239 results about "Wiring diagram" patented technology

The invention provides a power grid wiring diagram automatic analysis system and method based on deep learning. The method comprises the steps of S1, data preprocessing and model training; s2, carrying out equipment identification and association; step S3, topology reasoning and verification are carried out; s4, standardized output: converting the verified power grid network model into a regulation and control cloud platform standard graphic file, and optimizing graphic arrangement through an adaptive layout algorithm; step S5, performing closed-loop optimization; and S6, performing manual intervention and iteration. According to the method, the wiring diagram analysis efficiency and accuracy can be remarkably improved, manual intervention is reduced, standardized graph generation of a power grid regulation and control cloud platform is supported, and a reliable data basis is provided for power system analysis and decision making.

The invention provides a main wiring automatic mapping method and system for SCD file topological connection, and the method comprises the steps: forming a topological connection code of equipment in a gap by taking the gap as an object through automatic analysis and recognition of primary equipment described by transformer substation configuration in a transformer substation SCD file and a topological connection relation of the primary equipment; building and layout of device primitives in the interval and generation of connecting lines are completed based on topology connection coding, and therefore automatic generation of interval primitives is completed; the layout mode of the main wiring diagram is automatically deduced by analyzing various characteristics such as the voltage class, the bus type and the wiring mode of the transformer substation, and automatic matching of the layout of the main wiring diagram is achieved; by analyzing the position relation between the main transformer and the voltage grade, analyzing the direction of the connecting line and setting a proper connecting line inflection point, the intersection and overlapping of the connecting line are avoided, and the automatic generation of the main transformer and the main transformer interval connecting line is completed.

The invention discloses an automatic intelligent point alignment method for identifying a G file of a master station based on AI, and the method comprises the steps: analyzing a G file of a primary wiring diagram of a transformer substation through an AI server, and building a mapping relation between equipment topology and screen coordinates; host images are monitored in real time through computer vision, and automatic navigation of target intervals is achieved through an intelligent mouse module; remote signaling and telemetering signals are identified through an AI analysis system; and after the test parameters are configured by the management and control platform, an instruction is issued through the simulation terminal, and a monitoring background return value and an AI identification result are automatically compared. According to the method, the problems of low efficiency and high error rate of traditional manual debugging are solved, and full-process coverage, full-loop verification, full-channel comparison and full-information verification can be carried out on three-remote information of the transformer substation.

The invention provides a method for tracing a multi-power-source power supply path in a power system, and the method comprises the steps: obtaining topological data of a power transmission line network diagram, a transformer substation primary wiring diagram, a distribution network looped network diagram and a user low-voltage side wiring diagram in a specified power network, and carrying out the multi-source heterogeneous fusion, so as to construct an integrated transmission, transformation and distribution network topology, corresponding switch state data and mobile power supply equipment operation data are configured in the transmission, transformation and distribution integrated network topology so as to obtain a digital power grid operation model; constructing a node admittance matrix by taking each piece of equipment as a node based on a topological relation between lines in the digital power grid operation model so as to determine lines belonging to a loop, a branch and a cut set in the digital power grid operation model; and determining a multi-power-supply power supply path according to the circuits of the loop, the branches and the cut set in combination with the switch state information and the mobile power supply equipment operation information. According to the invention, the transmission, transformation and distribution integrated network can be comprehensively covered, and the problems of low efficiency, high use cost and the like can be solved.

The invention discloses a multi-modal depth perception and adaptive segmentation full-spectrum primitive intelligent identification method, system and device for a primary main wiring diagram of a plant station, and a medium, and belongs to the technical field of intelligent identification of engineering drawings of power system plant stations, and the method comprises the steps: fusing the dual-source data of a vector diagram and a grating diagram, and constructing a standard primitive library; secondly, primitive detection is realized by adopting a recognition and segmentation collaborative multi-stage network, and the contour precision is improved through model distillation; secondly, iteratively optimizing the recognition capability of the model on the difficult sample by using an active learning mechanism; and finally, finishing primitive classification and result optimization in combination with deep metric learning and electrical topology constraints. According to the invention, through multi-modal data fusion and multi-scale feature adaptive fusion, the recognition capability of a small target and a fine primitive is improved; by means of a double-branch heterogeneous architecture and a structural distillation mechanism, pixel-level high-precision edge segmentation is realized, and the identification problem of dense adhesion primitives is effectively solved.

The invention discloses a primary main wiring diagram on-site configuration method based on microcomputer protection, and relates to the technical field of power system microcomputer protection. Two-stage mapping from a grid position to a primitive identifier and two-stage mapping from the primitive identifier to a rendering handle are maintained in parallel in a nonvolatile memory through a double-page mirroring transaction; reading screen resolution, adaptively calculating blank space and grid scaling according to a residual pixel threshold value, and outputting global coordinates; monitoring virtual switching value change in an operation period, dynamically selecting local or whole-screen refreshing according to a dirty rectangle merging algorithm, and triggering a degradation strategy in combination with a comprehensive load score; the protection device can be safely and rapidly rearranged in a low-resource mode on site, a main wiring diagram is refreshed in real time, and the problems of factory returning upgrading, resolution ratio adaptation and refreshing lagging are solved.

A converter station AC filter group circuit breaker abnormal arcing modeling analysis method comprises the following steps: firstly, carrying out time mark alignment, amplitude calibration and filtering de-noising on wave recording signals such as AC bus voltage and filter branch current to obtain high-quality input data; then, combining a primary wiring diagram, enabling an alternating current system to be equivalent to an ideal power supply and series impedance, enabling circuit breaker arcing to be equivalent to arc resistance changing along with time, and constructing an equivalent circuit model comprising the power supply, the impedance, a capacitor, an electric reactor and dynamic arc resistance; establishing a state space equation according to a voltage and current constraint relation, and performing time domain solution by adopting a two-step two-order or three-order diagonal implicit Runge-Kutta method to obtain voltage and current responses of the whole process of the fault; on this basis, the arc resistance change is inversed according to the ratio of arc voltage to current, the current zero passage is identified through the resistance symbol change at adjacent moments, and the zero passage section is smoothed; and finally, the inversion result is verified again by means of the simplified equivalent model.

The invention provides an electrical drawing automatic identification method and system, and relates to the field of electrical drawing identification, and the electrical drawing automatic identification method comprises the following steps: 1, image preprocessing: collecting a multi-source original new energy electrical drawing, converting the multi-source original new energy electrical drawing into a uniform format, eliminating noise, and enhancing features; step 2, identifying and marking symbols; step 3, topological relation modeling; step 4, parameter verification; according to the method, the topological structure diagram of the electrical system is automatically generated by analyzing the connection relation between the elements, manual intervention is not needed, the recognition result and the BOM database of the new energy vehicle are integrated, automatic input and management of element information are achieved, production efficiency can be improved, human errors are reduced, and the method is suitable for popularization and application. The method can be expanded to automatic identification of transformer substation wiring diagrams and industrial robot electrical drawings, and promotes digital transformation of the manufacturing industry.

The invention relates to the technical field of smart home, and discloses a smart device auxiliary installation method and auxiliary installation system based on the AR technology, and the method comprises the steps that a terminal module obtains a device list and an installation task of a to-be-installed house from a cloud module; the terminal module constructs a 3D space model of a to-be-installed house through an SLAM algorithm, marks recommended installation positions of a plurality of devices in a device list, and superposes the recommended installation positions to a real-time image picture of the terminal module through an AR technology; and the terminal module generates a virtual wiring diagram of target equipment for the target equipment needing wiring in the multiple pieces of equipment, and the virtual wiring diagram is superposed to a real-time image picture of the terminal module through an AR (Augmented Reality) technology. According to the method provided by the embodiment of the invention, the 3D model of the to-be-installed house is automatically constructed through the SLAM algorithm, and the AR technology is adopted to provide guidance of the recommended installation position and the virtual wiring diagram, so that installation personnel can carry out visual reference, repeated measurement and trial and error are avoided, and the installation efficiency of the intelligent equipment is improved.

The invention relates to the technical field of image recognition, and discloses an automatic checking method, device and equipment for a power grid plant station wiring diagram and a storage medium, and the method comprises the steps: carrying out the image segmentation of a pre-obtained power grid plant station wiring diagram based on an edge detection algorithm, and obtaining a plurality of segmentation parts corresponding to the power grid plant station wiring diagram; a character recognition checking technology, a primitive feature matching checking technology and a topological connection relation checking technology are adopted to carry out recognition result checking on the multiple segmented parts, and abnormal recognition parts are obtained; and correcting the abnormal identification part to obtain an automatic identification result corresponding to the power grid plant station wiring diagram. Compared with a single checking technology, the overall recognition accuracy of the power grid plant station wiring diagram can be remarkably improved, recognition errors caused by manual or simple technologies are reduced, and full-process automation from picture segmentation, recognition result checking to abnormal part correction is achieved.

The invention relates to the technical field of graph generation, in particular to a self-adaptive topological plant station wiring diagram generation method, which comprises the following steps of: acquiring a wiring diagram image, extracting edge features, identifying an image type, positioning a line structure boundary, identifying text elements, classifying corresponding graph areas, and constructing a connection path to generate a topological wiring diagram. According to the method, clear recognition of the structure contour is realized through extraction of the edge features of the line structure of the image, the spatial position information of the graphic area is acquired in combination with a multi-scale primitive positioning technology, and then identity extraction and classification judgment are performed on the text elements contained in the image in a character recognition mode. And establishing a corresponding relationship between the equipment identifier and the graph region, constructing a complete path by combining the graph arrangement direction and the starting and ending coordinates of the connecting line, and determining a connection sequence and a relative position relationship between the equipment. Coordination and unification of spatial arrangement are realized through application of an attraction and repulsive force model between nodes, and clearness and coherence of a wiring diagram in the aspect of topological structure expression are improved.

The invention discloses a wiring diagram recognition method and device based on deep learning and electronic equipment, and the method comprises the steps: carrying out the correction and mask processing of an initial wiring diagram, obtaining a first wiring diagram, and fusing mask parameters and mask skeleton parameters; extracting circuit elements in the first wiring diagram based on fusion mask parameters and mask skeleton parameters, and classifying the circuit elements to obtain a first element set of the first wiring diagram; performing port connection line pairing processing and topological optimization processing on the first element set, and determining a second element set and a corresponding second wiring diagram; according to the second element set, performing multi-dimensional verification and global topological structuring processing on the second wiring diagram to obtain a topological graph data model; and determining a JSON file, a visual topological graph, a VF2 check report and a Neo4j graph database of the wiring diagram according to the structured data in the topological graph data model. According to the invention, a more accurate wiring diagram can be identified.

The invention provides a wiring diagram identification method based on a neural network and computer vision processing, and relates to the technical field of graph model identification, and the method comprises the following steps: inputting a power grid plant station wiring diagram, the wiring diagram format specifically comprising BMP, JPG and PNG; uniformly scaling the wiring diagram to a uniform pixel resolution, processing the power grid plant station wiring diagram, scaling, denoising, detecting circuit breakers and other primitives, correcting the direction, identifying characters, complementing to generate a structured record, reconstructing a wiring topology, associating the primitives, the characters and the topology through multi-modal matching, and finally obtaining the power grid plant station wiring diagram. And finally, a CIM file conforming to IEC61970 is mapped, automatic extraction and standardized conversion of wiring diagram information are achieved, the rotating frame detection and direction restoration technology is adopted, primitives rotating at any angle are accurately recognized, the direction of the primitives is unified, the problem of angle loss of traditional horizontal frame detection is effectively solved, and the detection efficiency is improved. The omission ratio and the repetition rate of dense or multi-scale primitives are reduced, and the robustness of primitive detection is improved.

The invention discloses a non-intrusive power grid wiring diagram switch state identification method and system based on deep learning, a processor and a medium. According to the method, electrical data and switching state data of a power grid are collected and marked, a '1D-CNN + BiLSTM + Transform 'hybrid network model is constructed after preprocessing and feature extraction, and the switching state of the whole network is recognized after training. Power grid equipment does not need to be transformed, an existing mutual inductor is used for collecting data, local mutation, time sequence dependence and global association of electrical data are effectively processed through multi-dimensional feature extraction and a hybrid network model, the accuracy and robustness of on-off state recognition are improved, and the method can be applied to power grid on-off state monitoring in real time. And reliable support is provided for intelligent power grid operation.

The invention relates to an automatic mapping method and system of a standard graph applied to power grid equipment homologous maintenance, and relates to the technical field of power grid mapping, and the method comprises the steps: extracting equipment ledger data, geographic coordinate data and topological connection relation data based on a preset power grid management system, and constructing a power grid equipment knowledge graph; layering the power grid equipment based on the voltage levels in the knowledge graph, and performing initial layout optimization on each layer of equipment by adopting a reinforcement learning model to generate a layered layout result; dynamically calculating the contact positions of the T-shaped wiring and the n-shaped wiring to generate contact coordinates; and generating a power grid logic view and a geographical wiring diagram based on superposition of the contact coordinates and the hierarchical layout result. By automatically generating the power grid logic view and the geographical wiring diagram, the dependency on manpower is greatly reduced, errors caused by personnel in drawing are reduced, and the drawing efficiency is remarkably improved.

The invention discloses an electric energy meter assembling standard intelligent identification system, which comprises an image acquisition module, an image preprocessing module, a wiring identification module and a standard comparison module, and is characterized in that the wiring identification module carries out NLM denoising and Gamma correction enhancement processing on a to-be-identified image, dynamically calculates a threshold value by using an Otsus method, accurately detects the position of a sleeve in an electric meter image, and carries out standard comparison on the to-be-identified image; and acquiring the coordinate of each casing, and identifying the casing fields by using PaddleOCR (Passive Optical Character Recognition). According to the method, through specific tests, wiring diagram shooting, threshold value dynamic calculation through an Otsus method, contrast enhancement through Gamma correction, NLM denoising, text detection, character recognition, geometric correction and other modes, OCR recognition is carried out on the image, algorithm and formula optimization of different degrees is carried out in the recognition process, the detection precision and efficiency are remarkably improved, and the method is suitable for popularization and application. And an effective solution is provided for automatic detection of the electric meter.

The invention provides an electric energy metering equipment wiring image-text recognition method and a related device, and the method comprises the steps: determining first image feature information based on image information and feature fusion operation, and determining first text feature information based on text information and depth context coding operation; determining a plurality of loss values under multiple levels based on the first image feature information and the first text feature information, and fusing features; determining a plurality of predicted loss values based on the fused features; and determining an optimized electric energy metering equipment wiring image-text recognition model based on the plurality of predicted loss values and the plurality of loss values under multiple levels, and determining an image-text recognition result based on the optimized electric energy metering equipment wiring image-text recognition model. Therefore, the semantic alignment precision of the image-text features is improved, the adaptability of the model to complex scenes is enhanced, the detection efficiency and accuracy are improved, the automation level of wiring specification detection of the electric energy metering equipment is further improved, and the operation safety and stability of an electric power system are improved.

The invention discloses an intelligent substation SCD automatic generation method, system and device and a storage medium. The method comprises the following steps: constructing a primary main wiring diagram, wherein the primary main wiring diagram comprises a topological relation in the intelligent substation; associating intelligent electronic equipment IED in a substation configuration description SCD file to a corresponding interval in the main wiring diagram; generating an IED configuration parameter based on the topological relation and the IED; and associating the ICD file of the IED to the SCD file, importing a loop configuration subscription relation table and a total station primary equipment interval classification, number and initial address table, and automatically generating a subnet, communication parameter configuration, port configuration and virtual loop configuration in combination with the IED configuration parameters so as to complete the configuration of the SCD file. According to the method, the SCD file meeting the specification can be automatically configured.

The invention relates to the technical field of computers, and discloses a primitive recognition method, device and equipment for a power grid plant station wiring diagram and a medium. The method comprises the following steps: acquiring a power grid station wiring diagram, and preprocessing the power grid station wiring diagram to obtain an original input image; the original input image is input into an improved YOLOV5 target detection network, and the improved YOLOV5 target detection network comprises an improved backbone network, an improved feature fusion network and an improved decoupling detection head; extracting feature information of an original input image through the improved backbone network to obtain a plurality of feature maps of feature layers with different scales; fusing the feature maps of the feature layers with different scales through an improved feature fusion network to obtain a plurality of fused feature maps; and identifying the fused feature maps through an improved decoupling detection head to obtain a primitive identification result. The invention provides an improved YOLOV5 target detection network, and the detection accuracy of the primitives is improved.

The invention relates to the technical field of disconnecting link maintenance and transformation, in particular to an isolation disconnecting link updating and transformation method. The method comprises the following steps: S1, performing inspection and quantitative analysis, performing equipment evaluation and problem diagnosis, and identifying operation problems and function deficiency; s2, performing model selection and material optimization, and setting a standard isolation switch and an electric operation assembly; s3, defining a transformation range, dividing transformation units based on the main wiring diagram, and covering the isolation disconnecting link and the grounding disconnecting link; s4, implementing a safety locking structure, adding an electrified locking and five-prevention system, and accessing whole plant management to prevent misoperation; and S5, accessing a monitoring system, and connecting an LCU screen and a computer monitoring system to realize remote centralized control. The contact resistance and temperature rise of the contact reach the standard, ablation is completely eradicated, the operating force and time are greatly reduced, jamming is returned to zero, the five-prevention locking success rate is extremely high, the misoperation drill can reach the zero penetration level, and the on-site actual effects of mechanical zero jamming, five-prevention zero penetration, remote control zero delay and old and waste utilization are achieved.

The invention relates to the technical field of intelligent power grid fault analysis, in particular to an intelligent power grid fault analysis method and system based on big data. Comprising the following steps: analyzing a power grid physical wiring diagram to construct a node admittance matrix, and extracting modal feature vectors through feature decomposition to generate a topological fingerprint database; synchronously collecting wide-area phasor data streams, dynamically intercepting a data time window, extracting resonance frequency band signals, and generating a fault feature data set through feature enhancement; a modal confidence factor is dynamically evaluated, when the dominant modal confidence coefficient is lower than a threshold value, redundant modals in the node-modal association map are activated, and fault feature vectors are fused and reconstructed; and matching the reconstruction vector with a fingerprint database, performing triple verification, and outputting a fault positioning coordinate or starting a safety protection instruction in combination with a confidence scoring decision. By introducing a dynamic confidence mechanism and a redundancy modal compensation strategy, the accuracy and fault-tolerant capability of fault identification in a new energy disturbance environment are improved.

The invention relates to a method and a system for generating a nuclear island relay protection system graphic structure. The method comprises the following steps: acquiring basic graphic layer data of an initial secondary wiring diagram, and establishing a unique identifier for each graphic element in the basic graphic layer data; and receiving a manual modification operation on the basic graphic layer data and recording the manual modification operation as a manual modification operation record. And when it is detected that the design input data is changed, new basic graph layer data is regenerated according to the changed design input data. Manual modification operation records are read, and for each manual modification operation record, the corresponding graphic element is searched in the new basic graphic layer data according to the unique identifier associated with the record. If the corresponding graphic element is found, the manual modification operation in the old basic graphic layer data is applied to the corresponding graphic element in the new basic graphic layer data according to the type and the parameter stored in the manual modification operation record, and a new-version secondary wiring diagram containing manual modification content is generated.

The application discloses an interactive adaptive generation method and device for a substation main wiring diagram, and belongs to the technical field of intelligent substations. The method comprises the following steps: loading a pre-constructed substation typical design framework template and a bus branch interval template; instantiating the substation typical design framework template and the bus branch interval template to generate a typical design template instance and a branch interval template instance; respectively performing graph module association on the typical design template instance and the branch interval template instance to generate the graph module associated typical design template instance and the graph module associated branch interval template instance; expanding a arrangeable interval according to the graph module associated typical design template instance and the number and form of the branch interval; and based on an interactive diagramming configuration mechanism, arranging the graph module associated branch interval template instance to the arrangeable interval to generate a substation main wiring diagram. The application can solve the problems of non-standard, low efficiency, easy-to-make errors in graph module association and difficult-to-control diagramming effect in automatic diagramming.

The application discloses a kind of test acceptance wiring scheme generation method, device, electronic equipment and storage medium.The method includes obtaining the description file of the secondary system to be accepted, including primary wiring diagram and secondary equipment description diagram;First type representation vector is obtained by graph information analysis to primary wiring diagram, and second type representation vector is obtained by graph information analysis to secondary equipment description diagram;With first type representation vector as the feature vector of simulation query, and with second type representation vector as the feature vector of simulation key, the query analysis based on local neighborhood optimization is carried out to simulation query feature vector and simulation key feature vector to capture the deployment relationship of primary equipment and secondary equipment and obtain wiring scheme semantic representation;Test acceptance wiring scheme diagram of the secondary system to be accepted is generated based on wiring scheme semantic representation.The technical scheme of the application ensures that the generated wiring scheme has comprehensiveness and reliability.

The application provides a power equipment wiring diagram data display method, device and storage medium. The method comprises the following steps: selecting a current data updating mode from a plurality of preset data updating modes; when a trigger condition required by the current data updating mode is met, updating an initialization xml file according to measurement point data, so as to display a graph element corresponding to the measurement point data in a browser. According to the embodiment of the application, when the trigger condition required by the current data updating mode is met, the initialization xml file can be updated according to the measurement point data, so as to display the graph element of different shapes and colors corresponding to the measurement point data in the browser. In addition, the graph element type can be customized according to requirements, and the display mode is flexible.

The invention provides a transformer substation wiring diagram automatic identification method and system, and the method comprises the steps: obtaining the canvas size and primitive coordinate information of a transformer substation wiring diagram in a transformer substation wiring diagram G file, obtaining the canvas size of a transformer substation wiring diagram screenshot, and calculating the size conversion coefficient of the transformer substation wiring diagram and the screenshot based on the canvas size of the transformer substation wiring diagram and the screenshot in the G file; calculating the coordinate information of each primitive in the screenshot based on the conversion coefficient and the primitive coordinate information of the G file transformer substation wiring diagram, marking each primitive in the screenshot, and identifying the information content of each primitive of the transformer substation wiring diagram in the screenshot; and carrying out variable quantity calculation on the current marked screenshot and the marked screenshot at the previous moment to obtain a change value of the primitive in the operation process of the transformer substation. The method supports display of images with various resolutions and various window images, meets the requirements of a mainstream dispatching automation master station system, and is wide in adaptability. In the whole recognition process, only the conversion rate is calculated for the first time, massive calculation is not needed, and the real-time monitoring requirement is met.

The invention discloses an automatic main wiring diagram forming method and system based on dynamic layout and template fusion. Carrying out regional division according to the voltage class of the transformer substation, and independently setting a main transformer region; optimizing region merging and splitting by adopting a genetic algorithm, setting a merging and splitting threshold value by calculating a region occupancy rate, and dynamically adjusting a region boundary to adapt to scenes such as transformer substation reconstruction and extension, interval increase or deletion and the like; and further aiming at the problems of large calculated amount of equipment layout in the interval, different voltage intervals of the same station and non-uniform styles of direct intervals of different stations, a standardized interval template matching method is adopted, the equipment primitive size, interval and alignment mode are set, the primitive position is automatically adjusted, the connecting line crossing condition is optimized, and finally the transformer substation main wiring diagram conforming to the specification is generated. The layout is dynamically adjusted based on the set threshold value, the overall space compactness is improved, the invalid blank is reduced, and the equipment arrangement is optimized in combination with template matching, so that the formed graph is more regular and conforms to the industrial standard.

The invention discloses a method and a system for identifying a connecting line relationship between primitives based on image identification, and relates to the technical field of image processing, the method comprises the following steps: constructing a primitive extraction model, and performing model pre-training in combination with a pre-constructed training data set, the primitive extraction model comprising a primitive identification sub-module and a neural coding sub-module; the wiring diagram to be recognized is input into the primitive extraction model to be analyzed and recognized, a primitive neural coding set is output, and the primitive neural coding set comprises a primitive entity set and an entity coding set; primitive connection relation reasoning is carried out on the primitive neural coding set based on the generative flow network, and a corresponding initial primitive connection network is generated; performing circuit semantic simulation verification on the initial primitive connection network by adopting a pre-constructed differentiable physical simulator to generate a simulation verification result; and iteratively executing the steps S3 to S4 based on a simulation verification result, and generating a precise primitive connection network, thereby realizing efficient, rapid and precise identification of the electrical wiring diagram.

The invention relates to an electric box wiring cross detection method and device, computer equipment and a storage medium. The method comprises the following steps: inputting an electric box wiring image into a YOLO backbone network to obtain N-1 feature images output by N-1 continuous first cross-stage local layers and a target image output by a spatial pyramid rapid pooling layer in the YOLO backbone network; inputting a target image into the global context attention network to obtain an Nth feature image, and determining the combination of the N-1 feature images and the Nth feature image as N feature images; inputting N feature images into the path aggregation feature pyramid network to obtain N to-be-detected images; and detecting the N to-be-detected images to obtain a detection result of wiring intersection. By adopting the method, the accuracy of wiring cross detection of the electric box can be improved.

The application belongs to the field of smart grid and computer vision, and particularly relates to a power station wiring diagram text robust generalization detection and recognition method based on an improved SwinTextSpotter v2. The method comprises the following steps: step 1: inputting an image into a text detection and recognition network based on multi-modal learning for training and prediction, obtaining a shared feature map through a shared feature extraction backbone network, and further inputting the shared feature map into a text detection module to obtain a text detection result and a text feature map; step 2: inputting the text feature map into a visual feature extraction and prediction module to obtain a feature sequence, and then matching the predicted feature sequence with a canonical representation obtained by a character structure feature extraction and prediction module to obtain a recognition result; and the like. The application robustly improves the detection and recognition accuracy of the model for irregular text and Chinese character text, and improves the generalization performance of the text detection and recognition of various types of wiring diagrams.