Fault detection method, device and equipment of power adapter and storage medium

Abstract

The invention relates to the technical field of fault detection, and discloses a fault detection method and device for a power adapter, equipment and a storage medium. The method comprises the steps of obtaining original power supply electrical data of preset multi-dimensional electrical parameters of a target power adapter, and performing feature decoupling on the original power supply electrical data based on electrical features corresponding to the electrical parameters to obtain basic electrical feature data; performing time-frequency characteristic decomposition and characteristic reconstruction of the operation state on the basic electrical characteristic data to obtain power state characteristic data; performing operation characteristic association of multilayer electrical parameters and evaluation of associated operation states on the power supply state characteristic data to obtain a power supply performance evaluation result, and performing state parameter mapping of fault characteristics on the power supply performance evaluation result to obtain power supply state fault characteristics; and carrying out fault feature counting and fault space positioning on the power supply state fault features to generate a fault detection result. According to the invention, the accuracy of the fault detection result of the power adapter is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and in particular to a method, device, equipment and storage medium for fault detection of a power adapter. Background Art

[0002] In the field of power adapter technology, fault detection is an important part of product development and quality control, and real-time monitoring and fault diagnosis of electrical parameters are the most critical and challenging stages of fault detection. Accurate detection of the operating status of power adapters is crucial for manufacturers, R&D institutions and testing centers, as it directly affects product reliability assessment, performance optimization and overall service life. Therefore, an effective fault detection method to monitor these electrical parameters is essential to ensure the operational stability and safety of power adapters.

[0003] Nowadays, digital signal processing technology is used to analyze sampled data, establish fault models to evaluate the operating status, or try to apply artificial intelligence technology to the fault diagnosis process to better evaluate the fault type and development trend. However, these methods still have challenges in integrating multi-dimensional electrical parameters, processing nonlinear characteristics, and adapting to dynamic loads. In addition, these detection methods often ignore some unique characteristics of power adapters, such as harmonic distortion, temperature drift, electromagnetic interference, etc., which may have a significant impact on fault judgment and location. That is, the existing fault detection methods of power adapters ignore multi-dimensional electrical parameters and some unique characteristics of power adapters, resulting in low accuracy of the final fault detection results. Summary of the invention

[0004] The main purpose of the present invention is to solve the problem that the existing power adapter fault detection method ignores the multi-dimensional electrical parameters and some unique features of the power adapter, resulting in low accuracy of the final fault detection result.

[0005] A first aspect of the present invention provides a power adapter fault detection method, the power adapter fault detection method comprising: obtaining original power electrical data of multi-dimensional electrical parameters preset by a target power adapter, and based on the electrical characteristics corresponding to each of the electrical parameters, performing feature decoupling on the original power electrical data to obtain basic electrical characteristic data; performing time-frequency feature decomposition and feature reconstruction of the basic electrical characteristic data of the operating state to obtain power state characteristic data; performing multi-layer electrical parameter operating feature association and associated operating state evaluation on the power state characteristic data to obtain a power performance evaluation result of the target power adapter, and performing state parameter mapping of the fault characteristics on the power performance evaluation result to obtain a power state fault characteristic; performing fault feature counting and fault space location on the power state fault characteristic to generate a fault detection result of the target power adapter.

[0006] Optionally, in a first implementation of the first aspect of the present invention, the step of obtaining original power electrical data of the multi-dimensional electrical parameters preset by the target power adapter includes: Perform analog-to-digital sampling and peak extraction on the input voltage of the target power adapter to obtain input voltage peak data, and perform variance calculation on the input voltage peak data to obtain voltage stability data; perform sampling and peak-to-peak value detection on the output voltage of the target power adapter within a preset output voltage bandwidth to obtain output ripple coefficient data, and collect the load jump response of the output ripple coefficient data to obtain voltage output response data; perform Hall detection and time domain decomposition on the input current of the target power adapter to obtain input harmonic component data, and perform phase detection processing on the input harmonic component data to obtain power factor data; perform shunt detection and ripple extraction on the output current of the target power adapter to obtain output current ripple data, and perform load change collection on the output current ripple data to obtain load characteristic data; The power devices of the target power adapter are subjected to hot spot extraction and temperature rise gradient calculation processing to obtain heat diffusion characteristic data; the radiation emission data of the electromagnetic field environment strength of the target power adapter are collected, and the electromagnetic intensity time-frequency transformation is performed on the radiation emission data to obtain electromagnetic interference characteristic data; the input and output power values ​​of the target power adapter are subjected to power conversion and multi-stage power consumption classification statistics to obtain loss distribution data at each stage, and the density of the preset unit volume power is calculated for the loss distribution data to obtain power density index data; the voltage stability data, the voltage output response data, the power factor data, the load characteristic data, the heat diffusion characteristic data, the electromagnetic interference characteristic data and the power density index data are timestamped and time-series aligned to obtain original power electrical data.

[0007] Optionally, in a second implementation of the first aspect of the present invention, the electrical characteristics corresponding to each of the electrical parameters are used to perform feature decoupling on the original power supply electrical data to obtain basic electrical characteristic data, including: performing frequency band decomposition on the voltage stability data in the basic electrical characteristic data to obtain grid voltage index data, and performing output voltage modal decomposition and linear transformation of transient voltage on the voltage output response data in the basic electrical characteristic data to obtain transient voltage characteristic data, and extracting current harmonic components and harmonic indicators from the power factor data in the basic electrical characteristic data to obtain current harmonic characteristic data, and performing wavelet packet decomposition and time-frequency decomposition of the load response on the load characteristic data in the basic electrical characteristic data to obtain load performance index data, and performing infrared spectroscopy on the heat diffusion characteristic data in the basic electrical characteristic data. The external heat map is constructed and the gradient of the thermal field distribution is calculated to obtain the device temperature index data, and the electromagnetic interference characteristic data in the basic electrical characteristic data are subjected to peak detection and spectrum energy calculation to obtain the electromagnetic index data, and the power density index data in the basic electrical characteristic data are subjected to loss decomposition and energy balance to obtain the power loss index data; the grid voltage index data, the instantaneous voltage characteristic data, the current harmonic characteristic data, the load performance index data, the device temperature index data, the electromagnetic index data, and the power loss index data are subjected to tensor decomposition to obtain the core power characteristic tensor, and the core power characteristic tensor is subjected to projection transformation to obtain the reduced-dimensional power characteristic data; the reduced-dimensional power characteristic data is subjected to power supply principal component extraction and electrical parameter index standardization transformation to obtain the basic electrical characteristic data.

[0008] Optionally, in a third implementation method of the first aspect of the present invention, the time-frequency characteristic decomposition and feature reconstruction of the operating state of the basic electrical characteristic data to obtain power supply state characteristic data includes: performing variational mode decomposition on the corresponding input and output voltage and current data in the basic electrical characteristic data to obtain an input and output fluctuation characteristic function set, and performing sequential Hilbert transform on the input and output fluctuation characteristic function set to obtain a voltage and current phase relationship matrix; performing power factor feature extraction on the voltage and current phase relationship matrix to obtain a power characteristic matrix, and calculating the input and output impedance of the power characteristic matrix to generate a power transfer characteristic matrix; performing closed-loop pole calculation and bandwidth margin calculation on the power transfer characteristic matrix to generate a power supply control stability matrix, and performing voltage and current loop coupling calculation on the power supply control stability matrix to generate a dual-loop characteristic matrix; performing transient response evaluation and power supply open-loop gain calculation on the dual-loop characteristic matrix to obtain power supply state characteristic data.

[0009] Optionally, in a fourth implementation method of the first aspect of the present invention, the power state characteristic data is subjected to multi-layer electrical parameter operation characteristic association and associated operation state evaluation to obtain the power performance evaluation result of the target power adapter, including: performing a timing network construction on the voltage stability data in the power state characteristic data to obtain a voltage characteristic subgraph, and constructing a current characteristic subgraph corresponding to the current harmonic characteristic data in the power state characteristic data, and performing topological feature merging on the voltage characteristic subgraph and the current characteristic subgraph to obtain an initial electrical characteristic graph; extracting voltage fluctuation characteristics from the initial electrical characteristic graph to obtain a voltage node matrix, and extracting current distortion characteristics from the initial electrical characteristic graph to obtain a current A node matrix is ​​prepared, and edge features are constructed for the voltage node matrix and the current node matrix to obtain an electrical parameter association matrix; multi-layer voltage and current coupling calculation is performed on the electrical parameter association matrix to obtain an electrical parameter coupling matrix, and harmonic distortion rate calculation is performed on the electrical parameter coupling matrix to obtain an electrical harmonic characteristic matrix; a synthesis calculation of harmonic power and fundamental power is performed on the electrical harmonic characteristic matrix to obtain a power quality matrix, and electrical characteristics are combined on the power quality matrix to obtain a comprehensive electrical characteristic matrix; a time-series evolution calculation is performed on the comprehensive electrical characteristic matrix to generate an electrical state evolution matrix, and feature space reconstruction and evaluation are performed on the electrical state evolution matrix to obtain a power performance evaluation result of the target power adapter.

[0010] Optionally, in a fifth implementation of the first aspect of the present invention, the state parameter mapping of the fault characteristics of the power supply performance evaluation results to obtain power supply state fault characteristics includes: separating the corresponding fundamental power data and harmonic power data in the power supply performance evaluation results to obtain a normal state power characteristic matrix and a fault state power characteristic matrix, and constructing a power supply state space for the normal state power characteristic matrix and the fault state power characteristic matrix to obtain a power supply state characteristic space; performing parameter mapping on multiple preset power supply performance parameters in the power supply state characteristic space to obtain a power supply performance characteristic matrix, and clustering multiple electrical parameters on the power supply performance characteristic matrix to obtain multiple fault type feature sets; based on preset power supply abnormality parameters, performing state probability calculation on the corresponding power supply abnormality parameter characteristics in the fault type feature set to obtain a fault probability matrix, and performing state transfer and fault association of multiple electrical parameters on each type of fault characteristics in the fault probability matrix to obtain a fault characteristic association matrix; performing power supply fault mode identification on the fault characteristic association matrix to obtain a fault mode feature vector, and performing state feature conversion on the fault mode feature vector to obtain a power supply state fault characteristic.

[0011] Optionally, in a sixth implementation manner of the first aspect of the present invention, the fault feature counting and fault spatial positioning of the power state fault features to generate the fault detection result of the target power adapter includes: based on preset power fault features, classifying and counting the fault features of the power state fault features to obtain various types of fault feature count matrices, and spatially positioning the voltage fault features in the fault feature count matrix to obtain a voltage fault spatial distribution map, and spatially positioning the current fault features in the fault feature count matrix to obtain a current fault spatial distribution map, and spatially positioning the power fault features in the fault feature count matrix to obtain a power fault spatial distribution map, wherein the power fault features include current fault features, voltage fault features, and power fault features; performing fault association and fault feature fusion on the voltage fault spatial distribution map, the current fault spatial distribution map, and the power fault spatial distribution map to obtain a comprehensive fault feature matrix; performing fault type determination on the comprehensive fault feature matrix to obtain fault type data, and performing fault degree assessment on the fault type data to obtain fault degree data; performing fault feature combination on the fault type data and the fault degree data to generate the fault detection result of the target power adapter.

[0012] A second aspect of the present invention provides a fault detection device for a power adapter, the fault detection device for the power adapter comprising: a feature decoupling module, used to obtain original power electrical data of multi-dimensional electrical parameters preset by a target power adapter, and based on the electrical characteristics corresponding to each of the electrical parameters, perform feature decoupling on the original power electrical data to obtain basic electrical feature data; a feature reconstruction module, used to perform time-frequency feature decomposition and feature reconstruction of the basic electrical feature data of the operating state to obtain power state feature data; a performance evaluation module, used to perform operating feature association of multiple layers of electrical parameters and evaluation of the associated operating state on the power state feature data to obtain a power performance evaluation result of the target power adapter, and perform state parameter mapping of the fault characteristics on the power performance evaluation result to obtain a power state fault characteristic; a fault detection module, used to perform fault feature counting and fault space location on the power state fault characteristic to generate a fault detection result of the target power adapter.

[0013] The third aspect of the present invention provides a power adapter fault detection device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the power adapter fault detection device performs each step of the above-mentioned power adapter fault detection method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute each step of the above-mentioned power adapter fault detection method.

[0015] The above-mentioned fault detection method, device, equipment and storage medium of the power adapter. By acquiring the original power electrical data of the preset multi-dimensional electrical parameters of the target power adapter, and based on the electrical characteristics corresponding to each electrical parameter, the original power electrical data is feature decoupled to obtain basic electrical characteristic data; the basic electrical characteristic data is decomposed into the time-frequency characteristics of the operation state and the feature reconstruction is performed to obtain the power state characteristic data; the power state characteristic data is associated with the operation characteristics of multiple layers of electrical parameters and the associated operation state is evaluated to obtain the power performance evaluation result of the target power adapter, and the state parameter mapping of the fault characteristics is performed on the power performance evaluation result to obtain the power state fault characteristics; the fault characteristics are counted and the fault spatial location is performed on the power state fault characteristics to generate the fault detection result of the target power adapter. Compared with the prior art, the present application obtains the basic electrical characteristics by multi-dimensional acquisition and feature decoupling of the electrical parameters of the target power adapter, and then the time-frequency decomposition and reconstruction of these characteristics are performed to obtain the power operation state characteristics; then the performance evaluation results are obtained and the fault characteristics are mapped through multi-layer parameter association analysis and state evaluation; finally, the counting statistics and spatial location are performed based on the fault characteristics to output the fault detection result. Through hierarchical data processing and feature analysis, the problem of accurate identification of power adapter fault detection is solved, especially in terms of voltage stability, current harmonics and power quality. The dynamic characteristics and fault characteristics of the power adapter are fully considered, and the detection accuracy is effectively improved; and a multi-level feature decoupling and fault mapping strategy is adopted to realize the correlation analysis between electrical parameters and enhance the reliability of fault diagnosis; in addition, through feature space reconstruction and fault location, the fault type and location are accurately identified, thereby realizing efficient and accurate detection of power adapter faults as a whole.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of a first embodiment of a method for detecting a fault of a power adapter according to an embodiment of the present invention; Figure 2 Schematic diagram of an embodiment of a fault detection device for a power adapter according to an embodiment of the present invention; Figure 3 FIG. 1 is a schematic diagram of an embodiment of a fault detection device for a power adapter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0021] To facilitate understanding of this embodiment, the specific process of the embodiment of the present invention is described below. Figure 1 , a first embodiment of a fault detection method of a power adapter in an embodiment of the present invention includes: 101. Acquire original power electrical data of multi-dimensional electrical parameters preset by the target power adapter, and perform feature decoupling on the original power electrical data based on electrical features corresponding to each electrical parameter to obtain basic electrical feature data; The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0022] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0023] In this embodiment, the multidimensional electrical parameters here include input and output voltage parameters, input and output current parameters, power device parameters, electromagnetic parameters, temperature parameters and power density parameters, which can also be increased or decreased according to the needs of fault detection; the input voltage of the target power adapter is sampled and the peak value is extracted to obtain the input voltage peak data, and the variance of the input voltage peak data is calculated to obtain the voltage stability data; the output voltage of the target power adapter is sampled and the peak-to-peak value is detected within the preset output voltage bandwidth to obtain the output ripple coefficient data, and the load jump response of the output ripple coefficient data is collected to obtain the voltage output response data; the input current of the target power adapter is subjected to Hall detection and time domain decomposition to obtain to the input harmonic component data, and perform phase detection processing on the input harmonic component data to obtain power factor data; perform shunt detection and ripple extraction on the output current of the target power adapter to obtain output current ripple data, and perform load change collection on the output current ripple data to obtain load characteristic data; perform hot spot extraction and temperature rise gradient calculation processing on the power device of the target power adapter to obtain thermal diffusion characteristic data; collect the radiation emission data of the electromagnetic field environment intensity of the environment in which the target power adapter is located, and perform time-frequency transformation of the electromagnetic intensity on the radiation emission data to obtain electromagnetic interference characteristic data; perform power conversion and multi-stage power consumption on the input and output power values ​​of the target power adapter Classify and count to obtain loss distribution data at each level, and perform density calculation of preset power per unit volume on the loss distribution data to obtain power density index data; perform timestamp marking and timing alignment on the voltage stability data, the voltage output response data, the power factor data, the load characteristic data, the heat diffusion characteristic data, the electromagnetic interference characteristic data and the power density index data to obtain original power supply electrical data; perform frequency band decomposition on the voltage stability data in the basic electrical characteristic data to obtain grid voltage index data, and perform output voltage modal decomposition and linear transformation of transient voltage on the voltage output response data in the basic electrical characteristic data to obtain transient voltage characteristic data, and Extracting current harmonic components and harmonic indicators from power factor data in the basic electrical characteristic data to obtain current harmonic characteristic data, performing wavelet packet decomposition and load response time-frequency decomposition on load characteristic data in the basic electrical characteristic data to obtain load performance indicator data, constructing an infrared thermal map and calculating the gradient of thermal field distribution on heat diffusion characteristic data in the basic electrical characteristic data to obtain device temperature indicator data, performing peak detection and spectrum energy calculation on electromagnetic interference characteristic data in the basic electrical characteristic data to obtain electromagnetic indicator data, and performing loss decomposition and energy balance on power density indicator data in the basic electrical characteristic data to obtain power loss indicator data;Perform tensor decomposition on the grid voltage index data, the instantaneous voltage characteristic data, the current harmonic characteristic data, the load performance index data, the device temperature index data, the electromagnetic index data, and the power loss index data to obtain a core power characteristic tensor, and perform projection transformation on the core power characteristic tensor to obtain reduced-dimensional power characteristic data; extract the power main component and perform standardized transformation on the electrical parameter index of the reduced-dimensional power characteristic data to obtain basic electrical characteristic data. ;

[0024] In practical applications, the input voltage signal of the target power adapter is first sampled at a high speed. The sampling frequency is usually set to more than 10 times the switching frequency to ensure that the detailed characteristics of the voltage waveform can be accurately captured. The peak data of the input voltage is extracted through the peak detection circuit, and the variance of the voltage peak sequence is calculated using the sliding window method to obtain the voltage stability index data reflecting the voltage stability. At the same time, the output voltage is sampled with limited bandwidth. The sampling bandwidth is usually set to 1 / 5 of the switching frequency to avoid the interference of switching noise. Then, the ripple factor of the output voltage is obtained through the peak-to-peak detection circuit, and the dynamic response characteristic data of the output voltage is recorded under load mutation conditions. In terms of current parameter acquisition, the input current is non-destructively detected using a Hall current sensor. The time domain current signal is decomposed into fundamental and harmonic components through Fourier transform, and the phase difference of the voltage and current signals is detected through a phase-locked loop circuit to calculate the power factor data. For the detection of the output current, a precision shunt resistor is used for sampling, and the current ripple information is extracted through bandpass filtering. At the same time, under different load conditions, the output current is sampled. The current change characteristics are recorded to obtain the load characteristic data; in order to monitor the temperature distribution of power devices, an infrared thermal imager is used to scan key devices, extract hot spot location information, and calculate the spatial gradient of the temperature field to obtain heat diffusion characteristic data; in terms of electromagnetic compatibility detection, the radiation emission intensity generated by the power adapter when working is collected by an electromagnetic field strength probe, and the electromagnetic signal is analyzed in the time and frequency domain using short-time Fourier transform to obtain the time-frequency distribution data reflecting the electromagnetic interference characteristics; for the analysis of power loss, a power loss model is established by measuring the difference between input and output power and combining the loss distribution of power conversion links at all levels, and then the power density index data per unit volume is calculated according to the volume of the power adapter; in order to ensure the timing consistency of various parameter data, high-precision timestamps are added to all collected data, and digital synchronous sampling technology is used to achieve timing alignment between different parameters, that is, firstly, a unified sampling reference is generated based on the clock signal of the power adapter, and then the synchronous trigger pulses of each sampling signal are generated through the digital phase-locked loop technology to ensure that all data acquisition channels complete sampling at the same time. Through the coordinated collection and synchronous processing of multi-dimensional electrical parameters, the original electrical data containing complete information such as voltage stability, output response, power quality, thermal characteristics, and electromagnetic characteristics is finally obtained. This not only takes into account the electrical characteristics and working characteristics of the power adapter, but also can fully reflect the operating status and potential fault information of the equipment.

[0025] Secondly, the voltage stability data is decomposed into multiple frequency bands using a bandpass filter group to extract voltage characteristics of different frequency bands such as the power frequency fundamental wave, high-frequency ripple and low-frequency fluctuation. The comprehensive index reflecting the quality of the power grid (i.e., the power grid voltage index data) is calculated through these characteristics. At the same time, the empirical mode decomposition technology is applied to the voltage output response data to decompose the complex voltage response signal into multiple intrinsic mode functions, and the linear expression of the transient voltage characteristics is realized through the Hilbert transform (Hilbert transform), so as to accurately capture the dynamic characteristics of voltage fluctuations (i.e., transient voltage characteristic data). In terms of power factor data processing, the fast Fourier transform is used to perform spectrum analysis on the input current, extract the amplitude and phase information of the fundamental wave and each harmonic component, calculate the total harmonic distortion rate and each harmonic content, and obtain comprehensive current harmonic characteristic data. For load characteristic data, the wavelet packet decomposition technology is used to perform multi-scale time-frequency analysis, separate the load response characteristics of different frequency bands, and obtain the load dynamic performance index data through reconstruction. In terms of thermal characteristic analysis Based on the heat diffusion characteristic data, an infrared thermal map model is constructed to calculate the spatial gradient distribution of the temperature field. The thermal resistance characteristic index of the device (i.e., the device temperature index data) is obtained by analyzing the relationship between the heat flux density and the temperature gradient. For the electromagnetic interference characteristic data, the peak detection algorithm is used to extract the interference peak of each frequency band, and the spectrum energy distribution is calculated to obtain the quantitative index reflecting the electromagnetic compatibility (i.e., the electromagnetic index data). In the power loss analysis, a multi-level power conversion model is established to decompose the loss components of each level of conversion link, and the power loss distribution characteristics (i.e., the power loss index data) are calculated in combination with the energy balance principle. The multi-dimensional feature data obtained above are then organized into a third-order tensor structure, the Tucker decomposition method is used to extract the core feature mode, and the high-dimensional features are mapped to the low-dimensional feature space through orthogonal projection transformation, so as to achieve data dimensionality reduction while retaining the key feature information. The principal component analysis method is then used to extract the main feature components of the feature data after dimensionality reduction, and the standardized basic electrical feature data is obtained through z-score standardization. This multi-level feature extraction and dimensionality reduction processing method not only ensures the integrity of the electrical features, but also improves the efficiency and accuracy of subsequent fault diagnosis.

[0026] 102. Perform time-frequency feature decomposition and feature reconstruction of the operating state on the basic electrical feature data to obtain power state feature data; In this embodiment, variational modal decomposition is performed on the input and output voltage and current data corresponding to the basic electrical characteristic data to obtain an input and output fluctuation characteristic function set, and the input and output fluctuation characteristic function set is subjected to sequential Hilbert transform to obtain a voltage and current phase relationship matrix; power factor feature extraction is performed on the voltage and current phase relationship matrix to obtain a power characteristic matrix, and input and output impedances are calculated on the power characteristic matrix to generate a power transfer characteristic matrix; closed-loop pole calculation and bandwidth margin calculation are performed on the power transfer characteristic matrix to generate a power control stability matrix, and voltage and current loop coupling calculation is performed on the power control stability matrix to generate a dual-loop characteristic matrix; transient response evaluation and power supply open-loop gain calculation are performed on the dual-loop characteristic matrix to obtain power state characteristic data.

[0027] In practical applications, the variational mode decomposition algorithm is applied to the input and output voltage and current signals in the basic electrical characteristic data. The algorithm decomposes the complex non-stationary signal into a series of intrinsic mode functions (where these functions can accurately reflect the fluctuation characteristics of voltage and current) by iteratively optimizing the center frequency and bandwidth of different modes, and performs Hilbert transform on the obtained set of fluctuation characteristic functions in turn to extract the instantaneous amplitude and phase information of each mode. The phase relationship between the input and output voltage and current is characterized by constructing a phase difference matrix, thereby obtaining a voltage and current phase relationship matrix; then, based on the voltage and current phase relationship matrix, the power factor characteristics corresponding to each frequency component are calculated to form a power characteristic matrix, which contains key information such as fundamental power and harmonic power, and the input and output at different frequencies are obtained by calculating the voltage and current ratio at the input and output ends. Impedance characteristics, construct a characteristic matrix that reflects the power transfer characteristics (wherein this matrix describes the gain and phase characteristics of the power supply at different frequencies), and generate a power transfer characteristic matrix; then the power transfer characteristic matrix is ​​used to calculate the pole distribution of the closed-loop transfer function and the bandwidth margin of the power adapter, the stability characteristics of the power adapter are evaluated, and the power control stability matrix is ​​generated. The coupling degree of the voltage loop and the current loop is analyzed on the matrix, and the interaction between the two control loops is calculated to obtain a characteristic matrix that characterizes the dual-loop control characteristics; then the dual-loop characteristic matrix is ​​used to analyze the response of the power adapter, evaluate the dynamic response characteristics of the voltage and current under transient conditions such as load mutations, and calculate the open-loop gain characteristics of the power adapter at different frequencies, and comprehensively form the power state characteristic data to fully reflect the dynamic performance and control characteristics of the power adapter.

[0028] 103. Perform multi-layer electrical parameter operation characteristic association and associated operation status evaluation on the power state characteristic data to obtain a power performance evaluation result of the target power adapter, and perform state parameter mapping of the fault characteristic on the power performance evaluation result to obtain a power state fault characteristic; In this embodiment, the preset power supply performance parameters include voltage stability, current harmonic rate, and power factor, and the parameters can also be increased or decreased according to the needs of fault detection; the preset power supply abnormality parameters include voltage abnormality, current abnormality, and power abnormality characteristics, and the parameters can also be increased or decreased according to the needs of fault detection; the voltage stability data in the power supply state characteristic data is constructed by a timing network to obtain a voltage characteristic subgraph, and a current characteristic subgraph corresponding to the current harmonic characteristic data in the power supply state characteristic data is constructed, and the voltage characteristic subgraph and the current characteristic subgraph are topologically merged to obtain an initial electrical characteristic graph; the initial The electrical characteristic diagram is used to extract voltage fluctuation characteristics to obtain a voltage node matrix, and the initial electrical characteristic diagram is used to extract current distortion characteristics to obtain a current node matrix, and the voltage node matrix and the current node matrix are used to construct edge characteristics to obtain an electrical parameter association matrix; multi-layer voltage-current coupling calculation is performed on the electrical parameter association matrix to obtain an electrical parameter coupling matrix, and the harmonic distortion rate is calculated on the electrical parameter coupling matrix to obtain an electrical harmonic characteristic matrix; the electrical harmonic characteristic matrix is ​​used to synthesize harmonic power and fundamental power to obtain a power quality matrix, and the power quality matrix is ​​used to perform electrical The invention relates to a method for evaluating the power performance of a target power adapter by combining the power characteristics of the target power adapter and the power supply characteristics of the target power adapter to obtain a comprehensive electrical characteristic matrix; performing time-series evolution calculation on the comprehensive electrical characteristic matrix to generate an electrical state evolution matrix, and performing feature space reconstruction and evaluation on the electrical state evolution matrix to obtain a power performance evaluation result of the target power adapter; separating the fundamental power data and the harmonic power data corresponding to the power performance evaluation result to obtain a normal state power characteristic matrix and a fault state power characteristic matrix, and performing power state space construction on the normal state power characteristic matrix and the fault state power characteristic matrix to obtain a power state characteristic space; and performing power state space reconstruction on a plurality of preset power performance parameters in the power state characteristic space. The power supply performance characteristic matrix is ​​mapped to a parameter, and the power supply performance characteristic matrix is ​​clustered with multiple electrical parameters to obtain multiple fault type feature sets; based on the preset power supply abnormality parameters, the state probability of the corresponding power supply abnormality parameter characteristics in the fault type feature set is calculated to obtain a fault probability matrix, and the various fault characteristics in the fault probability matrix are subjected to state transfer and fault association of multiple electrical parameters to obtain a fault feature association matrix; power supply fault mode recognition is performed on the fault feature association matrix to obtain a fault mode feature vector, and the fault mode feature vector is converted into a state feature to obtain a power supply state fault feature.

[0029] In practical applications, first, a time-series correlation network is constructed based on voltage stability data, in which network nodes represent voltage states at different times, and edges represent state transfer relationships, thereby forming a characteristic subgraph that reflects the dynamic characteristics of voltage; at the same time, a current correlation network is constructed based on current harmonic characteristic data, in which nodes represent harmonic components of different frequencies, and edges represent the mutual influence between harmonics, forming a current characteristic subgraph, and the voltage and current characteristic subgraphs are merged through a topological feature fusion algorithm to establish a unified electrical characteristic graph, which contains the complete dynamic characteristics of voltage and current; then, based on the initial electrical characteristic graph, a graph neural network is used to extract the topological features of voltage fluctuations to obtain a node matrix that reflects the voltage state distribution, and at the same time, the topological features of current distortion are extracted to obtain a node matrix that characterizes the current harmonic distribution, and these two matrices are connected through an edge feature construction algorithm to form a description of the voltage and current correlation relationship. Parameter association matrix; then a multi-level voltage-current coupling analysis is performed on the matrix, the coupling coefficients of different levels are calculated, the electrical parameter coupling matrix is ​​obtained, and the total harmonic distortion rate and each harmonic content of the power adapter are calculated based on the matrix to form an electrical harmonic characteristic matrix; then, through harmonic power calculation and fundamental power synthesis, a characteristic matrix reflecting the power quality of the power supply is obtained, which integrates information such as fundamental power factor and harmonic power loss, and the power quality matrix is ​​subjected to feature combination and dimensionality reduction processing to obtain a comprehensive matrix containing multi-dimensional electrical characteristics; then, by analyzing the time series evolution characteristics of the matrix, a state transition model is constructed, and the performance indicators of the power adapter are evaluated in the reconstructed feature space, and finally a performance evaluation result that comprehensively reflects the operating status of the power adapter is obtained. Through this multi-level and multi-dimensional evaluation, the dynamic characteristics and operating status of the power adapter can be accurately reflected.

[0030] Secondly, the power data in the power performance evaluation results are separated into fundamental and harmonic components, that is, the fundamental power component and each harmonic power component are extracted through Fourier analysis, and these components are classified into normal working state and fault state, and two power characteristic matrices (i.e., normal state power characteristic matrix and fault state power characteristic matrix) are constructed. Based on these two matrices, a state space model is constructed, in which state variables include key parameters such as voltage, current, and power. The dynamic characteristics of the power adapter are described by state equations to form a complete power state characteristic space; then, in the constructed state characteristic space, nonlinear mapping is performed on the pre-defined power performance parameters (such as voltage stability, power factor, harmonic distortion rate, etc.) to obtain a characteristic matrix reflecting the power performance, and through By using the improved K-means clustering algorithm, the electrical parameters in the feature matrix are subjected to multi-dimensional clustering analysis, similar fault features are classified, and different types of fault feature sets are formed; then, based on the pre-set power supply abnormality parameter threshold, the probability of occurrence of each fault type is calculated, a fault probability matrix is ​​constructed, and the transfer characteristics of the fault state are analyzed through the Markov chain model, and the fault association relationship between electrical parameters is established to obtain a complete fault feature association matrix; then, the fault feature association matrix is ​​analyzed using pattern recognition algorithms such as support vector machines, typical fault modes are identified and corresponding feature vectors are extracted, and these feature vectors are transformed into state space and mapped into standardized feature space to form the final power supply state fault feature description. Through the precise separation of power characteristics, the effective construction of state space, the accurate extraction of fault features, and the reliable identification of fault modes, the accurate characterization of the power adapter fault state is achieved.

[0031] 104. Count the fault characteristics of the power state fault characteristics and locate the fault space to generate a fault detection result of the target power adapter.

[0032] In this embodiment, the power supply fault features include voltage fault, current fault and power fault, and the power supply fault features can also be increased or decreased according to the needs of fault detection; based on the preset power supply fault features, the power supply state fault features are classified and counted to obtain various fault feature count matrices, and the voltage fault features in the fault feature count matrix are spatially located to obtain a voltage fault spatial distribution map, and the current fault features in the fault feature count matrix are spatially located to obtain a current fault spatial distribution map, and the power fault features in the fault feature count matrix are spatially located to obtain a power fault spatial distribution map, wherein the power supply fault features include current fault features, voltage fault features and power fault features; the voltage fault spatial distribution map, the current fault spatial distribution map and the power fault spatial distribution map are fault associated and fault feature fused to obtain a comprehensive fault feature matrix; the comprehensive fault feature matrix is ​​fault type determined to obtain fault type data, and the fault degree is evaluated to obtain fault degree data; the fault type data and the fault degree data are fault feature combined to generate a fault detection result of the target power adapter.

[0033] In practical applications, the state fault features are classified and counted based on the predefined power supply fault feature library, and the occurrence frequency of three types of fault features, voltage, current and power, are identified through the feature matching algorithm. The matrix construction formula is: ; in, is the feature count matrix element, is the feature weight coefficient, is the kth characteristic function, is the feature space coordinate, λ is the attenuation coefficient, and n is the number of features. For example, assuming that a power supply has 10 overvoltages and 15 undervoltages in the voltage fault feature, a voltage fault feature count submatrix can be constructed, thereby constructing a three-dimensional fault feature count matrix; and the spatial mapping algorithm is used to transform the spatial coordinates of the voltage fault features in the matrix to generate a spatial heat map reflecting the voltage fault distribution. At the same time, similar processing is performed on the current fault features and the power fault features to obtain the spatial distribution maps of the current fault and the power fault respectively (where these distribution maps intuitively show the location distribution characteristics of different types of faults in the power adapter); and then after obtaining the spatial distribution maps of the three types of faults, the image fusion algorithm is used to perform multi-layer superposition analysis on these distribution maps, calculate the spatial correlation between the fault points, and extract the mutual influence relationship between the faults through the feature fusion network (i.e., tensor network) to form a comprehensive fault feature matrix containing complete fault information. The feature fusion formula is: , where Y is the fused feature tensor, G is the core tensor, and U1 ,U 2 ,U 3 is the factor matrix, W is the weight matrix, and ×ᵢ is the i-mode tensor product. For example, the three types of fault feature maps of voltage, current, and power are fused through the tensor network to obtain a comprehensive fault feature matrix; then based on the comprehensive fault feature matrix, the multi-classifier integration algorithm is used to determine the fault type and obtain specific fault type data. At the same time, by establishing a fault severity assessment model, the hazard level of each fault type is calculated to obtain quantitative fault degree data; then the fault type data and the fault degree data are feature cascaded, and a complete fault description vector is constructed through the feature combination algorithm. The vector contains both the type information of the fault, the severity of the fault, and the spatial distribution characteristics of the fault, thereby forming a comprehensive power adapter fault detection report. Through the precise statistics of fault characteristics, the visual expression of spatial distribution, the in-depth analysis of fault associations, and the quantitative assessment of fault degrees, accurate detection and reliable diagnosis of power adapter faults are achieved.

[0034] In the embodiment of the present invention, the electrical parameters of the target power adapter are multi-dimensionally collected and feature decoupled to obtain basic electrical features, and then these features are time-frequency decomposed and reconstructed to obtain power supply operation status features; then, through multi-layer parameter association analysis and state evaluation, performance evaluation results are obtained and fault features are mapped; finally, counting statistics and spatial positioning are performed based on the fault features to output fault detection results. Through hierarchical data processing and feature analysis, the problem of accurate identification of power adapter fault detection is solved, especially in terms of voltage stability, current harmonics and power quality, the dynamic characteristics and fault characteristics of the power adapter are fully considered, and the detection accuracy is effectively improved; and the multi-level feature decoupling and fault mapping strategy are adopted, which not only realizes the association analysis between electrical parameters, but also enhances the reliability of fault diagnosis; in addition, through feature space reconstruction and fault location, the fault type and location are accurately identified, thereby realizing efficient and accurate detection of power adapter faults as a whole.

[0035] The above describes the fault detection method of the power adapter in the embodiment of the present invention. The following describes the fault detection device of the power adapter in the embodiment of the present invention. Figure 2 , an embodiment of a fault detection device of a power adapter in an embodiment of the present invention includes: The feature decoupling module 201 is used to obtain the original power electrical data of the multi-dimensional electrical parameters preset by the target power adapter, and perform feature decoupling on the original power electrical data based on the electrical features corresponding to each of the electrical parameters to obtain basic electrical feature data; A feature reconstruction module 202 is used to perform time-frequency feature decomposition and feature reconstruction of the operating state of the basic electrical feature data to obtain power state feature data; The performance evaluation module 203 is used to perform multi-layer electrical parameter operation characteristic association and associated operation state evaluation on the power state characteristic data to obtain the power performance evaluation result of the target power adapter, and perform state parameter mapping of the fault characteristic on the power performance evaluation result to obtain the power state fault characteristic; The fault detection module 204 is used to count the fault characteristics and locate the fault space of the power state fault characteristics, and generate a fault detection result of the target power adapter.

[0036] In the embodiment of the present invention, the electrical parameters of the target power adapter are multi-dimensionally collected and feature decoupled to obtain basic electrical features, and then these features are time-frequency decomposed and reconstructed to obtain power supply operation status features; then, through multi-layer parameter association analysis and state evaluation, performance evaluation results are obtained and fault features are mapped; finally, counting statistics and spatial positioning are performed based on the fault features to output fault detection results. Through hierarchical data processing and feature analysis, the problem of accurate identification of power adapter fault detection is solved, especially in terms of voltage stability, current harmonics and power quality, the dynamic characteristics and fault characteristics of the power adapter are fully considered, and the detection accuracy is effectively improved; and the multi-level feature decoupling and fault mapping strategy are adopted, which not only realizes the association analysis between electrical parameters, but also enhances the reliability of fault diagnosis; in addition, through feature space reconstruction and fault location, the fault type and location are accurately identified, thereby realizing efficient and accurate detection of power adapter faults as a whole.

[0037] above Figure 2 The fault detection device of the power adapter in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The fault detection device of the power adapter in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0038] Figure 31 is a schematic diagram of the structure of a power adapter fault detection device provided by an embodiment of the present invention. The power adapter fault detection device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage devices) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the power adapter fault detection device 300. Furthermore, the processor 310 may be configured to communicate with the storage medium 330 to execute a series of instruction operations in the storage medium 330 on the power adapter fault detection device 300.

[0039] The power adapter fault detection device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 3 The structure of the fault detection device of the power adapter shown does not constitute a limitation on the fault detection device of the power adapter, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0040] The present invention also provides a power adapter fault detection device, the computer device includes a memory and a processor, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes each step of the power adapter fault detection method in the above-mentioned embodiments.

[0041] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the various steps of the power adapter fault detection method.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0043] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or 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 to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0044] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0045] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting a fault of a power adapter, characterized in that: The power adapter fault detection method comprises: Acquire original power electrical data of multi-dimensional electrical parameters preset by the target power adapter, and perform feature decoupling on the original power electrical data based on electrical features corresponding to each of the electrical parameters to obtain basic electrical feature data; Performing time-frequency feature decomposition and feature reconstruction of the operating state on the basic electrical feature data to obtain power state feature data; Performing multi-layer electrical parameter operation characteristic association and associated operation state evaluation on the power state characteristic data to obtain a power performance evaluation result of the target power adapter, and performing state parameter mapping of fault characteristics on the power performance evaluation result to obtain a power state fault characteristic; Fault feature counting and fault spatial location are performed on the power state fault feature to generate a fault detection result of the target power adapter.

2. The power adapter fault detection method according to claim 1, characterized in that: The step of obtaining the original power supply electrical data of the preset multi-dimensional electrical parameters of the target power adapter includes: Performing analog-to-digital sampling and peak extraction on the input voltage of the target power adapter to obtain input voltage peak data, and performing variance calculation on the input voltage peak data to obtain voltage stability data; Sampling the output voltage of the target power adapter at a preset output voltage bandwidth and performing peak-to-peak detection to obtain output ripple coefficient data, and collecting load jump response of the output ripple coefficient data to obtain voltage output response data; Performing Hall detection and time domain decomposition on the input current of the target power adapter to obtain input harmonic component data, and performing phase detection processing on the input harmonic component data to obtain power factor data; Performing shunt detection and ripple extraction on the output current of the target power adapter to obtain output current ripple data, and performing load change collection on the output current ripple data to obtain load characteristic data; Perform hot spot extraction and temperature rise gradient calculation on the power device of the target power adapter to obtain thermal diffusion characteristic data; Collecting radiation emission data of the electromagnetic field environment strength of the target power adapter, and performing time-frequency transformation of the electromagnetic intensity on the radiation emission data to obtain electromagnetic interference characteristic data; Performing power conversion and multi-level power consumption classification statistics on the input and output power values ​​of the target power adapter to obtain loss distribution data at each level, and performing density calculation of preset unit volume power on the loss distribution data to obtain power density index data; The voltage stability data, the voltage output response data, the power factor data, the load characteristic data, the heat diffusion characteristic data, the electromagnetic interference characteristic data and the power density index data are time stamped and time aligned to obtain original power supply electrical data.

3. The power adapter fault detection method according to claim 2, characterized in that: The step of performing feature decoupling on the original power supply electrical data based on the electrical features corresponding to the electrical parameters to obtain basic electrical feature data includes: Perform frequency band decomposition on the voltage stability data in the basic electrical characteristic data to obtain grid voltage index data, perform output voltage modal decomposition and linear transformation of transient voltage on the voltage output response data in the basic electrical characteristic data to obtain transient voltage characteristic data, extract current harmonic components and harmonic indexes from the power factor data in the basic electrical characteristic data to obtain current harmonic characteristic data, perform wavelet packet decomposition and time-frequency decomposition of load response on the load characteristic data in the basic electrical characteristic data to obtain load performance index data, construct infrared thermal map and calculate gradient of thermal field distribution on the heat diffusion characteristic data in the basic electrical characteristic data to obtain device temperature index data, perform peak detection and spectrum energy calculation on the electromagnetic interference characteristic data in the basic electrical characteristic data to obtain electromagnetic index data, and perform loss decomposition and energy balance on the power density index data in the basic electrical characteristic data to obtain power loss index data; Performing tensor decomposition on the grid voltage index data, the instantaneous voltage characteristic data, the current harmonic characteristic data, the load performance index data, the device temperature index data, the electromagnetic index data, and the power loss index data to obtain a core power characteristic tensor, and performing a projection transformation on the core power characteristic tensor to obtain reduced-dimensional power characteristic data; The power supply characteristic data with reduced dimension is subjected to power supply principal component extraction and electrical parameter index standardization transformation to obtain basic electrical characteristic data.

4. The power adapter fault detection method according to claim 3, characterized in that: The step of performing time-frequency feature decomposition and feature reconstruction of the operating state of the basic electrical feature data to obtain power state feature data includes: Performing variational modal decomposition on the input and output voltage and current data corresponding to the basic electrical characteristic data to obtain an input and output fluctuation characteristic function set, and performing sequential Hilbert transformation on the input and output fluctuation characteristic function set to obtain a voltage and current phase relationship matrix; Extracting power factor characteristics from the voltage-current phase relationship matrix to obtain a power characteristic matrix, and calculating input and output impedances from the power characteristic matrix to generate a power transfer characteristic matrix; Performing closed-loop pole calculation and bandwidth margin calculation on the power transfer characteristic matrix to generate a power control stability matrix, and performing voltage-current loop coupling calculation on the power control stability matrix to generate a dual-loop characteristic matrix; Transient response evaluation and power supply open-loop gain calculation are performed on the dual-loop characteristic matrix to obtain power supply state characteristic data.

5. The power adapter fault detection method according to claim 3, characterized in that: The step of performing multi-layer electrical parameter operation characteristic association and associated operation state evaluation on the power state characteristic data to obtain a power performance evaluation result of the target power adapter includes: Performing a timing network construction on the voltage stability data in the power state characteristic data to obtain a voltage characteristic subgraph, and constructing a current characteristic subgraph corresponding to the current harmonic characteristic data in the power state characteristic data, and performing topological feature merging on the voltage characteristic subgraph and the current characteristic subgraph to obtain an initial electrical characteristic graph; Extracting voltage fluctuation characteristics from the initial electrical characteristic diagram to obtain a voltage node matrix, extracting current distortion characteristics from the initial electrical characteristic diagram to obtain a current node matrix, and constructing edge characteristics from the voltage node matrix and the current node matrix to obtain an electrical parameter association matrix; Performing multi-layer voltage-current coupling calculation on the electrical parameter correlation matrix to obtain an electrical parameter coupling matrix, and performing harmonic distortion rate calculation on the electrical parameter coupling matrix to obtain an electrical harmonic characteristic matrix; Performing synthetic calculation of harmonic power and fundamental power on the electrical harmonic characteristic matrix to obtain a power quality matrix, and performing electrical characteristic combination on the power quality matrix to obtain a comprehensive electrical characteristic matrix; The comprehensive electrical characteristic matrix is ​​subjected to time-series evolution calculation to generate an electrical state evolution matrix, and the electrical state evolution matrix is ​​subjected to feature space reconstruction and evaluation to obtain a power performance evaluation result of a target power adapter.

6. The power adapter fault detection method according to claim 5, characterized in that: The performing state parameter mapping of the fault characteristic on the power supply performance evaluation result to obtain the power supply state fault characteristic includes: Separating the fundamental power data and harmonic power data corresponding to the power performance evaluation result to obtain a normal state power feature matrix and a fault state power feature matrix, and constructing a power state space for the normal state power feature matrix and the fault state power feature matrix to obtain a power state feature space; Performing parameter mapping on a plurality of preset power performance parameters in the power state feature space to obtain a power performance feature matrix, and clustering a plurality of electrical parameters on the power performance feature matrix to obtain a plurality of fault type feature sets; Based on the preset power supply abnormality parameters, the state probability calculation is performed on the corresponding power supply abnormality parameter characteristics in the fault type feature set to obtain a fault probability matrix, and the state transition and fault association of various electrical parameters are performed on various fault characteristics in the fault probability matrix to obtain a fault feature association matrix; Power supply fault mode recognition is performed on the fault feature association matrix to obtain a fault mode feature vector, and state feature conversion is performed on the fault mode feature vector to obtain a power supply state fault feature.

7. The power adapter fault detection method according to claim 1, characterized in that: The performing fault feature counting and fault spatial positioning on the power state fault feature to generate a fault detection result of the target power adapter includes: Based on the preset power supply fault characteristics, the power supply state fault characteristics are classified and counted to obtain various fault characteristic count matrices, and the voltage fault characteristics in the fault characteristic count matrix are spatially located to obtain a voltage fault spatial distribution map, and the current fault characteristics in the fault characteristic count matrix are spatially located to obtain a current fault spatial distribution map, and the power fault characteristics in the fault characteristic count matrix are spatially located to obtain a power fault spatial distribution map, wherein the power supply fault characteristics include current fault characteristics, voltage fault characteristics and power fault characteristics; Performing fault correlation and fault feature fusion on the voltage fault spatial distribution map, the current fault spatial distribution map, and the power fault spatial distribution map to obtain a comprehensive fault feature matrix; Performing fault type determination on the comprehensive fault feature matrix to obtain fault type data, and performing fault degree evaluation on the fault type data to obtain fault degree data; The fault type data and the fault severity data are combined with fault characteristics to generate a fault detection result of the target power adapter.

8. A power adapter fault detection device, characterized in that: The fault detection device of the power adapter comprises: A feature decoupling module is used to obtain the original power electrical data of the multi-dimensional electrical parameters preset by the target power adapter, and based on the electrical characteristics corresponding to each of the electrical parameters, perform feature decoupling on the original power electrical data to obtain basic electrical feature data; A feature reconstruction module, used for performing time-frequency feature decomposition and feature reconstruction of the operating state of the basic electrical feature data to obtain power state feature data; A performance evaluation module, used to perform multi-layer electrical parameter operation characteristic association and associated operation status evaluation on the power state characteristic data to obtain a power performance evaluation result of the target power adapter, and perform state parameter mapping of fault characteristics on the power performance evaluation result to obtain a power state fault characteristic; The fault detection module is used to count the fault characteristics and locate the fault space of the power state fault characteristics, and generate the fault detection result of the target power adapter.

9. A power adapter fault detection device, characterized in that: The power adapter fault detection device comprises: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instruction in the memory to enable the power adapter fault detection device to perform each step of the power adapter fault detection method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the steps of the fault detection method of the power adapter as described in any one of claims 1-7 are implemented.

Images