, which should follow the t distribution t(n-2) with n-2 degrees of freedom. The critical value t can be obtained by looking up the table α / 2(n-2). When T>|t α / 2(n-2)|, H0 is rejected, the two variables are correlated, and the feature conforms to the trend; otherwise, H0 is accepted, and the feature has no trend.

[0117] S600, predicting the life of the target power output circuit according to the second feature to obtain a prediction result.

[0118] In some implementations, the target power output circuit is predicted using the second feature and time to obtain a prediction result, and the prediction result is compared with a preset trend to determine the life of the target power output circuit.

[0119] Exemplarily, when the trending life characteristics of the power output circuit are energy and time domain characteristics, and then predicted according to time and preset trends based on the trends of the energy and time domain characteristics, it can be understood that the preset trend is the entire power output circuit, which can be determined based on the model and performance of the power output circuit. When the sum of the trends of the energy and time domain characteristics reaches the preset trend time, it is the life time of the power output circuit.

[0120] The target power output circuit is continuously detected, the second feature is updated, and the life of the target power output circuit is updated.

[0121] In some embodiments, unified dynamic life prediction can also be performed on one or more power output circuits to achieve batch health trend monitoring.

[0122] Figure 7 1 is a diagram of a life prediction device for a power output circuit according to an embodiment of the present invention. The device comprises a first module 710 , a second module 720 , a third module 730 , a fourth module 740 , a fifth module 750 and a sixth module 760 .

[0123] Among them, the first module is used to determine the signal collection points according to the design data file of the target power output circuit; the second module is used to collect the first signal of the target power output circuit through the information collection points, and the first signal is used to characterize the original electrical signal of the target power output circuit; the third module is used to process the first signal using signal time-frequency transformation to obtain the second signal; the fourth module is used to perform feature extraction on the second signal to obtain the first feature of the target power output circuit, and the first feature includes at least one of energy, information entropy and time domain statistical indicators; the fifth module is used to perform feature screening on the first feature to obtain the second feature, and the second feature is used to characterize the trend of the first feature; the sixth module is used to predict the life of the target power output circuit according to the second feature to obtain a prediction result.

[0124] Exemplarily, under the cooperation of the first module, the second module, the third module, the fourth module, the fifth module and the sixth module in the device, the embodiment device can implement any of the aforementioned life prediction methods for the power output circuit, that is, determine the signal acquisition point according to the design data file of the target power output circuit;

[0125] The first signal of the target power output circuit is collected through the information collection point, and the first signal is used to characterize the original electrical signal of the target power output circuit; the first signal is processed by signal time-frequency transformation to obtain a second signal; the second signal is subjected to feature extraction to obtain a first feature of the target power output circuit, and the first feature includes at least one of energy, information entropy and time-domain statistical indicators; the first feature is subjected to feature screening to obtain a second feature, and the second feature is used to characterize the trend of the first feature; the life of the target power output circuit is predicted according to the second feature to obtain a prediction result. The beneficial effects of the present invention are as follows: the original signal is decomposed by the time-frequency transformation method, and then the part that can reflect the main information is selected, and the characteristic value is extracted from it in combination with the signal energy, information entropy and time-domain statistical indicators, and the signal can be decomposed according to the frequency component of the signal, reducing the data dimension and redundancy, and extracting features on this basis can effectively mine the deep information of the circuit; the Spearman correlation analysis is used for the extracted multiple electrical signal features, and the electrical signal features with the degradation trend of the power output circuit are screened, and the features without obvious trends can be eliminated, and the features with degradation trends are screened out as the life features of the electronic board power output circuit, thereby improving the accuracy of the life of the electronic board power output circuit.

[0126] An embodiment of the present invention further provides an electronic device, the electronic device comprising a processor and a memory;

[0127] The memory stores a program;

[0128] The processor executes the program to execute the aforementioned method for predicting the life of the power output circuit; the electronic device has the function of carrying and running the software system for predicting the life of the power output circuit provided in the embodiment of the present invention, for example, a computer, a speed regulator and a speed regulating device.

[0129] An embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the life prediction method of the power supply output circuit as described above.

[0130] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0131] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the life prediction method of the power output circuit.

[0132] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0133] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0134] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0135] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0136] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0137] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0138] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0139] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for predicting the life of a power output circuit, characterized in that: include: Determine the signal acquisition points according to the design data file of the target power output circuit; Collecting a first signal of the target power output circuit through the information collection point, wherein the first signal is used to represent an original electrical signal of the target power output circuit; Processing the first signal by using signal time-frequency transformation to obtain a second signal; Perform feature extraction on the second signal to obtain a first feature of the target power output circuit, wherein the first feature includes at least one of energy, information entropy, and time-domain statistical indicators; Performing feature screening on the first feature to obtain a second feature, where the second feature is used to characterize the trend of the first feature; The life of the target power output circuit is predicted according to the second feature to obtain a prediction result.

2. The life prediction method of a power output circuit according to claim 1, characterized in that: The collecting the first signal of the target power output circuit through the information collection point includes: Determining sampling settings according to the design data file; Acquiring the first signal by using an oscilloscope and the sampling setting using a minimum ground loop method; The first signal is preprocessed, and the preprocessing includes deleting residual errors in the first signal using a 3σ criterion of a normal distribution.

3. The life prediction method of a power output circuit according to claim 1, characterized in that: The step of processing the first signal by using a signal time-frequency transformation to obtain the second signal includes: The signal time-frequency includes at least one of wavelet packet decomposition and empirical mode decomposition; The wavelet packet decomposition is used to perform Fourier transform processing on the first signal to obtain the main peak frequency of the spectrum diagram; the wavelet packet decomposition layer number is determined according to the main peak frequency, the first signal is decomposed using the wavelet basis function according to the wavelet packet decomposition layer number to obtain a frequency band signal, and the main peak frequency is separated into different frequency band signals to obtain a reconstructed signal; The first signal is subjected to empirical mode decomposition through the empirical mode decomposition to obtain an intrinsic mode function signal, wherein the intrinsic mode function signal is used to characterize the frequency distribution of the first signal.

4. The life prediction method of a power output circuit according to claim 3, characterized in that: The performing feature extraction on the second signal to obtain a first feature of the target power output circuit includes: Perform segmented energy extraction on the reconstructed signal and the intrinsic mode function signal, and calculate the energy percentage, which is calculated as follows: Among them, E j，n and E n Respectively represent the energy of each frequency band of the reconstructed signal and the energy of each intrinsic mode function signal, Ep j，n and Ep n represents the energy percentage of the frequency band signal and the intrinsic mode function signal, respectively, where j is the number of wavelet packet layers, n is the signal sequence number, S j，n represents the reconstructed signal, x k represents the discrete point amplitude of the reconstructed signal, k is the discrete point number, N is the total number of discrete points, c n(i) is the nth intrinsic mode function signal component of the signal decomposition, and m is the number of intrinsic mode function signals; The information entropy is calculated for the frequency band signal and the intrinsic mode function signal after wavelet packet decomposition respectively. The calculation method is: S(X)=-∑p 2 (x)log p 2 (x),L(x)=-∑log p 2 (x), Where S(X) is the information entropy of the frequency band signal after wavelet packet decomposition, L(X) is the information entropy of the intrinsic mode function signal, and p(x) is one of the energy of the frequency band signal or the energy of the intrinsic mode function signal; The time domain statistical index is calculated for the frequency band signal where the main peak is located and the intrinsic mode function signal after wavelet packet decomposition, wherein the time domain statistical index includes at least one of mean, variance, root mean square value, kurtosis, skewness and kurtosis.

5. The life prediction method of a power output circuit according to claim 1, characterized in that: The step of screening the first feature to obtain the second feature includes: Calculating the Spearman correlation coefficient of the first feature; If the number of samples of the first feature is less than or equal to the preset sample value, the trend of the first feature is determined by comparing the Spearman correlation coefficient with the comparison table; If the number of samples of the first feature is greater than the preset sample value, a check value T is generated according to the Spearman correlation coefficient. where ρ xy is the Spearman correlation coefficient, and q is the number of samples, d is the level difference between the two variables, if the check value T is greater than the critical value of the comparison table, the two variables in the first feature have a trend, otherwise, there is no trend; The first feature with trend is retained, and the first feature without trend is deleted to obtain the second feature.

6. The life prediction method of a power output circuit according to claim 5, characterized in that: The predicting the life of the target power output circuit according to the second feature to obtain a prediction result includes: The target power output circuit is predicted based on the second feature and time to obtain a prediction result, and the prediction result is compared with a preset trend to determine the life of the target power output circuit.

7. The life prediction method of a power output circuit according to claim 6, characterized in that: The method further comprises: The target power output circuit is continuously detected, the second feature is updated, and the life of the target power output circuit is updated.

8. A life prediction device for a power output circuit, characterized in that: include: The first module is used to determine the signal collection points according to the design data file of the target power output circuit; A second module is used to collect a first signal of the target power output circuit through the information collection point, where the first signal is used to represent an original electrical signal of the target power output circuit; A third module is used to process the first signal by using signal time-frequency transformation to obtain the second signal; A fourth module is used to perform feature extraction on the second signal to obtain a first feature of the target power output circuit, where the first feature includes at least one of energy, information entropy, and time-domain statistical indicators; A fifth module is used to perform feature screening on the first feature to obtain a second feature, where the second feature is used to characterize the trend of the first feature; The sixth module is used to predict the life of the target power output circuit according to the second feature to obtain a prediction result.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the life prediction method of the power output circuit according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program is executed by a processor to implement the life prediction method of a power supply output circuit according to any one of claims 1 to 7.

Images