51 results about "Coupling (electronics)" patented technology

The invention discloses a photovoltaic fault active compensation control system and method, and relates to the technical field of power electronics and new energy, and the system comprises a data collection module which is used for collecting a three-phase grid-connected current signal of a photovoltaic system at a common coupling point in real time; the fault feature extraction module is used for performing multi-scale decomposition and spatial-temporal feature fusion based on the three-phase grid-connected current signal to obtain a deep fusion feature; the quantization factor generation module is used for converting the deep fusion features into quantization factors representing fault dynamic influences; the active compensation control module is used for generating a compensation control instruction through a reinforcement learning algorithm according to the quantization factor and the power grid state information; and the compensation execution module is used for receiving and executing the compensation control instruction, driving the grid-connected inverter to inject compensation current and actively counteracting fault disturbance. And the stability and the electric energy quality of the high-proportion new energy power system are improved.

The utility model discloses an MOS tube hard switch control surge suppression circuit, which belongs to the technical field of power electronics, and comprises a control circuit, a drive circuit and a power switch circuit, the power switch circuit is composed of a high-power suppression resistor, a first MOS tube and a second MOS tube, the first MOS tube and the second MOS tube are connected in parallel, the source electrode of the MOS tube is connected with the positive electrode of the output power supply through the high-power suppression resistor, and the drain electrode of the MOS tube is connected with the negative electrode of the output power supply. Compared with the prior art, the control circuit of the circuit can automatically detect the voltage and change the conducting state of the triode based on the change of the voltage so as to control the on-off state of the MOS tube, when the voltage is too large, the high-power suppression resistor will intervene in the circuit to consume the voltage spike, it is ensured that the rear-end power supply voltage is stable, and the purpose of protecting rear-end equipment is achieved.

The system 10 comprises an electronics module 100 removably mechanically coupled to a 5 wearable article 200. The electronics module 100 forms a communicative connection with the wearable article 200 to receive measurement data from the wearable article 200. A processor analyses measurement data obtained when the electronics module 100 is mechanically coupled with the wearable article 200 to determine whether an anomaly condition is present. The anomaly condition is indicative of an anomaly in the communicative connection between the electronics module 100 and the wearable article 200. If the anomaly condition is present, the processor triggers the generation of a notification to prompt the user to adjust the mechanical coupling between the electronics module 100 and the wearable article 200 so as to adjust the communicative connection between the electronics module 100 and the wearable article 200.

The application belongs to the technical field of microelectronics, discloses a micromechanical resonator and a preparation method thereof, and improves the Q value of the micromechanical resonator Q The application provides the micromechanical resonator, which comprises a resonant unit and a coupling block; the coupling block is used for realizing mechanical energy transmission between adjacent resonant units; at least one coupling block contained in the micromechanical resonator is provided with a thermal resistance enhancement structure; the thermal resistance enhancement structure is an etching groove formed in the coupling block, or the thermal resistance enhancement structure is composed of the etching groove formed in the coupling block and a low-thermal-conductivity material located in the etching groove; the thermal conductivity of the low-thermal-conductivity material is less than that of a main material of the coupling block. Q The application can effectively improve the Q value of the micromechanical resonator without affecting mechanical coupling between the resonant units.

The present disclosure provides a fluid detection system that includes a pipe coupling member having a perforated portion having one or more openings formed in the perforated portion. A conductive fluid sensor is disposed over the openings and an electronics module is configured to apply a charging voltage to the fluid sensor and measure a discharge time of a capacitance when fluid couples the fluid sensor to a pipe.

Disclosed is a system comprising an electronics module removably mechanically coupled to a wearable article. The electronics module forms a communicative connection with the wearable article to receive measurement data from the wearable article. A processor analyses measurement data obtained when the electronics module is mechanically coupled with the wearable article to determine whether an anomaly condition is present. The anomaly condition is indicative of an anomaly in the communicative connection between the electronics module and the wearable article. If the anomaly condition is present, the processor triggers the generation of a notification to prompt the user to adjust the mechanical coupling between the electronics module and the wearable article so as to adjust the communicative connection between the electronics module and the wearable article.

This invention proposes a data-driven back-end parameter identification method for underwater wireless power transfer systems, belonging to the fields of wireless power transfer technology and power electronics technology. By employing a data-driven method based on the XGBoost data-driven algorithm, a nonlinear complex mapping model between voltage parameters and coupling parameters is established. At the back end, the multivariate parameters to be identified in the coupling mechanism are inverted based on the Buck input voltage. This method only requires DC voltage acquisition to achieve online identification of coupling parameters, eliminating the need for bilateral communication modules or high-frequency sampling modules, thus reducing hardware costs. It enables rapid identification of multivariate coupling parameters in underwater magnetically coupled wireless power supply systems, exhibiting better adaptability to marine environments. This method solves the problems of existing underwater wireless power transfer systems relying on complex analytical models and requiring bilateral communication and acquisition for parameter identification, while also addressing the difficulty of rapid online identification of coupling parameters in complex underwater environments such as seawater.

The utility model discloses a high-coupling structure of a single-path transformer, which belongs to the technical field of power electronics and is characterized in that primary windings and secondary windings of the transformer are wound by enameled wires or Litz wires in a layered manner, and in each layer, the primary windings and the secondary windings are respectively wound on a magnetic core; a primary winding wire coil and a secondary winding wire coil which are formed after coiling are closely adjacent to each other up and down; one or more copper ring sheets are arranged between the layers; the primary windings between the adjacent layers adopt a parallel structure, the secondary windings between the adjacent layers adopt a parallel structure, and the primary windings and the secondary windings between the adjacent layers are alternately stacked. Compared with the prior art, the primary and the secondary are crossed and stacked in the scheme, so that not only can full coupling be realized, but also the peripheral size can be remarkably reduced. In addition, the copper ring sheets are smaller in size and occupied space, so that the transformer is more compact, the primary and secondary coupling degree is improved, the copper strips have better heat dissipation performance, the secondary temperature rise of the transformer can be effectively reduced, and the transformer is very suitable for a narrow space.

A coupling tool (400) for coupling together an electronics unit and a base unit of a wearable device for continuous analyte monitoring, such as continuous glucose monitoring, includes a carrier (404) comprising a receiving feature and a carrier retention device (632) configured to retain an electronics unit in proximity to the receiving feature. The coupling tool also includes an activator (402) comprising a first component at least partially receivable in the receiving feature and a contact component (514) configured to release the electronics unit from the carrier retention device in response to movement of the activator relative to the carrier. The coupling tool is in a locked configuration when the carrier retention device (632) is configured to retain the electronics unit and in an unlocked configuration when the carrier retention device (632) is configured to release the electronics unit from the carrier retention device (632). Other embodiments and methods are also disclosed. Because the wearable device and associated electronics unit can be very small so as not to interfere with the movement of the user or annoy the user, the coupling tool (400) can facilitate manually coupling together the electronics unit and the base unit.

The invention discloses a direct current reactor and a frequency converter comprising the same, and belongs to the technical field of power electronics. The direct-current reactor comprises a magnetic core module, a sleeving framework, a winding, a support and a plugging assembly. The magnetic core module adopts a box-type integrated magnetic circuit and is composed of two middle blocks, four end side blocks, four side middle plates and two U-shaped blocks which are buckled up and down, so that a highly-symmetrical and closed double-magnetic loop is formed. The invention further provides and limits four collaborative optimization core design parameters, namely a winding coupling degree control parameter kappa, a magnetic circuit sectional area continuity ratio eta, a vertical heat dissipation path optimization ratio lambda and a structural rigidity and shielding parameter xi. Through the parametric design, collaborative optimization is achieved on a magnetic circuit, a thermal circuit and a mechanical structure, and the problems that an existing integrated direct-current reactor is large in size, poor in heat dissipation, poor in performance consistency and the like are effectively solved.

A technique facilitates collecting data, related to a fracturing, run-in-hole, or pull-out-of hole operation. A frac plug is provided with electronics for obtaining the desired information related to the operation. For example, the frac plug may be constructed with electronic sensors, a digital storage device, and a power supply or other power related equipment. Depending on the application, the frac plug may be a composite frac plug, degradable frac plug, dummy frac plug, or other suitable frac plug. Retrieval of the information obtained may be done via coupling with an associated device, e.g. inductive or physical coupling, by utilizing a broadcast transmitter / receiver, by physical retrieval of the storage device, or by physical retrieval of the dummy frac plug.

The application discloses a kind of double DC bus power decoupling type hybrid power system structure, belong to the technical field of hybrid power system in power electronics and new energy technology.Existing technology, traditional structure exists steady power and fluctuation / pulse power coupling, super capacitor response speed is slow, battery has the problem of impact current.The application proposes a kind of double DC bus power decoupling type hybrid power system structure, the structure is provided with two DC buses, one is connected with battery and steady load, the other is connected with super capacitor and high fluctuation / pulse load, and two buses realize energy interflow and balance by bidirectional DC converter.Battery and super capacitor have independent voltage node, and steady power and fluctuation power are separated in space.Super capacitor is directly connected with fluctuation load without interface converter, and response speed is extremely fast.Only one bidirectional DC-DC converter can realize the collaborative operation control of hybrid power system, and hardware cost and control freedom are well balanced.

This invention belongs to the field of automotive electronics reliability testing technology, and relates to a method and apparatus for accelerated life testing of controllers based on multi-stress coupling. The method includes: selecting an on-board electronic controller sample and performing performance testing; determining the core failure mechanism and stress limit boundary; constructing a multi-stress coupling life conversion model; designing a cyclic accelerated test profile; setting the test condition parameters of the on-board electronic controller sample; conducting accelerated life testing based on the cyclic accelerated test profile; determining the failure state of the on-board electronic controller sample according to multi-level failure criteria; solving the life conversion model using the failure data corresponding to the failure state; inputting the target operating condition parameters into the solved life conversion model; and calculating the corresponding actual service life according to a preset reliability target. This invention addresses industry pain points such as low life prediction accuracy, incomplete failure identification, and high verification costs for complex on-board electronic systems like smart cockpits throughout their entire lifecycle.

The application discloses a switching power supply protection control method and system, and relates to the technical field of power electronics. The method comprises the following steps: collecting a primary side current signal and preprocessing the same to obtain a current waveform purification data set; performing shape enhancement and joint time-frequency analysis on the current waveform purification data set to obtain time-domain shape features and frequency-domain energy features, and performing nonlinear fusion through feature coupling mapping to generate a current waveform multi-dimensional feature vector frame; performing a multi-stage energy release operation according to a residual energy state set to dynamically adjust a controlled energy release path; and when it is monitored again that residual energy parameters are released to a safety threshold and output voltage is stable, generating a switching power supply protection control report. Through the generation of the current waveform multi-dimensional feature vector frame, the application realizes the deep fusion and high-dimensional representation of the time-domain shape and frequency-domain energy features in the switching current waveform, and improves the integrity and separability of fault feature information.

The invention discloses an optimal control method for improving the power tracking precision of a hybrid synchronous converter, and belongs to the field of power electronic technology and power system control. The invention provides an improved control strategy based on a series high-pass filter in order to solve the problem of low active power tracking precision caused by steady-state coupling of a voltage synchronization loop in existing hybrid synchronization control. According to the invention, the high-pass filter is connected in series in the voltage synchronization loop, continuous interference of a steady-state power angle to the frequency is blocked by utilizing the direct current blocking and alternating current alternating current characteristics of the high-pass filter, and the power synchronization loop independently dominates the steady-state frequency, so that the steady-state power error is thoroughly eliminated, and zero-static-error tracking is realized. And meanwhile, the response capability of the voltage synchronization loop to transient high-frequency disturbance is reserved, so that the excellent transient damping characteristic of the hybrid synchronization control strategy still exists. In addition, the method is simple in structure, does not need to depend on complex system parameter identification, and has high robustness and engineering adaptability.

This invention discloses a three-phase power grid phase sequence identification system and method, relating to the field of power electronics technology. The system includes a three-phase power grid, at least one AC source electrically coupled to a single phase of the power grid, at least one AC load electrically coupled to a single phase of the power grid, and a power acquisition device. The power acquisition device is communicatively connected to the AC source. This invention defines the phase sequence of the three-phase power grid based on the phase sequence of the power acquisition device. By controlling the output power of each AC source to change abruptly, it synchronously detects the corresponding changes in the three-phase power grid electrical signals and automatically determines the electrical coupling phase sequence of each AC source. This invention requires no additional hardware costs; automatic phase sequence identification can be achieved solely through software logic. This significantly reduces the construction and commissioning difficulty of three-phase AC systems such as photovoltaic power generation systems, solves the problems of time-consuming, labor-intensive, and error-prone manual phase sequence checks, and provides a reliable phase basis for phase-by-phase anti-reverse current control.

The utility model discloses a dual-channel nuclear pulse acquisition daughter card based on a JESD204B interface, which adopts a dual-channel high-speed ADC (analog-to-digital converter) chip based on the JESD204B interface as a core module of hardware design, and selects a universal MCX connector in the industry as an input interface of two paths of analog pulse signals. Wherein one path of pulse signal enters the ADC for sampling through the alternating current coupling transformer driving module, the other path of pulse signal enters the ADC for sampling through the direct current coupling amplifier driving module, and a high-integration positive and negative power supply module is designed to provide working voltage for a hardware system, so that nuclear electronics data acquisition with two channels, 14-bit sampling precision and 1 GHz sampling rate is realized; and the nuclear pulse signal data acquisition capability of a nuclear electronics system can be greatly improved.

The invention provides an automobile pressure sensor semiconductor device based on an MEMS technology. The method belongs to the technical field of crossing of a micro electro mechanical system (MEMS) technology and intelligent automobile electronics. The device comprises a three-dimensional heterogeneous sensitive structure, a thermal activation type vacuum reference cavity, an on-chip intelligent diagnosis engine, a signal conditioning circuit module, a multi-source stress isolation layer and an anti-interference packaging shell, and is characterized in that the three-dimensional heterogeneous sensitive structure is used for simultaneously sensing coupling stress; the signal output end of the microprocessor is connected with the signal input end of the signal conditioning circuit module; the thermal activation type vacuum reference cavity is in thermal coupling connection with the three-dimensional heterogeneous sensitive structure and is used for providing a stable vacuum reference datum to eliminate environmental drift; through the three-dimensional heterogeneous sensitive structure, the pressure, the shearing force and the bending moment in the vertical direction can be synchronously and independently sensed, the comprehensive measurement capability of complex coupling stress is remarkably improved, and richer mechanical dimension information is provided for chassis control systems, battery safety systems and the like.

The invention provides a three-dimensional nonlinear magnetic material equivalent circuit model and a test method, and is applied to the technical field of power electronics. According to the model, a current source module applies bias current and excitation current to a nonlinear coupling packaging module; a plurality of node ports of the nonlinear coupling packaging module are connected with physical coupling nodes of a plurality of nonlinear coupling devices; the mapping sub-module maps the plurality of physical coupling nodes and the plurality of nonlinear coupling devices, constructs an analog nonlinear coupling circuit according to a connection coupling relation between the linear damping module and the plurality of nonlinear coupling devices, and responds to an applied bias current and an excitation current, so that the nonlinear coupling circuit is simulated. The analog nonlinear inductance unit is coupled to generate an analog induced voltage, and the analog induced voltage is applied to the linear damping module and other analog nonlinear devices through a plurality of node ports to reproduce a magnetic domain rotation coupling effect and an anisotropic hysteresis effect; the linear damping module responds to the analog induced voltage and reproduces a damping effect in combination with the analog nonlinear resistor unit.

This invention discloses a GFM-VSG system suitable for grid-type virtual synchronous machines, relating to the fields of power electronics and new energy power generation technology. The system aims to solve problems such as inaccurate stability assessment and unclear oscillation propagation mechanisms in grid-type virtual synchronous machines (VSGs) under weak grid conditions due to grid strength disturbances and internal coupling effects. The system constructs a unified frequency-voltage coupling model, introduces a virtual inertia and damping dynamic response calculation module, derives the coupling transfer function set using small-signal linearization methods, and defines the propagation gain M. X ᵧ serves as a quantitative criterion for oscillation propagation effects, forming a complete stability analysis framework. This system can quantitatively assess the impact of grid intensity disturbances on the output electrical quantities of the VSG, accurately identify the main risk sources of oscillation propagation, and provide a theoretical basis for the optimal design of controller parameters.

Embodiments of the present application relate to the technical field of electronics, and provide a chip and a preparation method therefor, a package substrate and a preparation method therefor, a chip package structure, a terminal device, and a communication system, for use in improving signal transmission performance in the chip package structure. The chip comprises a first dielectric layer, a first capacitor plate, and a second capacitor plate, wherein the second capacitor plate, the first capacitor plate, and the first dielectric layer located between the second capacitor plate and the first capacitor plate form a coupling capacitor. The capacitor formed in the chip can mitigate the problem of unequal return ground voltages between the chip and other chips caused by common‑mode matching requirements. Therefore, when the chip and the other chips are co‑packaged, a signal return structure in the package substrate may be implemented as a full‑layer structure, and a signal line does not need to cross splits in the package substrate, thereby improving signal transmission performance.

This invention provides a multi-physics coupling test method and system for intelligent vehicle hybrid network architecture, belonging to the field of automotive electronics testing technology. The method includes: applying single-physics coupling, dual-physics coupling, and system-level multi-physics coupling loads through a tiered testing process, simultaneously acquiring electrical signal, optical signal performance parameters, and environmental parameters; calculating a unified signal health index (HI) and degradation rate based on a preset algorithm; analyzing the contribution of each physical field factor to performance degradation when the HI is below a health threshold or the degradation rate exceeds a rate threshold, and identifying the dominant failure factor; and intelligently matching the corresponding optimization strategy from a pre-built optimization strategy library. This invention solves the problems of media isolation, data fragmentation, and inefficient optimization in traditional testing methods, achieving unified cross-media testing, real-time health status assessment, and precise automated optimization, thus improving the reliability verification efficiency of intelligent vehicle hybrid network architecture.

The invention provides a loss junction temperature prediction method of a power module, which relates to the technical field of power electronics, and comprises the following steps: reading a working condition file containing voltage, current, temperature ratio and resistance, and importing characteristic parameters of a power module data manual; based on the characteristic parameters, using an interpolation algorithm to construct a loss and thermal resistance model dynamically changing along with temperature, current and resistance; calculating a pulse sequence in a current half cycle, and calculating conduction loss and switching loss corresponding to each pulse through an accumulation method to obtain total loss; and calculating an initial junction temperature based on the loss and thermal resistance model, and feeding back the initial junction temperature to a loss calculation process for loop iteration until the junction temperature difference value of two successive calculations is smaller than a preset threshold value, and taking the last calculation result as the highest junction temperature for life prediction. The method has the beneficial effects that the pulse loss is accurately calculated through an accumulation method, a multi-dimensional dynamic model is constructed, thermoelectric coupling feedback iterative logic is introduced, and efficient and accurate junction temperature calculation under mass working conditions is realized.

The present application relates to the field of power electronics, specifically to an active damping control method and system for LC network resonance suppression, comprising: detecting the current instantaneous value flowing through the passive damping resistor R in real time to obtain a current detection signal; and obtaining the voltage at the common coupling point of the LC filter network output end; identifying the resonance degree of the current LC filter network based on the frequency spectrum characteristic parameter, and determining the identification result; dynamically adjusting the adjustment coefficient K according to the identification result; calculating the target compensation current according to the real-time detected current detection signal and the adjusted adjustment coefficient K; controlling the class-D amplifier to output the compensation current and inject the compensation current into the parallel node of the passive damping resistor R; and starting the energy feedback unit when the DC bus voltage exceeds the preset feedback starting threshold voltage. By dynamically adjusting the coefficient K, online stepless adjustment of the equivalent damping resistor can be realized, and different working conditions and different intensity resonances can be adaptively coped with.

This invention belongs to the field of microelectronics technology and discloses a high-precision eight-phase voltage-controlled oscillator based on a transformer superharmonic coupling network. The oscillator includes four voltage-controlled oscillator units and a superharmonic coupling network, with each unit outputting a pair of differential oscillation signals. The superharmonic coupling network consists of two stage I coupling transformers and one stage II coupling transformer. The stage I coupling transformers are connected to the second harmonic output terminals of the first and second groups of voltage-controlled oscillator units, respectively, creating a 90° phase difference between the corresponding fundamental signals. The stage II coupling transformer is connected to the fourth harmonic coupling terminal of the stage I coupling transformer, creating a 90° phase difference between the second harmonic signals of the four units, thus ensuring that the eight fundamental output signals are sequentially 45° out of phase. This invention achieves high-precision eight-phase output by precisely controlling the fourth harmonic phase to regulate the second harmonic phase of each unit, while not interfering with the fundamental oscillation. It has advantages such as symmetrical structure, small phase error, and consistent output amplitude.

A transducer includes a diaphragm support portion having a hollow portion, a vibrating diaphragm which is displaceable in a diaphragm thickness direction and is coupled to the diaphragm support portion, and a piezoelectric element on the vibrating diaphragm, having a pair of electrodes and a piezoelectric film sandwiched by the pair of electrodes, in a region overlapping the hollow portion, a plurality of first regions having a first total film thickness which is a sum of a diaphragm thickness of the vibrating diaphragm and a film thickness of the piezoelectric element, and a plurality of second regions having a second total film thickness which is a sum of the diaphragm thickness of the vibrating diaphragm and the film thickness of the piezoelectric element and is different from the first total film thickness, the first regions and the second regions being alternately arranged, one of the first regions being adjacent to a coupling portion of the diaphragm support portion and the vibrating diaphragm.

This invention discloses a method, device, and storage medium for assessing the lifetime of MMC-HVDC IGBT modules based on multi-scale thermal cycling in the field of power electronics evaluation. It aims to solve the problems of traditional evaluation methods neglecting loss-temperature coupling and low computational efficiency. The method includes: acquiring task profile data and sequentially inputting it into pre-constructed electromagnetic operation, loss, and thermal network models; using an iterative algorithm to alternately correct power loss and junction temperature to obtain a steady-state average junction temperature sequence; equating the thermal network model to an RC circuit and analytically calculating the junction temperature fluctuation amplitude of the fundamental frequency thermal cycle; extracting low-frequency thermal cycle characteristic parameters from the steady-state average junction temperature sequence using a rainflow counting algorithm; and combining the Bayer lifetime model and Miner's linear cumulative damage theory to calculate and linearly superimpose the damage degree of the fundamental frequency and low-frequency thermal cycles to obtain the total lifetime of the IGBT module. This invention balances accuracy and efficiency and is suitable for reliability assessment under complex operating conditions.

This invention provides a method, apparatus, electronic device, and dielectric for determining the resonant parameters of a DC-DC converter, relating to the field of power electronics technology. The method includes: determining the initial electrical parameters of the power switching devices in the DC-DC converter based on preset electrical parameters; determining the initial resonant inductance value of the resonant network in the DC-DC converter based on the initial electrical parameters and preset electrical parameters; determining the optimized magnetizing inductance value corresponding to the resonant network based on the initial electrical parameters and the high-frequency resonant phase angle during the dead time; ensuring the optimized magnetizing inductance value satisfies the zero-voltage turn-on condition; and determining the target resonant parameters in the DC-DC converter that satisfy the target voltage gain based on the optimized magnetizing inductance value and the initial resonant inductance value. This invention optimizes the parameter range of the magnetizing inductance based on the coupling relationship between the magnetizing inductance and the high-frequency resonant phase angle during the dead time, ensuring that the optimized magnetizing inductance value of the resonant network can satisfy the zero-voltage turn-on condition.

The application discloses a simulation method for quantitatively analyzing the influence mechanism of a power converter on a proton exchange membrane fuel cell, and belongs to the cross technical field of fuel cells and power electronics. The core of the application is that: firstly, an electro-chemical-thermal-flow multi-physical field coupling model of the proton exchange membrane fuel cell is established, and is exported as a functional model unit file based on a functional model interface standard; secondly, a power electronic converter circuit and a control algorithm model are built, and the functional model unit is imported to construct a closed loop system; finally, the load current of the converter is transmitted to the functional model unit to drive multi-physical field calculation, and meanwhile, the functional model unit feeds back the internal state parameters of the fuel cell to the control algorithm in real time. The method breaks through the technical limitation that the control system and the multi-physical field simulation are mutually separated, and can accurately reveal the influence law of the converter topology and the control strategy on the external characteristics and the internal working state of the proton exchange membrane fuel cell from the mechanism level.

The invention discloses a Boost converter physical information identification method, and belongs to the crossing field of power electronics and artificial intelligence. Aiming at the problems of discontinuous current, differential coupling and parasitic parameter identification caused by non-minimum phase characteristics of a Boost converter, the invention provides a physical information neural network architecture fusing a switch modulation impedance network and a time sequence adaptive mask. A nonlinear state space model is constructed, and a coupling relation between a capacitor equivalent series resistor and a load is embedded by adopting an algebraic method, so that differential item nesting is eliminated, and the numerical stability is improved; meanwhile, a time sequence index mask layer is designed, automatic decoupling and joint identification of global inherent parameters and multi-section local load parameters are achieved in single training, and multi-working-condition online parameter updating is supported. According to the method, a whole set of circuit parameters including parasitic resistance can be identified from sparse sampling data with high precision, and support is provided for predictive maintenance of a power electronic system.