2473 results about "Switching frequency" patented technology

The invention discloses a charging chip real-time power scheduling control method, and relates to the technical field of new energy automobile charging pile power scheduling, and the method comprises the following steps: S1, obtaining the operation parameters of charging equipment in real time, including the voltage fluctuation value of an input power grid, the output current of a charging interface, and the temperature of a charging chip; according to the method, the power grid voltage fluctuation, battery SOC / temperature and chip working condition multi-source data are collected in real time, a multi-target optimization model of dynamic weight distribution is combined, the service life loss of the battery is minimized, the charging efficiency and the power grid load are balanced at the same time, the user charging mode and the battery state difference are taken into consideration by adopting a priority scheduling strategy, and the charging efficiency is improved. Millisecond-level power fine control is achieved through cooperative adjustment of the PWM duty ratio and the switching frequency, personalized charging requirements are met, the overall energy efficiency is improved, a dual safety defense line is constructed through a graded load reduction protection mechanism and a fuzzy control algorithm, and a power distribution scheme is intelligently reconstructed when a power grid is abnormal or a battery is overheated.

A method for actuating a DC-to-DC converter, in particular a clocked buck-boost converter that includes at least two switch elements and an inductor or throttle and via which an input voltage is converted into a regulated output voltage, wherein the switch elements are actuated with a variable switch frequency using pulse- and frequency-modulated control signals, where the pulse- and frequency-modulated control signals are derived from a manipulated variable from a regulator, a voltage curve on the switch elements is continuously ascertained or monitored to determine the activation time of the switch elements and to start a new switch cycle, the switch elements are activated if a specified minimum period duration has been exceeded upon detecting a voltage minimum of the ascertained voltage curve on the switch elements.

The invention provides a single-tube boost PFC circuit control method and system for single-cycle control, and relates to the technical field of data processing, and the method comprises the steps: generating a self-adaptive PWM control signal according to a duty ratio adjustment amount and a frequency compensation factor, and dynamically limiting the pulse current amplitude of a power switch tube through a peak current prediction algorithm, a load current change rate is generated based on the real-time relevance between the pulse current amplitude and the load current; setting a first monitoring point in a dynamic duty ratio change path of the PWM control signal, and extracting a duty ratio change rate and a switching frequency fluctuation parameter in real time; setting a second monitoring point in a feedback path of a load current change rate, and extracting a load current slope characteristic and a transient change amplitude in real time; and based on the parameters of the first monitoring point and the second monitoring point, generating a correction factor through a dynamic response correlation model. According to the invention, a high power factor can be realized under wide input voltage and load fluctuation.

The invention relates to the technical field of power electronics, in particular to a voltage ripple suppression and compensation method for a multi-output AC-DC switching power supply, and the method comprises the steps: obtaining the real-time voltage ripple signal of each AC-DC output channel and the working parameters of the switching power supply, and the working parameters comprise the switching frequency, the duty ratio, the inductance and capacitance parameters and the load current; inputting the real-time voltage ripple signal and the working parameters of the switching power supply into a preset coupling interference model, and generating ripple coupling characteristics between AC-DC output channels; determining a ripple suppression compensation coefficient through the ripple coupling characteristics; and based on the ripple suppression compensation coefficient and a preset PWM control signal, adjusting the output voltage of the switching power supply to suppress the ripple, thereby breaking through the limitation of a traditional method on independent processing of a single channel, and fundamentally solving the problem of a ripple suppression blind area caused by cross interference between channels.

The invention provides a load current and inverter load discharge combined remote discharge method and system, and relates to the technical field of discharge control, and the method comprises the steps: collecting the parameters of a battery pack through a remote monitoring system, evaluating the health state and dischargeable capacity of the battery pack based on a fuzzy control algorithm, and calculating the optimal discharge power and duration according to a prediction correction algorithm; and generating a load and inversion discharge power curve. When the battery pack is healthy and the dischargeable capacity is sufficient, the load and the inverter discharge unit are remotely controlled, the duty ratio of a load switch and the switching frequency and output voltage of an inverter are controlled according to respective power curves, and two discharge loops are formed. And collecting discharge current and battery pack parameters in real time, and stopping discharging when the state of charge of the battery pack reaches a preset threshold value. According to the invention, by combining the load current and the inverter load for discharging and performing intelligent control based on the battery state, efficient and safe remote discharging is realized, and the service life of the battery pack is prolonged.

A system includes a first device connected to a power line and a second device connected to a circuit breaker switch upstream of the first device. A transmitter of the first device communicates a signal pertaining to amplitude, phase, frequency, switching frequency and duty cycle of current, including multiple current signatures that include an unique first current signature from the first device and at least one second current signature associated with one or more load fixtures on the power line, with a receiver of the second device. A processor in communication with the receiver processes the signal using Fast Fourier Transform (FFT) to obtain the unique first current signature and the at least one second current signature, perform line and load mapping, and generate user intelligible data representation of the confirmation and determination.

The invention discloses a storage battery capacity attenuation trend prediction method, and belongs to the technical field of storage battery prediction. By collecting voltage, current and temperature data of each monomer in real time and combining historical capacity attenuation and internal resistance growth data, the method identifies a voltage and capacity difference value, evaluates cyclic stress non-uniform distribution, and determines a current sharing proportion and a load unbalance degree. Identifying an abnormal mode of new battery overload and aged battery deep discharge, constructing a mixing abnormal working condition identification mode, if the unbalance degree exceeds the standard, adaptively adjusting the charging and discharging time and the current switching frequency, establishing a load balance control framework, predicting the capacity attenuation rate and the residual cycle index of each monomer, and determining the capacity matching degree and the life matching degree; finally, a comprehensive residual life estimation value and a credibility interval are generated through fusion; the performance balance of the mixed battery pack is remarkably improved, the overall service life is prolonged, and the method is suitable for real-time monitoring of a battery management system.

Power converters including electronic-embedded transformers for current sharing and load-independent voltage gain are described. An example power converter system includes an input, an output, a power converter between the input and output, and a controller. The converter includes a first bridge, a second bridge, and an electronic-embedded transformer between the first and second bridge. The electronic-embedded transformer includes a bidirectional coupling switch bridge. The controller generates drive control signals for quasi-trapezoidal current modulation control of the bidirectional coupling switch bridge. The controller is configured to generate the drive control signals based on a commutation coefficient k and a switching frequency for the power converter. The commutation coefficient k can be set based on an inductance in the electronic-embedded transformer, a capacitance in the bidirectional coupling switch bridge, and a margin for resonant current commutation.

The invention belongs to the field of circuits, and provides a direct-current conversion circuit for suppressing overshoot, a switching power supply and electronic equipment. The circuit comprises a ripple injection module, a conversion module, a driving module, a feedback module, a signal detection module and a signal generation module, the conversion module is connected with the feedback module, the driving module, the ripple injection module and a load, and the feedback module is connected with the driving module and the ripple injection module. The signal generation module is respectively connected with the driving module and the signal detection module, and the signal detection module is connected with the ripple injection module. The first signal is generated through the signal generation module, the frequency of the first signal is the same as the switching frequency of the direct current conversion circuit, the first signal is detected through the signal detection module, and when the frequency of the first signal is smaller than the preset frequency, the ripple injection module is controlled to stop outputting the ripple signal; the upper tube is prevented from being started in the process that the load jumps from a heavy load to a light load, and overshoot suppression of the output voltage is achieved.

The invention relates to the technical field of adaptive control, in particular to an adaptive control method and system for a power conversion system driven by a multi-mode neural network. The method comprises the following steps: extracting a high-order transient feature vector of a load path and a high-order transient feature vector of a voltage path based on a dual-path nonlinear coupling model, and jointly inputting the high-order transient feature vector and the high-order transient feature vector into a multi-modal neural network in combination with an aging index of a power supply for feature fusion to generate an initial switching frequency demand currently required by the power supply; evaluating the heat dissipation capability, inputting the heat dissipation capability into a preset heat dissipation capability three-state model to determine a heat dissipation state, performing temperature compensation and amplitude limiting processing on the initial switching frequency demand, and outputting the corrected switching frequency demand; according to the method and system, the frequency range can be dynamically adjusted according to the real-time heat dissipation capacity, overheating is avoided when heat dissipation is insufficient, efficient operation is kept when heat dissipation is sufficient, the magnetic saturation risk can be restrained in an emergency state, and frequency out-of-control is avoided.

The invention discloses a high-voltage integration and cooperative control method and system for an energy storage system, and the method comprises the steps: carrying out the three-dimensional space coupling design of a high-voltage box and a power conversion system PCS through a topological optimization algorithm according to the power level and heat distribution demands of the energy storage system, and generating a physical layer architecture scheme of a high-voltage all-in-one machine; based on a physical layer architecture scheme of the high-voltage all-in-one machine, constructing a collaborative optimization data set by using a timestamp alignment technology; inputting the collaborative optimization data set into a model predictive control framework to generate dynamically adjusted PCS working parameters; and according to the PCS working parameters, the internal temperature distribution of the high-voltage all-in-one machine is calculated in real time through the three-dimensional thermal field reconstruction model, the rotating speed of a cooling fan and the switching frequency of a power device are dynamically adjusted, and a cooperative control instruction is generated. According to the embodiment of the invention, the integration level and the cooperative control capability of the high-voltage energy storage system can be improved, and the efficient and stable operation of the energy storage system is realized.

The invention relates to a load matching optimization method of a multi-channel gate driver and related equipment thereof. The method comprises the following steps of: acquiring grid current, grid voltage and switching frequency of the multi-channel parallel grid driver to obtain driver load operation data; performing region division processing on the operation space to obtain a safe operation region and a transition region; inputting a first adaptive prediction model for feature extraction to obtain a load balancing prediction result; parallel optimization calculation is carried out on the control strategy to obtain each driving channel control parameter, and each driving channel control parameter comprises a grid driving voltage regulating variable and a driving time sequence regulating variable; performing real-time adjustment on each driving channel control parameter to obtain a load balance adjustment instruction; and performing performance index evaluation calculation to obtain a performance optimization parameter, and updating the first adaptive prediction model to obtain a second adaptive prediction model. According to the invention, through cooperative control of voltage regulation and time sequence compensation, dynamic balance of each channel load is realized.

The invention discloses a planning method and system of a power distribution communication access network, particularly relates to the technical field of power distribution communication network dynamic topology management, and is used for solving the problems of fault misjudgment and inaccurate positioning caused by topology update delay in the prior art. A dynamic topological data set is generated by dynamically monitoring the operation state of each optical fiber node and a link transmission signal, a real-time topological view with a dynamic version identifier is generated based on a preset topological planning strategy, and a network change area is quickly locked by combining a topological difference identification technology of an alarm moment and a historical view. Candidate fault links are screened through the conflict relation between link switching frequency and service priority, and the vibration characteristics of the construction machinery are further analyzed by using distributed acoustic sensing data and external force damage interference is eliminated; and finally, optical cable breakage faults are confirmed through optical power interruption verification, heartbeat signal loss detection and redundant optical fiber phase mirror symmetry verification. And multi-dimensional accurate positioning of the topological data fault root cause is realized.

The invention relates to a lightweight AI outbound model hot switching framework construction method. Defining a runtime context state unit to form an independent data structure; the AI model is divided into modules responsible for process reasoning execution; the module is in communication with the module responsible for dialogue memory and intention tracking state scheduling through a lightweight intermediate interface; establishing a set of semantic alignment mapping table based on the semantic difference between the models; performing semantic label conversion on a current dialogue state through a mapping table, and injecting a pseudo state vector into the new model to reconstruct an initial reasoning condition; calculating a tension score based on the intention hopping frequency, the slot position switching frequency and the user reply uncertainty; if the controller judges that the switching window is a safe switching window, hot switching logic can be triggered; a prediction module is introduced before switching to perform rehearsal on the reachable intention coverage of the new model, and switching is delayed if the risk is too high. The whole model hot switching process is not perceived by users, dialogue continuity and service continuity are guaranteed, the system resource utilization rate is improved, and model scheduling intelligence is enhanced.

The invention relates to the technical field of electric energy meter data analysis, and discloses a power consumption data analysis operation and maintenance method and system of an electric energy meter. The method comprises the following steps of: acquiring and preprocessing electricity utilization state data, namely synchronously acquiring voltage, current and power data, and performing timestamp alignment, data cleaning and resampling; classifying according to preset peak, flat and valley time intervals, and extracting time-frequency domain characteristics such as voltage fluctuation amplitude and current change rate in each interval to generate a set; detecting time period switching frequency, and marking an abnormal interval if the time period switching frequency exceeds a threshold value; generating an abnormal feature subset according to similarity decomposition of historical anomalies and current features; correcting the abnormal feature subset based on the electrical and time sequence parameter association relationship; and reversely backtracking a potential fault source along the fault association map, determining an operation and maintenance priority according to the number of abnormal parameters, position depth and the like, and generating an operation and maintenance instruction. The system comprises corresponding modules such as data acquisition and processing. According to the scheme, data can be accurately analyzed, a fault source is positioned, and the operation and maintenance efficiency is improved.

Disclosed in the present invention are a dual three-phase asymmetric stator-type double-permanent-magnet vernier electric motor and an improved model predictive current control method therefor. The electric motor comprises a stator core, stator permanent magnets, a rotor core and a rotor ferrite, wherein the stator core is formed by alternately arranging umbrella-shaped armature teeth and quasi-rectangular armature teeth in the circumferential direction, the stator permanent magnet, which is magnetized radially outwards, is adhered to each of openings on two sides of the umbrella-shaped armature teeth, and a fan-shaped auxiliary slot is formed between the umbrella-shaped armature teeth and the stator permanent magnet; and the shape of the rotor core is similar to a gear, the rotor ferrite is embedded in the rotor core using a Halbach-based alternating-pole array, and the length of the rotor ferrite is greater than the length of the rotor core and the length of the stator core. In addition, the present invention uses a model predictive current control system based on a virtual voltage vector to construct a cost function to select the voltage vector having the minimum switching frequency, and make same operate preferentially. The present invention not only improves the torque density of the electric motor, but also significantly reduces switching losses in model predictive control.

The present application relates to the technical field of safety management, in particular to a mine air compressor safe operation management method and system based on deep learning, comprising the following steps: collecting output pressure and input current, calculating slope change rate and direction switching frequency, clustering fluctuation amplitude and frequency by time period, generating working condition fluctuation label, identifying state mutation point and duration, calculating risk level label, analyzing load fluctuation coefficient, judging periodic fatigue early warning, combining synchronous rate to identify abnormal section, judging the need to switch state and generating instruction set. In the present application, by constructing the load risk level sequence on the basis of state duration and mutation point identification, the quantitative identification of multi-period load deviation is realized, the early warning period is marked by combining risk level and fatigue deviation ratio, and the abnormal load trend is judged according to the change direction of pressure and return gas flow, which strengthens the capture ability of abnormal operation section and improves the accuracy of operation safety risk identification and the adaptability of response strategy.

A power tool device (102) is controlled to emit sounds that can identify the power tool device and / or be used to encode data for acoustic data transmission. Modulating the switching frequency of electronic components (370) (e.g., transistors, semiconductor switched devices) of the power tool device (102) can create an acoustic signature detectable by a microphone-equipped device (104), such as a cell phone, tablet, security camera, or the like. In this way, power tool identification and / or acoustic data transmission can be realized with existing power tool device circuitry and hardware components.

The invention belongs to the technical field of charging pile control, and particularly relates to an intelligent charging pile control system and method based on dynamic power distribution, and the method comprises the steps: a slave control board collects battery data in real time, and uploads the battery data to a main control board; the main control board receives the battery data, predicts the maximum allowable charging power of the battery through an LSTM model, and determines the safety boundary of the battery according to the maximum allowable charging power and the safety operation parameters of the battery; generating a power distribution strategy by adopting a Nash equilibrium algorithm based on the battery safety boundary in combination with the power grid load and the user priority; issuing the power distribution strategy to an intelligent charging module; the intelligent charging module adjusts the switching frequency of the LLC resonant converter based on the power distribution strategy to realize target power output; the intelligent charging module monitors output current, voltage and temperature in real time and feeds back monitoring data to the main control board to form closed-loop control. And flexible control and data feedback of the charging module are realized.

The invention discloses a single-stage bidirectional power conversion topological structure capable of realizing wide-range soft switching and a control method of the single-stage bidirectional power conversion topological structure. In-bridge and inter-bridge phase shift control of a double-half-bridge circuit and a high-frequency switch matrix provides a basis for realizing soft switching of all power tubes; the switching frequency is adjusted in real time according to the magnitude of the port voltage, soft switching within a wide voltage change range is achieved, the current of a bridge arm connected with the primary side of the second transformer is the sum of the primary side current of the second transformer and the current of an inductor connected in parallel with the second transformer, and the inductance value of the inductor connected in parallel with the primary side of the second transformer is designed. The current of the bridge arm connected with the primary side of the second transformer can be effectively changed, so that the soft switching implementation range of the switching tube of the bridge arm connected with the primary side of the second transformer in a wide power transmission range can be expanded.

The invention relates to the technical field of chip design, and discloses a chip dynamic power consumption scheduling method and system based on an intelligent algorithm. The method comprises the following steps of: firstly, acquiring instruction stream data operated by a chip in real time, extracting a feature vector comprising an instruction dynamic change vector and context associated data, and determining a power consumption prediction mapping parameter according to the feature vector; and when the parameter exceeds a preset threshold value, an accurate power consumption prediction result is generated by adjusting the weight of the convolutional neural network. Subsequently, a synchronous timing demand is calculated based on the instruction switching frequency and the data dependency, and an initial power supply configuration is determined. By monitoring task load classification signals, the power consumption distribution proportion is adjusted when the signals are lower than a threshold value, the optimized power supply configuration is obtained, and the improvement index of the resource distribution efficiency is calculated according to the optimized power supply configuration. And finally, according to the index, dynamically adjusting a limiting condition of a scheduling period, and forming a self-adaptive optimization framework, thereby realizing accurate prediction and dynamic optimization scheduling of the chip power consumption.

The invention relates to the technical field of converter starting, in particular to an LLC full-bridge converter hybrid control soft start method, comprising the following steps: selecting a hybrid soft start strategy combining down-conversion starting and PWM starting, and respectively introducing nonlinear orders of a control switch frequency decline curve in the down-conversion starting process and a duty ratio rise curve in the PWM starting process; defining a mathematical formula of the change of the switching frequency and the duty ratio along with time when two nonlinear orders are different in value; establishing a simulation model, and substituting a mathematical formula that the switching frequency and the duty ratio change along with time when the two nonlinear orders are different in value into the simulation model; performing simulation testing on the nonlinear orders with different values, extracting the maximum resonance current, and determining an order combination with the best maximum resonance current suppression effect; and performing soft start of the LLC full-bridge converter based on the determined order combination. According to the invention, the current in the soft start process can be effectively suppressed, and meanwhile, quick start and device safety are considered.

The invention relates to the technical field of capacitor manufacturing and state monitoring, in particular to a capacitor aging curve optimization method based on deep learning, which comprises the following steps: step 1, multi-modal data acquisition and standardization processing; 2, modeling and training a physical constraint neural network; 3, executing a dynamic control and maintenance strategy: dynamically adjusting the switching frequency of the inverter based on a comparison result of the residual life prediction value output in the step 2 and a preset threshold value, and enabling the ripple frequency to deviate from a capacitor impedance sensitive frequency band; and according to the aging mode classification result output in the step 2, triggering a corresponding temperature sampling frequency adjustment or bus voltage reduction operation. Through fusion of multi-modal data, a physically constrained neural network model and a dynamic control strategy based on a prediction result, the aging process of the capacitor is successfully optimized, the use efficiency of the capacitor is improved, and the life cycle of the capacitor is effectively prolonged.

The invention relates to the technical field of intelligent sound channel switching, and provides an intelligent sound channel switching method for audio equipment, which comprises the following steps: S1, converting an audio signal to a preset standard sampling rate, and carrying out amplitude normalization and framing processing; extracting a time domain feature and a frequency domain feature of the audio signal, and identifying a content type of the audio signal by using a deep learning model; s2, acquiring environmental acoustic characteristics through a microphone of the audio equipment; the audio equipment performs playing time period statistics of the user, monitors a volume adjustment mode of the user, monitors a sound channel switching frequency of the user, and constructs a user behavior model based on the Bayesian theorem; s3, constructing a sound channel switching decision model, and performing weighted calculation to obtain a final decision score; when the difference value between the decision score and the switching threshold value is greater than a preset value, performing sound channel switching; and S4, realizing sound channel switching by using a high-fidelity down-mixing algorithm and an intelligent sound field reconstruction algorithm. Audio signal characteristics and a playing environment can be detected in real time, and seamless switching between a sound channel 6 and a sound channel 2 can be automatically carried out.

The present invention relates to the technical field of energy storage. Disclosed is a wired parallel operation method applied to grid connection and grid disconnection of an energy storage system. The energy storage system comprises a plurality of inverters, a communication module, an electric meter and a load. The inverters are connected in parallel, each inverter being connected to the communication module, and each inverter being connected to a battery; the electric meter is connected to one of the inverters, and the electric meter is connected to the load. The method comprises: each inverter determines whether the inverter itself is a master inverter or a slave inverter according to the communication state with the electric meter; the master inverter sends a synchronization signal to each slave inverter, such that the switching frequency and the working mode of each slave inverter are synchronized with that of the master inverter; the master inverter communicates with the electric meter to acquire load power consumption information and outputs same to the communication module; and the communication module communicates with the master inverter and the slave inverters to acquire battery information, and adjusts the battery charge and discharge power according to the load power consumption information and the battery information. The present application ensures a reasonable distribution of power among inverters, thus avoiding the phenomenon of high-frequency circulating currents.

The invention provides a transient current capability improvement control system and method of a wind and light power generation system. The system comprises a self-synchronizing voltage source control module, a junction temperature prediction module, a transient current improvement control module, a current instruction calculation module and a transient current precise control module. According to the invention, online observation of the junction temperature of the wind-solar converter power device can be realized, the maximum short-circuit current which can be output can be accurately predicted, the DC bus voltage, the switching frequency and the modulation strategy are adjusted when the power grid has a fault, the junction temperature of the power device is reduced, and the transient current endurance capability of the power device is improved. And the transient current is accurately controlled to limit the utilization of the transient capability of the power device.

The invention discloses a power adapter parameter dynamic optimization method for prolonging the service life of equipment, and relates to the technical field of adapter performance optimization, and the method comprises the steps: monitoring and recording the output voltage, current and power factor parameter changes of a power adapter in real time under static and dynamic load conditions by using a test instrument; identifying the trend and reason of abnormal parameter change; performing thermal design optimization and temperature management on the power adapter; performing over-power and over-current limitation on the pulses one by one; introducing a power factor correction circuit; optimizing the switching frequency and the design of an output filter; a load line adjusting circuit is introduced, and space is reserved for dynamic ripples when output current jumps; and evaluating whether the optimization effect reaches an expected target or not. By introducing the power factor correction circuit and the load line adjustment circuit, the power factor of the power adapter is improved, the harmonic pollution is reduced, the stability of the output voltage of the power adapter is improved, and the performance of the power adapter under the dynamic load condition is improved.

The invention relates to the technical field of data processing, and provides a dynamic load test method and device for a chiplet, and the method comprises the steps: collecting transient heat flux data generated by the chiplet in a multi-frequency load switching process to generate a thermal dynamic behavior map, calculating a thermal coupling interference index map according to the thermal dynamic behavior map, and carrying out the testing of the dynamic load of the chiplet according to the thermal coupling interference index map. And inputting the thermal coupling interference index diagram into a preset thermal stability mapping matrix, and generating a small chip dynamic load behavior evaluation parameter group. By inputting the transient heat flux data into the preset thermal stability mapping matrix to generate the small chip dynamic load behavior evaluation parameter group, the thermal stability and the overall performance of the small chip are improved, and the problem that in practical application, the prior art depends on single thermal monitoring data, and thus the reliability of the small chip is improved is solved. And when the load switching frequency is relatively high or the load disturbance is relatively large, the potential thermal coupling interference and the influence of the potential thermal coupling interference on the system performance cannot be accurately identified.

The invention relates to the technical field of inverter pulse width modulation, in particular to a neutral-point potential balance and low common-mode voltage modulation method and system based on an SVPWM inverter, and the method comprises the steps: determining a basic voltage vector based on different switching states of an NPC three-level inverter; respectively determining the midpoint current of each fundamental voltage vector, and determining the fundamental voltage vector causing midpoint potential deviation; combining with a large vector and a zero vector, and performing mode division of a single vector space region; determining an expected current according to the DC side capacitor voltage of the inverter, and constructing a target function based on the expected current; and performing region judgment according to the output synthesized voltage vector, determining a corresponding optimal division mode, obtaining each basic voltage vector of the synthesized voltage vector and a corresponding duty ratio, and determining a switching sequence to obtain a PWM wave. While the neutral-point potential balance and common-mode voltage suppression of the NPC three-level inverter are realized, the switching frequency of the inverter is effectively reduced, and the switching loss is reduced.

A multi-phase resonant power converter includes: a power stage on a primary side of the multi-phase resonant power converter, the power stage including bridge converter legs each configured to receive an AC input voltage and implement a separate phase of the power converter; a transformer device having a primary side winding for each bridge converter leg and a secondary side winding for each primary side winding; a power circuit electrically connected to the secondary side windings on a secondary side of the multi-phase resonant power converter; a primary-side controller configured to operate the bridge converter legs at a switching frequency; and a separate resonant tank electrically connected to a midpoint of each bridge converter leg. Each resonant tank includes a resonant capacitor in series with the primary side winding for that bridge converter leg. A resonant frequency of each resonant tank is tuned to within + / −50% of the switching frequency.