136 results about "Power topology" patented technology

The present invention provides a wind power off-grid hydrogen production power supply topology and control method without a step-down transformer, including: obtaining wind speed data, using a wavelet threshold algorithm for denoising, predicting power based on an LSTM-GRU hybrid neural network, and extracting power fluctuation characteristics through multi-scale entropy analysis; obtaining load data, using Kalman filtering and recursive least squares to establish an impedance model, and extracting dynamic characteristics in combination with Hilbert transform; using a differential evolution algorithm for multi-objective optimization to generate an optimal switching sequence and power allocation strategy; achieving system control through space vector modulation, balanced control, soft switch control and harmonic compensation; achieving state estimation and disturbance suppression based on a sliding mode observer and a nonlinear observer; and achieving fault diagnosis and protection using a deep belief network. The present invention improves system efficiency and reliability and improves dynamic performance.

The invention discloses a power supply fault rapid detection system and method based on an intelligent diagnosis engine, and belongs to the technical field of power electronic equipment fault diagnosis, and the system comprises a function region division module which is mainly used for dividing a plurality of function regions according to a power supply topological structure; the real-time data acquisition module is mainly used for acquiring parameter data of each functional area in a power supply operation stage; the current model building module is mainly used for building a current time sequence prediction model; the trend prediction module is mainly used for generating a future current change trend; the abnormity judgment module is mainly used for identifying an abnormal condition by performing similarity matching with real-time data; the fault positioning module is mainly used for matching with a fault mode based on an abnormal current waveform and constructing a directed graph to position a fault position; the method has the advantages that early warning and accurate positioning of the power supply fault are realized through multi-modal data fusion and a deep learning model, and the operation and maintenance efficiency and reliability of the power electronic equipment are remarkably improved.

The invention discloses a photovoltaic hydrogen production power supply topological structure based on MPPT and a control method, and relates to the technical field of renewable energy electrolysis hydrogen production, and the photovoltaic hydrogen production power supply topological structure mainly comprises a photovoltaic power module, a direct current bus, a water electrolysis hydrogen production unit, an energy storage unit and a control unit. The photovoltaic power modules are connected in parallel to the direct current bus through the independent DC / DC converter, and are matched with the automatic identification and scheduling functions of the control unit, the system operation strategy is automatically adjusted when the modules are newly added or faults exit, and plug-and-play, fault isolation and power reconstruction of the photovoltaic power modules can be supported; the duty ratio change step length of the DC / DC converter is limited when the current rapidly fluctuates due to the change of the photovoltaic output power, the current change rate of the electrolytic cell stack is effectively controlled, the service life protection of the electrolytic cell can be considered, the high modular access capability is achieved, multi-source collaborative scheduling is supported, and bus voltage stabilization and system current sharing optimization can be achieved.

The invention relates to the technical field of intelligent power grid communication, and discloses an HPLC communication-oriented power topology adaptive reconstruction and energy consumption optimization control method, which comprises the following steps of: acquiring HPLC communication characteristic data of power network nodes in real time; constructing a dynamic spatio-temporal topological graph based on the communication feature data and predicting a future topological state; generating a communication power distribution and routing path joint optimization strategy by using a multi-target reinforcement learning algorithm in combination with the predicted future topology state and the real-time energy consumption parameter; verifying the electromagnetic compatibility and protocol compliance of the optimization strategy through a digital twin model, and generating a control instruction which can be executed by equipment; and the transmitting power and the communication route of the HPLC node are dynamically adjusted through the control instruction. According to the method, the communication quality stability and the energy efficiency level are improved, topology dynamic perception, multi-target collaborative optimization and strategy security verification are realized, and efficient and reliable operation of HPLC communication in a complex power grid environment is guaranteed.

The present invention discloses a method for tracing power supply paths in important locations based on digital twins. In order to solve the problem that it is time-consuming and laborious in the prior art due to the fact of requiring manual data collection and mapping in the case where the mapping of a single object in a digital space needs to be restored according to the actual situation on the spot, the present invention hierarchically establishes digital twin scenarios, constructs a complete power supply path according to a signal flow relationship between power topology and equipment, and conducts power supply path tracing in the event of equipment failure to determine a scope of influence, thus being fast and accurate in tracing.

The invention relates to the technical field of data processing, and discloses a power sample data acquisition and classification method and system. The method comprises the following steps: synchronously acquiring active power, reactive power fluctuation and voltage harmonic data through a multi-point terminal, and constructing six types of characteristic index vector groups; a power load genetic optimization algorithm is used to optimize a classification threshold under power flow constraint; based on the optimal threshold value, clustering analysis is carried out through an EFC-KMeans algorithm in combination with impedance matrix characteristics; and inputting the clustering center into a multi-head power topology attention mechanism modeling node coupling relationship to realize power sample data classification and identification. The physical constraint conditions of the power system are effectively fused in the power sample data acquisition and classification process, so that the physical feasibility and engineering practicability of the classification result are improved.

The invention relates to the technical field of relay protection, in particular to an automatic checking method and system for a relay protection setting value of a power plant, and the method comprises the steps: obtaining relay protection data, including obtaining electrical parameters, equipment parameters and power topological structure information; constructing a topological graph model of the power network by using the power topological structure information, constructing an electrical model based on elements according to the electrical parameters and the equipment parameters, and combining the topological graph model with the electrical model by using node parameter association to obtain a power system integrated model; and performing constant value check calculation based on the power system integrated model, and generating a report according to a constant value check calculation result. Hidden layers are set for different fault types by adopting the multi-branch structure network, so that the neural network can perform specialized learning for different types of fault features, and the adaptability of the system to different fault scenes is enhanced.

The invention belongs to the technical field of charging pile power distribution, and discloses a charging pile power distribution method and system based on a depth-first algorithm, and the method comprises the steps: building a host power topology adjacency matrix as an initial adjacency matrix; obtaining a host fault state, and optimizing the initial adjacency matrix to obtain a real-time adjacency matrix; when a host fault state or charging gun power demand change is detected, each charging gun is taken as a starting point, the real-time adjacency matrix calculates each charging path and path efficiency through a depth-first algorithm, the charging paths with the path efficiency lower than a preset value are eliminated, weighting priorities of the remaining paths are calculated, and the charging paths are calculated according to the weighting priorities; selecting the charging path with the highest weighting priority as a target path; and performing cross verification on the target path and the parallel contactor path, if verification succeeds, outputting the target path, otherwise, outputting the path successfully verified last time, so that power distribution can be dynamically and flexibly adjusted according to the charging power demand changing in real time.

The invention provides a direct current optimal power flow solution-oriented data mechanism combined driving method and system, and is applied to the technical field of power data processing. Inputting the power demand into the DNN to obtain an initial solution; and if the initial solution does not meet the problem constraint, processing the problem to be solved and the initial solution by using a variable elimination method, and converting the problem to be solved and the initial solution into an optimization problem without equality constraint and a corresponding initial solution. Iteratively adjusting the position of the initial solution based on a constraint set only comprising the inequality constraint and the normal vector violating the inequality constraint until a non-feasible solution of the initial solution falls into a feasible region; and restoring the initial solution to obtain an intermediate feasible solution meeting problem constraints. And inputting the intermediate feasible solution into the FONET neural network to obtain adjustment information, and adjusting the intermediate feasible solution into a target feasible solution based on the adjustment information. The method is suitable for the DCOPF problem with a feasible solution under any power topology, the solving efficiency and the feasibility of the solving result can be guaranteed at the same time, and the obtained solving result has higher optimality.

The invention discloses an energy consumption type pulse special power supply topology circuit, particularly relates to the technical field of power supply circuits, and is technically characterized in that a forward energy storage unit is used for outputting forward current to an external load coil and providing energy required by main compression for a field inversion configuration device; the negative energy storage unit is used for outputting negative current to an external load coil and providing bias field energy for the field inversion configuration device; the charging unit is also used for charging the current source unit; the current source unit is used for providing bias current for an external load coil and creating a bias condition for the field inversion configuration device; the resonance unit is used for outputting resonance current to an external load coil and providing energy required by pre-ionization for the field inversion configuration device; the first consumption unit is used for consuming residual electric energy after the external load coil and the current source unit work; the first control unit is used for controlling short circuit of the first consumption unit; the second control unit is used for controlling the on-off of the resonance unit; and the third control unit is used for controlling the on-off of the negative energy storage unit.

The invention relates to the technical field of electric power emergency control, and discloses an electric power facility emergency processing system based on earthquake early warning, and the system comprises a region division module which is used for obtaining the data of an electric power topological structure, and generating a plurality of region sections; the data acquisition module is used for acquiring earthquake early warning data, equipment operation data and equipment environment data and representing the data through a multi-source data set; the fault risk assessment module is used for calculating the failure probability of each power device; the isolation decision and control module is used for calculating a region reliability index of the sub-region and determining a target sub-region; the off-grid operation regulation and control module is used for switching the target sub-region to off-grid operation and adjusting the control parameters; the operation state evaluation module is used for generating an operation health score; and the reconnection recovery module is used for recovering the connection of the target sub-region to the main power grid. According to the invention, intelligent isolation and safe reconnection control of the power subareas under earthquake early warning driving are realized.

The invention relates to the technical field of new energy automobile charging, and discloses a charging pile gun address dynamic allocation method and system based on power topology identification. A host control module, a terminal equipment module, a charging gun access module, a power transmission detection module, a network communication interaction module, an address intelligent distribution module, a fault diagnosis isolation module and a data persistent storage module are established; the host control module is connected with the charging gun access module, the power transmission detection module and other subsequent function modules through power lines, each module is correspondingly connected with a terminal and a charging gun, a system physical power topology and a multi-terminal network communication topology are constructed based on a hardware connection relationship, and the host and the terminal finish power-on initialization respectively. And the host starts an address allocation preparation process, and the terminal starts network initialization and address request preparation, so that a plug-and-play function is realized, the deployment efficiency and operation and maintenance reliability of the charging pile are improved, and finally the problems of low efficiency, easy error and complex after-sales maintenance of traditional manual address configuration are solved.

The invention relates to the technical field of power systems and intelligent power grids, and discloses an automatic monitoring method and system for a booster station of a wind power plant, and the method comprises the steps: collecting the operation data of a main transformer, a circuit breaker and a bus through arranging edge equipment in a power grid system of the booster station of the wind power plant, and constructing a power topological graph; constructing a graph neural network anomaly detection model, and outputting a node anomaly probability and an edge anomaly probability through node embedding aggregation and edge weight prediction; determining the importance score of each node based on the node influence factor of the topological graph; sorting all the nodes from high to low according to the importance scores, and selecting the m nodes before sorting to participate in each round of federated learning until the loss function of the neural network anomaly detection model is smaller than a set threshold value; and monitoring the operation state of the booster station of the wind power plant in real time by using a graph neural network anomaly detection model, and triggering a multi-stage alarm signal. And the dynamic adaptability and monitoring precision of wind power plant booster station monitoring are improved.

The invention relates to a high-frequency high-voltage power supply topological structure, which comprises a cascade multi-level module, a charging power supply module, an isolation high-frequency alternating current module and an isolation direct current module, and is characterized in that a series circuit of the charging power supply module and the isolation high-frequency alternating current module is connected to the head alternating current end and the tail alternating current end of the cascade multi-level module; a series circuit of the isolation direct current module and the equipment to be tested is connected to the head and tail alternating current ends of the cascade multi-level module; according to the invention, the structure is simple, high-frequency and high-voltage output can be realized by only needing power electronic devices and constructing a power frequency power supply, an inductor, a resistor and a capacitor, the voltage amplitude and frequency output range is wide, high-frequency and high-voltage alternating current can be output, the voltage endurance capability of a high-voltage to-be-tested product can be tested, the topological structure is simple, the real-time operation is convenient, and the cost is low.

The invention belongs to the technical field of voltage transformation, and particularly relates to a multistage power topology self-adaptive temperature control method and a voltage transformation system.The method comprises the steps that first temperature data are generated by collecting the real-time temperature of target equipment, and the power topology level where a current load is located is divided according to the first temperature data; generating a corresponding voltage adjusting instruction based on the level, and generating a control signal matched with the input characteristic to drive topology switching; after the voltage is adjusted, second temperature data are continuously collected, a temperature compensation value is calculated through comparative analysis and used for updating the voltage instruction, and closed-loop control is formed; the dynamic linkage control between the voltage output and the temperature change is realized, the response capability of the system to the load fluctuation and the thermal management precision are improved, and the problem of system instability caused by untimely voltage regulation or temperature control lag in the prior art is effectively solved, so that the overall operation efficiency and reliability are improved.

The invention provides a dynamic power distribution method, system and device and a medium, and the method comprises the steps: generating a path comprehensive state table between a parking space and a power module according to the charging condition and the power topology state of each parking space in a current station; according to the real-time state of each power module in the power topology state and the charging condition of each parking space, performing priority ranking on the charging demand of each parking space to obtain a parking space distribution ranking result; therefore, a power module distribution mode of each parking space can be generated by using a charging power dynamic distribution algorithm based on the charging condition of each parking space, the path comprehensive state table and the parking space distribution sorting result, and power distribution is performed on the parking space having a charging demand according to the power module distribution mode of each parking space. The charging requirement of the required parking space can be met on different power topologies, and the flexibility of the charging pile system can be improved to the maximum extent.

The invention provides a high-voltage cable sheath grounding fault diagnosis method and device and electronic equipment, and relates to the technical field of electrical fault diagnosis and detection. The method comprises the following steps: acquiring a multi-source state parameter, wherein the multi-source state parameter is used for representing a real-time operation state of a sheath grounding system corresponding to a high-voltage cable sheath; performing multi-source heterogeneous data processing on the multi-source state parameter to obtain a fundamental component and a harmonic component of the sheath current and a vibration energy value of the cable joint; determining a current fault type based on the multi-source state parameter, the fundamental component, the harmonic component and the vibration energy value; determining fault section position information based on the multi-source state parameters and the electric power topological graph of the high-voltage cable sheath; and determining a fault diagnosis result based on the current fault type and the fault section position information. The fault diagnosis method and device are used for improving the comprehensiveness and accuracy of fault diagnosis, reducing the possibility of misjudgment and missed judgment, and being convenient for maintenance personnel to quickly position and solve the fault problem so as to guarantee stable operation of a power system.

The invention discloses an energy feedback type pulse special power supply topology circuit, particularly relates to the technical field of power supply circuits, and is technically characterized in that a forward energy storage unit is used for outputting forward current to an external load coil and providing energy required by main compression for a field inversion configuration device; the current source unit is used for receiving and storing electric energy, providing bias current for an external load coil and creating bias conditions for the field inversion configuration device; the negative energy storage unit is used for outputting negative current to an external load coil and providing bias field energy for the field inversion configuration device; the charging unit is also used for charging the current source unit; the resonance unit is used for outputting resonance current to an external load coil and providing energy required by pre-ionization for the field inversion configuration device; the first bidirectional combination switch is used for controlling the on-off of the resonance unit; and the second bidirectional combination switch and the third bidirectional combination switch are used for controlling the positive on-off and negative on-off of the circuit.

The invention provides an ultra-wide voltage regulation stable output circuit which comprises a BUCK power supply module and a main power supply module. The main power supply module comprises a DCDC deformation control unit and a transformer output unit, the transformer output unit comprises at least two transformers, and the transformers are connected in a manner that primary sides are connected in parallel and multiple sides are connected in series, so that the output voltage of a transformer sub-circuit is adjusted according to a PWM2 signal. The ultra-wide voltage regulation stable output circuit has an ultra-wide voltage regulation range, stable output can be realized, a two-stage power supply topological structure is adopted, bus voltage reduction or direct connection can be realized by utilizing the first-stage BUCK power supply module, and the output voltage of the bus can be reduced by utilizing the second-stage BUCK power supply module. And the second-stage main power supply module is utilized to realize adjustable output in a larger range and stable output, so that the applicability and reliability of the circuit can be effectively improved, and the use requirements of different power supplies are met.

The invention discloses a mobile photovoltaic energy storage charging system based on a distributed direct current coupling architecture, belongs to the technical field of photovoltaic energy storage charging equipment, and aims to solve the problems of relatively high investment cost and relatively low energy conversion efficiency caused by relatively many component controllers and couplings. Comprising a mobile distributed DC coupling architecture module and an all-in-one controller. A distributed direct current coupling framework is adopted, on a mobile energy storage charging vehicle or a fixed energy storage charging pile, by multiplexing a DCDC power module and through algorithm switching, photovoltaic charging of an energy storage battery and charging of a new energy vehicle by a mobile charging trolley are achieved, and meanwhile charging or discharging of a micro-grid can be achieved by matching with an energy storage inverter; the main hardware power topology of the photovoltaic MPPT controller is a bidirectional Boost topology, a Buck-Boost topology or a bidirectional DAB topology, and the off-grid mobile energy storage charging core power module can realize wide voltage output through the Buck-Boost by means of a DCDC converter, so that the charging of most new energy batteries on the market is met.

Embodiments of the present application provide a thermal failure processing circuit, a thermal failure processing method, an electronic device, a medium and a product. The thermal failure processing circuit comprises a power topology module, a signal processing module and a control module, wherein the power topology module comprises a first transistor and a second transistor, the control module is configured to control the first transistor and the second transistor to be turned on, a first detection unit is configured to collect a first voltage corresponding to a second end of the first transistor and send the first voltage to the control module, a second detection unit is configured to collect a second voltage corresponding to a second end of the second transistor and send the second voltage to the control module, and the control module is configured to perform thermal failure detection on the first transistor and / or the second transistor based on the first voltage and / or the second voltage, and start a preset processing mechanism when the first transistor and / or the second transistor is thermally failed. Based on the thermal failure processing circuit provided by the present application, the stability and reliability of the system can be improved.

The invention relates to the technical field of electric power early warning, and provides an electric power system multi-mode fault intelligent early warning method and system based on edge computing, and the method comprises the steps: collecting the operation state information of each electric power node based on edge computing equipment disposed at each electric power node in an electric power system; performing operation state analysis according to the operation state information of each power node to obtain abnormal fault nodes in the power system, and performing fault type analysis based on the operation state information of each abnormal fault node to obtain an abnormal fault result of each abnormal fault node; performing propagation path traversal in the power topology according to a fault traversal mode corresponding to the abnormal fault result of each abnormal fault node, and determining a target fault area; and generating fault early warning of a corresponding level based on the area range of the target fault area. The safe and stable operation of the power system is improved.

The invention relates to a direct-current side rapid discharging technology, and discloses a converter direct-current side rapid discharging method, which comprises the following steps of: during boosting, monitoring a voltage value of a direct-current side capacitor in real time, comparing the voltage value with a preset target voltage, and when the detected capacitor voltage is lower than a set value, adjusting the output of a power supply circuit and increasing a charging current by a control system; on the contrary, when the voltage of the capacitor is close to the set value, the control system can gradually reduce the charging current, so that stable boosting is achieved, and voltage overshoot is avoided. According to the converter direct-current side rapid discharging method, the circuit can be actively disconnected, so that the influence on a power grid is cut off, the problem of direct-current side capacitor discharging existing in an existing APF or SVG hardware power topology is solved, and the converter direct-current side rapid discharging method has very high practical value, so that under the condition that the cost is not increased too much, the direct-current side rapid discharging of the converter is achieved through a local inversion method. And the voltage of the direct current side capacitor can be rapidly reduced, so that the safety of maintenance personnel is protected, and the practicability is very high.

The utility model discloses an improved topological structure based on BOOST and LLC power topology, which comprises a switching circuit, a full-bridge LLC bridge arm and a main switching tube which are connected in parallel with the input end of the switching circuit, and a first inductor connected with the full-bridge LLC bridge arm, and the output end of the switching circuit is connected with a transformer through a first filter capacitor and a second inductor. The secondary side of the transformer is connected in parallel with a switching tube assembly and a third filter capacitor; according to the utility model, the efficiency upper limit of the two-stage power topology architecture can be improved, the output power of the BOOST circuit is reduced by half, the use of power devices can be reduced, the cost is saved, the weight and the size of an inductor can be effectively reduced, the weight and the size of a DC / DC converter can be obviously reduced, the gain of a full-bridge LLC resonant cavity is ensured, and the output power of the two-stage power topology architecture is improved. And the stability of the output voltage of the DC / DC converter under the working condition of low input voltage is ensured.

The present invention relates to a mutual inductance type switched reluctance motor power topology with an auxiliary excitation winding, comprising: a DC power supply, a power supply filter capacitor connected in parallel with the DC power supply, a plurality of power switching tubes connected to the DC power supply, a plurality of diodes connected to each power switching tube, a plurality of auxiliary excitation windings connected between the power switching tubes, and a three-phase mutual inductance type switched reluctance motor connected between the power switching tubes. The present invention can control the working mode of the auxiliary excitation winding by switching on and off, exciting or cutting off the faulty armature structure, improving the output torque of the existing mutual inductance type switched reluctance motor while ensuring the reliability of system operation.

The present invention discloses a power supply fault rapid detection system and method based on an intelligent diagnostic engine, which belongs to the technical field of fault diagnosis of power electronic equipment, and includes: a functional area division module mainly used to divide the power supply into multiple functional areas according to the power supply topology structure; a real-time data acquisition module mainly used to obtain parameter data of each functional area during the power supply operation stage; a current model construction module mainly used to establish a current time series prediction model; a trend prediction module mainly used to generate future current change trends; an abnormality judgment module mainly used to identify abnormal situations by performing similarity matching with real-time data; a fault location module mainly used to match abnormal current waveforms with fault modes, and construct a directed graph to locate the fault position; the advantages of the present invention are: through multimodal data fusion and deep learning models, early warning and precise positioning of power supply faults are achieved, and the operation and maintenance efficiency and reliability of power electronic equipment are significantly improved.

This application relates to a method, apparatus, and storage medium for dynamic operation of a power grid based on seasonal scenarios. The dynamic operation includes: constructing a power topology model of a flexible microgrid; obtaining operational data for each grid partition on a target day of the current seasonal scenario; determining update indicators for each grid partition based on the operational data; adjusting the operational status of energy storage devices and the on / off states of remote control switches according to the comparison results of each update indicator with a preset update threshold; determining the grid partition structure; using the grid partition structure and the operational status of the energy storage devices as the initial partition configuration for the next seasonal scenario; and again adjusting the energy storage devices and remote control switches according to the update indicators; and adjusting the operational status of the power topology model in each seasonal scenario, using the current seasonal scenario and the next seasonal scenario respectively. This application improves the power grid's ability to dynamically operate under different seasonal scenarios.

The present application belongs to the technical field of power supply for large-capacity high-frequency transformer testing, and particularly relates to an energy feedback large-capacity high-frequency transformer testing power supply topology and control method, aiming to solve the problem of resource waste caused by large power consumption in the temperature field testing of large-capacity high-frequency transformers in the prior art. The present application comprises: a three-phase multi-winding transformer, including one primary three-phase winding and M+N secondary three-phase windings; M+N AC / AC modules corresponding to the secondary three-phase windings; a first resonant capacitor connected in series between the first AC / AC module and the high-voltage side of the large-capacity high-frequency transformer to be tested; a second resonant capacitor connected in series between the M+1th AC / AC module and the low-voltage side of the large-capacity high-frequency transformer to be tested; and a first bypass switch and a second bypass switch connected in parallel with the first resonant capacitor and the second resonant capacitor, respectively. The present application saves electric energy and occupies less resources, has a modular structure, a wide test voltage range, a low rated voltage of a single module, and strong reconfigurability.

The invention discloses a main network SVG graph model updating method, system and device based on an overlapping sliding window mechanism and dynamic correction and a medium, and the method comprises the steps: carrying out the sliding cutting of a main network SVG graph according to a preset window size and an overlapping rate, and generating a sub-graph set containing an overlapping region; electrical equipment primitives and coordinates in a sub-graph set are recognized through a target detection model, connecting lines between the equipment are extracted based on a straight line detection algorithm, the equipment primitives and the coordinates of the connecting lines are accurately positioned, and the problems of missed recognition and false recognition in a high-density graph are solved; the connection relation is dynamically corrected according to a power topology rule, reconnection operation is executed on the connecting lines crossing the bus to correct abnormal topology, a final to-be-rendered area is generated, and it is ensured that a graph model is consistent with an actual power grid structure; calculating a comprehensive complexity index of the region according to the attribute value of the to-be-rendered region; and on the basis of the comprehensive complexity index and the set threshold value judgment result, the judgment area is subjected to CIM model data rendering or updating operation, interface jamming is avoided, and the main network SVG graph model updating requirements of different complexities based on an overlapped sliding window mechanism and dynamic correction can be met.

The invention discloses a high-frequency isolation high-reliability UPS device power topological structure which is characterized in that commercial power is connected to a UPS distribution box, is output to a UPS host through a change-over switch, and is output after being connected to a power correction unit; the battery is connected to the UPS host, outputs DC360V through the boost unit, outputs the DC360V after being connected to the power factor correction unit through the voltage stabilization unit, and is connected to the high-frequency transformer isolation unit in parallel with the output of the commercial power. After rectification and voltage stabilization, the power is connected to the inversion unit and is output to the UPS distribution box, and multiple paths of load power are output through the circuit breaker. Commercial power input adopts a normal power supply mode and a maintenance power supply mode, and when the UPS host is abnormal, the UPS distribution box is switched to the maintenance mode to directly supply power to the load. The problem that a traditional inverter is not isolated when commercial power is input is solved, input and output full isolation of the UPS device is achieved, the anti-interference capability of the device is improved, and the reliability of the device is enhanced.