71 results about "Differential correction" patented technology

The invention relates to the technical field of image data processing, in particular to an automatic driving model based on a mixed low-rank expert and a multi-domain adaptive fine tuning method, and the model comprises a general model base in a frozen state and a multi-domain adaptive component in a trainable state; the universal model base comprises an image encoder, a measurement information encoder and a trajectory planner; the multi-domain adaptation component comprises a mixed low-rank expert group and a domain sensing router; the mixed low-rank expert group is used for calculating a difference correction signal according to the input characteristics; and the domain sensing router is used for receiving the shared image features extracted by the image encoder, outputting probability vectors and dynamically activating corresponding experts in the mixed low-rank expert group according to feature distribution of input data. According to the method, plug-and-play type efficient adaptation of a single universal model to multiple downstream domains is achieved with extremely low calculation and storage cost, and the pain point of a traditional multi-domain adaptation method is perfectly solved.

The invention relates to an intelligent ship water diversion method and system based on RTK and inertial navigation fusion, and the method comprises the steps: firstly, deploying and calibrating an RTK base station at a wharf commanding height, generating a differential correction in real time, and broadcasting the differential correction to a shipborne end; utilizing RTK surveying and mapping equipment to collect shoreline feature points, adopting a cubic spline interpolation algorithm to construct a continuous and smooth three-dimensional shoreline curve model, and fusing wharf geographic information to generate a shoreline visual map with geographic coordinates; establishing a ship coordinate system, converting inertial navigation attitude data into the coordinate system through a rotation matrix and a translation vector, fusing the inertial navigation attitude data with RTK position data, and outputting a high-precision ship pose; fusion data are transmitted through a 5G network, a shore end converts the fusion data into a local plane coordinate system and generates a real-time dynamic visual interface, the real-time dynamic visual interface is pushed to a tug terminal through an RTSP protocol, water diversion operation is assisted, high-precision perception, visual guidance and shore ship cooperation of the ship berthing process are achieved, and the water diversion safety and the intelligent level are improved.

The invention relates to a high and steep slope real-time monitoring method and system based on multi-source data fusion and cooperative enhancement. The method comprises the following steps: scanning a slope provided with a corner reflector array by using a synthetic aperture radar (SAR); performing real-time differential correction on data acquired by the radar by taking the corner reflector in the stable region as a reference so as to eliminate common noise; meanwhile, a three-dimensional laser scanner is used for obtaining point cloud data of the slope; coordinate registration and fusion of radar data and point cloud data are realized through the corner reflector array, so that a three-dimensional deformation field of the slope is generated. The method provided by the invention has the advantages of high monitoring precision, high signal-to-noise ratio and good data reliability.

The invention discloses a high-precision receiver analog signal comprehensive detection method based on a Beidou satellite, and particularly relates to the technical field of signal detection. The method comprises the following steps: acquiring satellite navigation signal data received by a receiver under a preset simulation motion track condition, respectively constructing a navigation calculation reference track evolution model and a high-dynamic track evolution model, analyzing an evolution change trend of a differential correction number, and extracting a dynamic response parameter of differential data; carrying out joint modeling on the dynamic response parameters and the motion state calculation model data, and outputting synchronous offset degree data; based on the synchronization offset degree data and the time sequence trajectory prediction data, analyzing a residual error between an actual resolving position and an expected trajectory of the receiver at each sampling time point; and whether a false detection risk caused by error response lag exists in the satellite navigation signal data or not is judged by combining a high-dynamic trajectory evolution model and residual error distribution, so that the test accuracy and evaluation reliability of the differential positioning performance of the receiver are effectively improved.

The invention provides a reference station network differential correction precision evaluation method, device and equipment and a medium, and the method comprises the steps: taking the coverage of each reference station in a reference station network as a purpose, constructing a virtual monitoring station for the reference station, and enabling every three virtual monitoring stations to form a triangular monitoring unit; performing exception elimination on the to-be-selected satellites which can be observed by the reference station differential correction based on the monitoring unit to obtain a target satellite after exception elimination; and based on the observation data of the target satellite and all the monitoring units, evaluating the positioning precision of the base station network. According to the embodiment of the invention, a mode of constructing the virtual monitoring station for each reference station is adopted, all-weather evaluation of the precision of the reference stations is realized, and the accuracy of the used data is ensured.

The invention relates to the technical field of electricity approaching early warning, in particular to a transformer substation operation vehicle electricity approaching early warning method and system based on space logic calculation. The method comprises the following steps: constructing a static safety electronic fence of the electrified equipment of the transformer substation, continuously generating differential correction data, simultaneously carrying out carrier phase differential calculation, and outputting real-time three-dimensional coordinate data of an operation vehicle and a moving part; the space coordinates of the static safety electronic fence and the real-time three-dimensional coordinate data of the operation vehicle and the moving part are input into a space logic calculation unit for real-time space distance comparison and logic operation, and the grade of an early warning area where the operation vehicle is located is judged; a forced stop signal is sent according to the judged early warning area grade; and after receiving the forced stop signal, the active stop device is driven by the controller to realize forced stop of the vehicle, and continuously feeds back a braking state to the spatial logic calculation unit to form a safe closed loop. According to the invention, the operation safety of the substation operation vehicle can be obviously improved.

The invention relates to the technical field of land surveying and mapping, in particular to a land surveying and mapping field acquisition system based on a mobile terminal, which comprises a hardware resource scheduling layer, a high-precision positioning processing engine, a field operation interaction engine, a real-time data quality verification module, an off-line and on-line asynchronous collaborative system and a multi-source heterogeneous data encapsulation center. According to the system, differential correction data and inertial navigation data are fused by using a positioning engine to realize centimeter-level positioning; executing geometric topology check and spatial precision evaluation through a verification module; and realizing incremental synchronization of end cloud data by adopting an asynchronous collaborative system. The invention aims to solve the problems that surveying and mapping depend on special hardware and the working continuity is poor in a weak network environment. By adopting the technical scheme, high-precision and high-reliability field acquisition is realized, the deployment cost is remarkably reduced, and the operation efficiency and the data quality are improved.

The invention provides a positioning time service base station, a positioning time service method, a storage medium and a program product, relates to the field of communication, and can solve the problems that an existing positioning time service scheme is high in construction, operation and maintenance cost and has a service blind area. The base station comprises a GNSS receiver chip, a crystal oscillator, a data acquisition and processing unit and a radio frequency unit, the GNSS receiver chip is used for receiving more than two frequency point signals at the same time; wherein the frequency point signal comes from a satellite navigation system; the crystal oscillator is used for providing time reference; the data acquisition and processing unit is used for receiving observation data of the satellite navigation system; performing differential calculation at least based on the observation data to obtain differential correction data; wherein the differential correction data is used for the terminal equipment to carry out positioning correction or time service correction; a radio frequency unit; and the radio frequency unit is used for sending the differential correction data to terminal equipment located in a service area of the positioning time service base station.

The invention discloses a feeding and discharging device of automatic connecting rod machining equipment. The feeding and discharging device comprises a rack, and a chain type conveying module and a grabbing and positioning module are arranged on the rack; the chain type conveying module comprises a conveying chain, a differential correction structure for adjusting the pitch of the conveying chain, guide blades arranged on chain links of the conveying chain, and a profiling guide rail which is arranged on the rack and is matched with the guide blades for angle adjustment; the grabbing positioning module adopts a vertical nesting layout, and comprises a parallel clamping mechanism, a composite buffer system and a reference positioning system which are arranged right above the output end of the conveying chain; the chain type conveying module outputs connecting rods with uniform postures, and the grabbing and positioning module completes grabbing, transferring and accurate positioning and placing. Through the synergistic effect of an angle self-adaptive guide system composed of a variable pitch mechanism, a differential correction structure, a guide blade and a profiling guide rail in the chain type conveying module, the new energy automobile engine / driving motor connecting rod can be accurately conveyed, and it is ensured that the connecting rods flowing into a grabbing station have highly-unified preset postures.

The present application relates to the field of aerospace technology, specifically to a kind of multi-pulse parallel construction method of gravitational wave detection formation considering timing optimization, each spacecraft of formation simultaneously departs from virtual center, each spacecraft is moved to the preset specified position by applying multiple pulses, so as to form a kind of spacecraft formation that can be used for space gravitational wave detection, by giving the steps of calculating the initial position and velocity of each spacecraft and applying initial control to reach the initial position and velocity, the state quantity of each spacecraft at each time in the formation construction process is calculated, so as to provide reference for the design of satellite formation task in actual engineering application, and the accuracy of the linearized relative motion dynamics model established by the multi-pulse parallel space gravitational wave detection formation is improved by differential correction, the error in the relative motion dynamics model is reduced, the initial state set in advance can be effectively reached, and the working efficiency of formation component is improved.

This application relates to the field of image data processing technology, and particularly to an autonomous driving model based on hybrid low-rank experts and a multi-domain adaptation fine-tuning method. The model includes: a general model base in a frozen state and a multi-domain adaptation component in a trainable state. The general model base includes: an image encoder, a measurement information encoder, and a trajectory planner. The multi-domain adaptation component includes: a hybrid low-rank expert group and a domain-aware router. The hybrid low-rank expert group is used to calculate differential correction signals based on input features. The domain-aware router is used to receive shared image features extracted by the image encoder, output probability vectors, and dynamically activate the corresponding experts in the hybrid low-rank expert group according to the feature distribution of the input data. This application achieves highly efficient "plug-and-play" adaptation of a single general model to multiple downstream domains with extremely low computational and storage costs, perfectly solving the pain points of traditional multi-domain adaptation methods.

This invention discloses a method and apparatus for estimating the height of monocular remote sensing images based on semantic distribution and region modulation, belonging to the field of computer vision and remote sensing image processing technology. The method first obtains a multi-scale feature map through bidirectional multi-scale feature extraction; then, it uses a height interval semantic modeling module to introduce global height semantic information to modulate the height interval representation, obtaining a preliminary height prediction; finally, it employs a region-aware pixel-level height modulation module to explicitly model the height region to which a pixel belongs and generate modulation parameters, performing differential correction on the preliminary prediction. This invention can significantly improve the accuracy, spatial consistency, and robustness of building height estimation, and is suitable for applications such as urban modeling and remote sensing mapping.

The invention discloses a differential broadcast telegraph text integration method for a Roland time service system. The method comprises the following steps: extracting differential correction parameters of differential stations in a broadcast station coverage area within a preset time period; obtaining a plurality of differential correction values of each differential station in a preset time period; grouping all the differential stations according to the differential correction deviation of the differential stations; determining the newest statistical average value of each differential station in an updating period of a differential correction parameter; for each group, determining all statistical average values of each group based on the latest difference correction difference value of each difference station; obtaining a differential correction parameter of each group according to all statistical average values of each group; and loading the differential correction parameter of each group, the latest statistical average value of each differential station in the group and the data age of the differential correction parameter of the group into a broadcast message for broadcast. According to the invention, the space occupied by the differential telegraph text is saved, the time consumed for broadcasting the differential telegraph text is reduced, and the actual service life of the differential correction parameter is prolonged.

The application discloses a response method and device for differential enhanced service request, equipment and storage medium, the method comprises the following steps: obtaining the differential enhanced service request sent by a service requester; determining the grid range type corresponding to the differential enhanced service request according to the grid coding in the differential enhanced service request; determining the differential correction data in the corresponding grid range and the broadcast frequency of the differential correction data according to the grid range type; sending the generated differential correction data to the service requester according to the determined broadcast frequency; by implementing the application, the differentiated needs of different service requesters for correction data can be met.

The application provides a real settlement monitoring method and monitoring equipment for earth and rockfill dam construction period, adopts a full-terrain walking robot to carry a high-precision GNSS receiver, realizes autonomous plane positioning and static measurement when walking on the earth and rockfill dam filling yard, carries out differential calculation with the differential correction data obtained by the reference station, obtains accurate three-dimensional positioning information and displacement change data, measures after each filling layer of the earth and rockfill dam filling yard is rolled and before the next filling layer is filled, respectively, and obtains the real settlement data of the earth and rockfill dam construction period, and feeds back the displacement change data and the real settlement data to the full-terrain walking robot through a server, which is used for adjusting the measurement plan of the full-terrain walking robot. Compared with the traditional manual measurement or fixed monitoring equipment, the application has autonomous movement and high-precision positioning capability.

The invention provides a near-infrared displacement monitoring method and system based on multi-reference point correction, and the method comprises the steps: obtaining a near-infrared target image sequence, and extracting a candidate reference point set and a measuring point set; performing multi-dimensional feature extraction on the two to obtain a feature vector set; inputting the feature vectors into a preset multi-dimensional similarity model for similarity evaluation, and generating a comprehensive similarity matrix; self-adaptive matching is carried out based on the matrix and a similarity threshold value, and a reference point-measuring point matching relation table is established; and finally, performing differential correction on the measuring point set according to the matching relation to obtain a measuring point displacement value set. According to the near-infrared displacement monitoring method and system based on multi-reference point correction provided by the invention, the reference stability is improved through multi-reference point screening and multi-dimensional feature extraction, and adaptive matching is performed through the preset multi-dimensional similarity model to ensure the matching accuracy, so that the displacement data accurately reflects the real deformation of the structure, and the accuracy of the real deformation of the structure is improved. And the accuracy and robustness of a monitoring result are improved.

The invention discloses a multi-mode fusion yak positioning and environment sensing system, and belongs to the technical field of livestock breeding information, the system comprises a collar type intelligent sensing terminal, a fixed environment sensing base station, an edge computing gateway and a cloud platform, the terminal collects GPS, inertia, vision and microenvironment data; the base station provides differential correction and environment data; the gateway executes differential positioning, dead reckoning and environment correlation analysis; and the cloud performs deep behavior recognition and abnormal situation reasoning through multi-modal fusion and a graph neural network model. The invention aims to solve the problem of information fragmentation caused by poor positioning robustness and multi-source data isolation in a complex terrain, high-precision all-weather positioning is realized, and early warning accuracy and decision support effectiveness are improved through comprehensive cognition of'individual-cattle-environment '.

This application provides a satellite terminal satellite acquisition method, electronic device, storage medium, and program product, relating to the field of satellite communication technology. This method, by introducing position confidence assessment, enables the satellite terminal to adaptively select the satellite acquisition path based on its own positioning quality. When positioning is reliable, it directly utilizes local ephemeris to quickly acquire low-Earth orbit (LEO) satellites. When positioning is unreliable or ephemeris is outdated, it switches to using high-Earth orbit (HEO) satellites to obtain differential correction and auxiliary position information. This effectively compensates for ephemeris and position errors without relying on real-time precise positioning, reduces the searching pointing deviation of LEO satellites, and narrows the search range. It avoids long waiting times and wasted power due to blind scanning, while improving the acquisition success rate. This allows the terminal to quickly join the network even under harsh network conditions, ensuring the timeliness of user service establishment and the terminal's survivability in extreme environments.

The application provides a network RTK differential correction number generation method based on Beidou and low-orbit constellation fusion, which comprises data acquisition, receiving pseudorange observation values and carrier phase observation values of Beidou satellite navigation system satellites and low-orbit constellation satellites through a ground reference station receiver simultaneously; constructing a joint observation model of Beidou satellite navigation system and low-orbit constellation, and using an extended Kalman filtering algorithm to estimate receiver clock difference and system bias between the two systems in real time; adaptive weighted differential correction number calculation, calculating pseudorange differential correction numbers of Beidou satellites and low-orbit satellites respectively based on the estimated system bias; correction number broadcasting, dynamically determining the effective duration of the differential correction number of the low-orbit satellite according to the orbit height of the low-orbit satellite, and encoding the correction numbers of the Beidou satellites and the low-orbit satellites into RTCM text format to broadcast to the user end; the user end receives the correction numbers, corrects the local Beidou satellite and low-orbit satellite observation values, performs joint RTK solving, and verifies and confirms ambiguity fixing.

The invention discloses an unmanned aerial vehicle intelligent distribution control method and system based on GNSS differential calculation, relates to the technical field of unmanned aerial vehicle intelligent distribution control, and aims to solve the problems that the unmanned aerial vehicle intelligent distribution control technology in the prior art is insufficient in positioning precision, lags in cold chain control response and excessively depends on a communication link. The system comprises a differential GNSS positioning module, a dynamic cold chain control module, an autonomous decision and emergency control module and a cooperative control module. The differential GNSS positioning module comprises a GNSS signal receiving unit arranged at an unmanned aerial vehicle end and a differential signal transmitting unit arranged at a ground reference station, the GNSS signal receiving unit is used for receiving satellite positioning signals, and the differential signal transmitting unit is used for transmitting differential correction data; and the differential GNSS positioning module carries out fusion processing on the satellite positioning signal and the differential correction data through a filtering algorithm. The method has the advantage that the precision, stability and autonomy in the intelligent distribution control process of the unmanned aerial vehicle are improved.

The application relates to an intelligent ship pilotage method and system based on RTK and inertial navigation fusion, which comprises the following steps: deploying and calibrating an RTK base station at a high point of a wharf, generating a differential correction quantity in real time and broadcasting the differential correction quantity to a shipborne terminal; collecting shore feature points by using an RTK surveying and mapping device, constructing a continuous and smooth three-dimensional shore curve model by using a cubic spline interpolation algorithm, and generating a shore visual map with geographic coordinates by fusing wharf geographic information; establishing a ship coordinate system, converting inertial navigation attitude data to the coordinate system by using a rotation matrix and a translation vector, fusing the RTK position data, and outputting high-precision ship position and posture; transmitting the fused data through a 5G network, converting the data to a local plane coordinate system at a shore terminal, generating a real-time dynamic visual interface, pushing the interface to a tugboat terminal through an RTSP protocol, and assisting in pilotage operation, so that high-precision perception, visual guidance and shore-ship cooperation in a ship berthing process are realized, and the safety and intelligent level of pilotage are improved.

The application relates to the field of high-precision positioning technology and discloses a troposphere delay product generation method combining GNSS and polar orbit remote sensing satellites, which comprises the following steps: calculating the difference between real-time GNSS true value points and a remote sensing satellite troposphere delay background field to generate a group of differential anchor points; after reliability arbitration is performed on the differential anchor points, a continuous troposphere differential correction field is generated through a terrain-constrained interpolation algorithm, the weight of the interpolation algorithm is based on the effective path distance considering the terrain obstacle penalty, and finally the correction field is superimposed on the background field; through generation and application of a dynamic correction field based on physical deviation, the method replaces the direct mathematical fitting of different source data in the prior art, the interpolation method of terrain constraint is used, space propagation of correction information can follow the restriction of real physical barriers, and unrealistic correction results in complex terrain areas are avoided.

This application proposes a simulation training method, system, electronic device, and storage medium, which includes: acquiring a preset geographic mapping model, fixed virtual geographic coordinates of a virtual reference station, and satellite ephemeris data; real-time acquisition of physical coordinate data fed back by the mobile platform; converting the physical coordinate data according to the geographic mapping model to obtain the virtual geographic coordinates of the mobile station; calculating the first satellite observation data using the fixed virtual geographic coordinates based on a unified time reference and precise satellite ephemeris data, and calculating the second satellite observation data using the virtual geographic coordinates of the mobile station; generating differential correction data based on the first satellite observation data; acquiring the positioning result output by the receiver of the tested mobile station; calculating the deviation index between the positioning result and the virtual geographic coordinates of the mobile station, and generating a training evaluation report.

The application relates to a robot differential correction positioning method and device. The robot differential correction positioning method comprises the following steps: calibrating a hand-eye camera and a fixed camera; photographing an object and a calibration board by the fixed camera to obtain fixed camera coordinates of the object and a reference station; photographing the calibration board by the hand-eye camera to obtain hand-eye camera coordinates of the reference station; converting the object coordinates to obtain hand-eye camera coordinates of the object; and calculating correction errors of a mobile station and the reference station by using a differential correction algorithm, and applying the correction errors to the converted hand-eye camera coordinates of the object. The robot differential correction positioning method effectively improves the accuracy of object coordinate detection, realizes sub-millimeter positioning, and meets the demand for high-precision operation by setting the object to be positioned as the mobile station, setting the calibration point as the reference station, simultaneously detecting the reference station coordinates when detecting the object coordinates, and differentially correcting the object coordinates according to the reference station coordinates.

The application provides a satellite terminal satellite searching method, an electronic device, a storage medium and a program product, and relates to the technical field of satellite communication. The method introduces a position confidence evaluation, and a satellite terminal can adaptively select a satellite searching path according to the positioning quality of the satellite terminal. When the positioning is reliable, the local ephemeris is directly used to quickly capture a low-orbit satellite, and when the positioning is unreliable or the ephemeris is aging, the high-orbit satellite is used to obtain a differential correction quantity and auxiliary position information, so that the ephemeris error and the position error are effectively compensated without relying on real-time accurate positioning, the search pointing deviation of the low-orbit satellite is reduced, the search range is reduced, the long waiting time and power waste caused by blind scanning are avoided, the capture success rate is improved, the terminal can still realize fast network access under poor network conditions, and the timeliness of user service establishment and the survival ability of the terminal under extreme environment are ensured.

The invention discloses an operation vehicle factory safety management and control system and method based on Beidou differential positioning, and the system comprises a positioning enhancement module which carries out the tight coupling fusion processing of satellite observation data, differential correction data and inertial data, and obtains the precise pose of a vehicle; the digital twin module is used for carrying out registration and three-dimensional reconstruction on the point cloud data of the working vehicle factory, constructing a three-dimensional factory environment model and carrying out dynamic updating by adopting an incremental updating algorithm; the safety management and control module is used for fusing the accurate pose of the vehicle and the three-dimensional factory environment model to obtain a safety situation map; and performing trajectory prediction on the security situation map based on a kinematic model, performing space-time intersection analysis on the predicted trajectory and the dynamic security region to generate a risk assessment result, and generating an early warning signal according to the risk assessment result. The method can solve the problems that Beidou signals are affected by the multipath effect, model updating is difficult, and real-time safety management and control challenges exist in the complex factory environment, and the safety management and control level of operation vehicles is improved.

This invention relates to an intelligent monitoring and management system for anti-slip track shoes, belonging to the field of anti-slip track shoe technology. The system includes: a data acquisition module: the anti-slip track shoe intelligent agent receives satellite signals through a BeiDou high-precision positioning submodule and generates centimeter-level positioning coordinates by combining differential correction data output from a BeiDou differential base station; a data transmission module: the collected positioning data, status data, and operation records are securely transmitted to the server in the interaction layer via an industrial-grade router, switch, and security gateway. This invention, through a design combining "track shoe intelligent agent sensors + low-power wireless communication," collects real-time data on track shoe contact status and displacement, reducing the impact of data reporting delays on operations; and, combined with a visual terminal, dynamically displays the status of all track shoes at the station, allowing for synchronous updates of status information without manual intervention.

The invention provides a Beidou time service signal analog output method and system based on security isolation. The method comprises the following steps: firstly, receiving a differential correction data stream and a Beidou satellite original time service signal, and transmitting the data stream and the Beidou satellite original time service signal to a security isolation environment for fusion to generate a high-precision space-time reference signal; secondly, in a security isolation environment, according to a preset formation and a cooperative track, resolving the high-precision space-time reference signal into a multi-path simulation time service signal for each unmanned aerial vehicle node; respectively embedding identity identification codes of the corresponding unmanned aerial vehicle into the multiple paths of analog time service signals, and carrying out anti-deception coding; then, generating a safety time service signal sequence and directionally injecting the safety time service signal sequence into an unmanned aerial vehicle flight control network; and finally, the unmanned aerial vehicle nodes are synchronized according to the corresponding safety time service signals. According to the technical scheme provided by the invention, the cooperative positioning and synchronous motion precision of the whole unmanned aerial vehicle cluster is improved, and high-precision safe synchronization of the anti-interference unmanned aerial vehicle cluster is also realized.

The invention relates to the technical field of power pole micro-deformation monitoring, in particular to a multi-source data power pole micro-deformation monitoring system and method based on artificial intelligence, and the system comprises a multi-source data collection module, an intelligent analysis module, a risk analysis module and a decision output module. Millimeter-level positioning is realized by adopting a carrier phase difference (RTK) technology, and a reference station RTCM3.2 format difference correction number is acquired in real time through an NTRIP protocol; a three-dimensional monitoring framework is constructed through a multi-source data acquisition module integrating Beidou / GPS dual-frequency RTK positioning, JSINSAR interferometry and foundation micro-strain sensing, the problem that a traditional single sensor is insufficient in monitoring dimension is solved, collaborative analysis of deformation, environmental stress and visual characteristics is achieved in combination with an environmental sensor and visual data, and the system has the advantages of being simple in structure, convenient to operate and high in practicability. And the monitoring comprehensiveness in a complex scene is obviously improved.

The invention, which relates to the technical field of power operation monitoring, discloses a 5G communication base station power operation monitoring system comprising a connection construction module, a difference acquisition module and an anomaly propagation module. The connection construction module is used for generating a two-dimensional correlation intensity matrix by fusing the power line topological graph and the base station geographic coordinates after differential correction; the difference acquisition module divides a risk level for each cluster through a clustering line, and configures a differentiated data acquisition frequency according to the risk level; and the anomaly propagation module calculates a line propagation coefficient and a geographical influence coefficient so as to judge an anomaly type. The operation stability of the 5G base station power is guaranteed, and the operation and maintenance efficiency is improved.