Electrical equipment metal detection auxiliary system and electrical equipment metal detection method

Abstract

The invention belongs to the technical field of power equipment detection, and particularly relates to a power equipment metal detection auxiliary system and a power equipment metal detection method.The power equipment metal detection auxiliary system comprises a nameplate recognition module, a data management module, a judgment module and an automatic report generation module; the equipment nameplate image is acquired through the mobile terminal, and the equipment model and nameplate information are automatically identified, so that the detection efficiency is improved; nameplate information is automatically matched with a preset equipment standard library, so that the accuracy and consistency of metal detection information are ensured, and errors caused by human negligence are reduced; the collected detection data is compared with the pre-stored detection standard data, so that the qualification of metal detection is objectively judged, and the subjectivity and errors of manual judgment are avoided; and the pre-stored corresponding metal detection report template is automatically called according to the nameplate information, so that the uniformity and normalization of report formats are ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power equipment detection, and particularly relates to a power equipment metal detection auxiliary system and a power equipment metal detection method. BACKGROUND

[0002] In the whole life cycle management of the power system, the metal detection of the electrical equipment of the substation occupies a pivotal position. The performance of the metal components is directly related to the reliability, safety and economy of the equipment, and is a key factor for guaranteeing the stable operation of the power grid. With the rapid development of the power industry, the complexity of the equipment is continuously improved, the intelligent transformation is accelerated, and the new construction and expansion projects of the substation are increasing, which puts forward higher requirements for the metal detection work.

[0003] However, in the traditional mode, the metal detection work has many disadvantages. First, the detection personnel need to manually formulate a detection plan, consult paper standards, and record data item by item. This process is not only time-consuming and laborious, but also inefficient and prone to errors. Secondly, since there are many detection projects, there are various corresponding report templates. Manual filling of the report not only consumes time, but also easily leads to problems such as non-uniformity of the report format and inaccuracy of the content due to negligence. In addition, the traditional mode is also prone to missed detection and misjudgment, which seriously affects the quality and efficiency of the metal detection work. SUMMARY

[0004] In view of the above deficiencies of the prior art, the application provides a power equipment metal detection auxiliary system and a power equipment metal detection method. The collected detection data is compared with the pre-stored detection standard data to objectively determine the qualification of the metal detection, avoiding the subjectivity and errors of manual determination. The corresponding metal detection report template is automatically retrieved according to the nameplate information, ensuring the uniformity and standardization of the report format.

[0005] In a first aspect, the application provides a power equipment metal detection auxiliary system, comprising: a nameplate recognition module configured to receive and process an equipment nameplate image of a power equipment to be detected, and output nameplate information and corresponding metal detection information of the power equipment to be detected; the metal detection information includes metal detection items, instruments used for detection, detection positions and detection methods; a data management module configured to receive and store detection data in real time, and synchronize the detection data with a cloud database; a determination module configured to compare real-time detection data with pre-stored detection standard data to determine the qualification of metal detection; The report automatic generation module is configured to call a pre-stored corresponding metal detection report template from a standard template library based on the nameplate information of the equipment to be detected, and dynamically fill the nameplate information of the equipment to be detected, detection data and metal detection determination results into the called metal detection report template to generate a corresponding standardized metal detection report.

[0006] Further improvements of the technical solution include that the nameplate recognition module comprises: The image acquisition unit is configured to capture the equipment nameplate image through the configured camera; The image preprocessing unit is configured to preprocess the equipment nameplate image collected by the mobile terminal, and the preprocessing includes illumination equalization, deblurring processing and distortion correction processing. The text recognition unit is configured to locate the text region of the preprocessed equipment nameplate image by using the YOLOv8 model, and call a pre-stored OCR engine to recognize the characters in the located text region to extract the nameplate information of the equipment. The data matching unit is configured to associate the extracted nameplate information of the equipment with a pre-stored equipment standard library to output the corresponding metal detection information of the equipment to be detected.

[0007] Further improvements of the technical solution include that the data management module comprises: The data acquisition interface is configured to call a pre-stored JSON Schema to define a detection data form structure according to the extracted metal detection information, and dynamically generate an input interface. The cloud synchronization unit is configured to use the MQTT protocol to upload the collected detection data to a cloud database in real time. The local storage unit is configured to cache data when the network is disconnected, and automatically synchronize the data to the cloud database when the network is restored.

[0008] Further improvements of the technical solution include that the determination module is configured to: compare the real-time detection data set item by item, and if the following condition is met: the detection data is determined to be qualified, otherwise the detection data is marked as abnormal; wherein S is a pre-stored corresponding industry standard threshold value, is the i-th measured value detected according to the metal detection information, and the industry standard threshold value S can be dynamically adjusted according to the type of the equipment.

[0009] Further improvements of the technical solution include that the report automatic generation module comprises: The template matching unit is configured to call corresponding PDF and Word dual-format templates from a standard template library according to the equipment model in the nameplate information. The data filling engine is configured to dynamically fill the nameplate information, detection data and metal detection determination results into the specified fields of the templates. A format conversion unit is configured to generate a standardized metal detection report conforming to the power industry specification from the filled-in template and support local download and cloud storage.

[0010] In a second aspect, the present application provides a power equipment metal detection method based on the power equipment metal detection auxiliary system of any one of the above, comprising: Step S1, acquiring the equipment nameplate image of the power equipment to be detected through the mobile terminal, identifying the nameplate information containing the equipment model based on the acquired equipment nameplate image, and matching with the pre-set equipment standard library to obtain the corresponding metal detection information; the metal detection information includes metal detection items, instruments used for detection, detection parts and detection methods; Step S2, collecting the detection data obtained by metal detection according to the metal detection information and uploading to the cloud; Step S3, determining whether the collected detection data is qualified and marking the abnormal detection data; Step S4, retrieving the pre-stored corresponding metal detection report template from the standard template library according to the nameplate information, and dynamically filling the nameplate information, detection data and metal detection determination result of the equipment to be detected into the retrieved metal detection report template to generate the corresponding standardized metal detection report.

[0011] Further improvement of the technical solution is that step S1 comprises: Step S11, preprocessing the acquired equipment nameplate image, including: adopting a pre-stored histogram equalization algorithm to perform illumination equalization processing on the acquired equipment nameplate image to adjust the image contrast; processing the equipment nameplate image after adjusting the image contrast by a pre-stored Wiener filter algorithm to eliminate image blur; detecting the edge of the equipment nameplate image based on Hough transform, and applying a pre-stored perspective transform algorithm to correct geometric distortion; Step S12, performing text region positioning on the preprocessed equipment nameplate image based on a pre-stored YOLOv8 model to output text bounding box coordinates; Step S13, calling a pre-stored OCR engine to perform character recognition on the positioned text region to extract the nameplate information containing the equipment model; Step S14, performing fuzzy matching of the equipment model with the equipment types recorded in the pre-set equipment standard library, and outputting the corresponding metal detection information after successful matching.

[0012] Further improvement of the technical solution is that step S2 comprises: According to the detection items in the metal detection information, a pre-set JSON Schema definition data form structure is called to dynamically generate an input interface matching the detection items; Collecting detection data obtained by metal detection according to the metal detection information, entering the collected detection data into the generated input interface, and performing real-time verification on the input detection data according to the preset verification rule, triggering an abnormal prompt and prohibiting the submission of the input data when the verification fails; After the verification is completed, the detection data that passes the verification is uploaded to the cloud through the MQTT protocol.

[0013] Further improvement of the technical solution is that step S3 includes: Step S31, extracting the detection rule corresponding to the current device from the cloud database to obtain the preset standard value S; Step S32, comparing the real-time detection data set Determine whether the detection data is qualified, and the determination formula is: ; Wherein, S is the pre-stored corresponding industry standard threshold, is the measured value detected according to the metal detection information, and the industry standard threshold S can be dynamically adjusted according to the device type. Step S33, visualizing the detection points corresponding to the unqualified detection data, and the marking method is to highlight the abnormal value in the generated standardized metal detection report and generate a warning log.

[0014] Further improvement of the technical solution is that step S4 includes: Step S41, based on the device model and metal detection item type in the nameplate information, retrieve the corresponding metal detection report template from the standard template library through the pre-stored keyword matching algorithm; Step S42, automatically fill in the basic information bar in the template with the identified nameplate information; the nameplate information includes the device manufacturer, device model, and device number; batch import the received detection values according to the template preset table structure; according to the received determination result, automatically trigger the generation of the corresponding conclusion paragraph in the template, and if it is determined to be unqualified, add an abnormal data mark; Step S42, use the pre-stored DOCX document object model or PDF generation tool to automatically layout according to the template preset format; and automatically embed the detection site diagram when generating the report, and the source of the diagram includes the built-in graphics in the template or the corresponding device structure diagram retrieved from the pre-stored graphics library according to the nameplate information.

[0015] The beneficial effects of the present application are: Efficiency improvement: Through the nameplate recognition module, the system can automatically collect and process the device nameplate image, quickly output the nameplate information and corresponding metal detection information of the power equipment to be detected, avoid the tedious process of manually formulating detection plans and consulting paper standards, and greatly save time. At the same time, the data management module realizes real-time collection and cloud synchronization of detection data, ensuring the timeliness and accuracy of data, and further improving the detection efficiency.

[0016] Enhanced data traceability: The data management module not only collects and stores detection data in real time, but also synchronizes data to the cloud database, ensuring data integrity and consistency. When needed, historical data can be easily traced and queried, providing strong support for equipment maintenance and management.

[0017] Improved decision accuracy: The decision module compares the collected real-time detection data with the pre-stored detection standard data to automatically determine the qualification of metal detection, avoiding the subjectivity and errors of manual judgment. At the same time, the decision module can dynamically adjust the industry standard threshold according to the equipment type, improving the accuracy and flexibility of the decision.

[0018] Standardized report format: The report automatic generation module retrieves the pre-stored corresponding metal detection report template from the standard template library based on the nameplate information of the equipment to be detected, and dynamically fills in the nameplate information, detection data and metal detection judgment results of the equipment to be detected into the template to generate a standardized metal detection report. This not only avoids the tediousness and error-prone of manual report filling, but also ensures the uniformity and standardization of the report format, improving the readability and credibility of the report. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0020] Figure 1 The schematic block diagram of the system of one embodiment of the present application.

[0021] Figure 2 The schematic flow chart of the method of one embodiment of the present application.

[0022] 210 is a nameplate recognition module, 220 is a data management module, 230 is a decision module, and 240 is a report automatic generation module. DETAILED DESCRIPTION

[0023] In order to make the objectives, characteristics and advantages of the present application more obvious and easy to understand, the technical solutions in the present application will be clearly and completely described below in combination with the drawings in the specific embodiments. Obviously, the embodiments described below are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terminology used in the description of the present application herein only for the purpose of describing specific embodiments and is not intended to limit the present application.

[0025] The power equipment metal detection auxiliary system provided by the present application can include a plurality of function modules composed of computer program segments. The computer programs of each program segment in the power equipment metal detection auxiliary system can be stored in the memory of a computer device and executed by at least one processor.

[0026] In the present embodiment, the power equipment metal detection auxiliary system can be divided into a plurality of function modules according to the functions it performs, as shown in the figure. The function modules can include a nameplate recognition module 210, a data management module 220, a judgment module 230 and a report automatic generation module 240. The module referred to in the present application refers to a series of computer program segments that can be executed by at least one processor and can complete a fixed function, which are stored in the memory. In the present embodiment, the functions of each module will be described in detail in subsequent embodiments. Figure 1

[0027] The nameplate recognition module is configured to receive and process the equipment nameplate image of the power equipment to be detected, and output the nameplate information and corresponding metal detection information of the power equipment to be detected. The metal detection information includes metal detection items, instruments used for detection, detection positions and detection methods. The data management module is configured to receive detection data in real time and store the detection data, and synchronize the detection data with a cloud database. The judgment module is configured to compare the real-time detection data with pre-stored detection standard data to determine the metal detection qualification. The report automatic generation module is configured to retrieve a pre-stored corresponding metal detection report template from a standard template library based on the nameplate information of the equipment to be detected, and dynamically fill the nameplate information of the equipment to be detected, the detection data and the metal detection judgment result into the retrieved metal detection report template to generate a corresponding standardized metal detection report.

[0028] Specifically, the nameplate recognition module includes: An image acquisition unit is configured to capture the equipment nameplate image through a configured camera; ​An image preprocessing unit is configured to preprocess the equipment nameplate image collected by the mobile terminal, and the preprocessing includes light equalization, deblurring processing, and distortion correction processing. A character recognition unit, the YOLOv8 model performs text region positioning on the preprocessed equipment nameplate image, and calls a pre-stored OCR engine to perform character recognition on the positioned text region to extract the nameplate information of the equipment. A data matching unit is configured to associate the extracted equipment nameplate information with a pre-stored equipment standard library and output metal detection information corresponding to the to-be-detected power equipment.

[0029] In a substation scenario, a mobile terminal (such as an explosion-proof industrial tablet) is equipped with a high-definition camera (resolution ≥ 16 million pixels, supporting automatic focusing). The detection personnel hold the terminal close to the equipment nameplate (distance about 20 cm to 30 cm), trigger the photographing function of the terminal, and the camera quickly captures the nameplate image and stores it as a JPEG format. During image acquisition, the camera automatically adapts to the environment light (such as strong light, shadow scene), avoids overexposure or underexposure through wide dynamic range (WDR) technology, and ensures that the nameplate text and model identification are clear and identifiable.

[0030] The collected original image is sequentially processed in three steps: Light equalization: an adaptive histogram equalization algorithm (CLAHE) is used to divide the image into 8x8 pixel sub-blocks, adjust the contrast block by block, and eliminate the interference of local strong light or shadow on text recognition (such as character blur caused by reflection).

[0031] Deblurring processing: Gaussian filter algorithm (kernel size 5x5) is used to smooth image noise, reduce edge blur caused by shooting jitter or device surface texture, and highlight the text outline.

[0032] Distortion correction: based on the pre-calibrated intrinsic matrix of the camera (obtained by chessboard calibration), the perspective transformation algorithm is used to correct the geometric distortion caused by the shooting angle (such as trapezoidal distortion caused by inclined shooting of the nameplate), and the image is corrected to the orthographic view angle.

[0033] Based on the YOLOv8 target detection model, the preprocessed image is input, and the boundary box coordinates (such as the upper left corner coordinates (x1, y1) and the lower right corner coordinates (x2, y2)) containing the nameplate text region are output. The model is pre-trained on a power equipment nameplate dataset and can recognize key regions such as "model", "manufacturer", and "production date".

[0034] Extract the text area image within the bounding box, call the pre-integrated OCR (Optical Character Recognition) engine for multi-language recognition (supporting Chinese, English, and mixed text of numbers), and output structured text data (such as device model "KYN28-12" and manufacturer "XX Electrical Co., Ltd.".

[0035] Use the recognized device model (such as "KYN28-12") as a keyword to search the local or cloud device standard library (stored in a SQL (Structured Query Language) database with fields including device model, detection item, instrument, position, and method). After a successful match, automatically retrieve the corresponding metal detection information: Detection item: such as "copper bar conductivity detection" and "shell plating thickness detection"; Detection instrument: such as eddy current conductivity meter and ultrasonic thickness gauge; Detection position: such as "cabinet bus room copper bar connection" and "equipment shell weld area"; Detection method: such as "eddy current method" and "ultrasonic pulse reflection method".

[0036] If there is no complete matching model in the standard library, the system automatically triggers fuzzy matching (such as matching to the "KYN28 series" general detection rule) and prompts the detection personnel for confirmation. The detection personnel perform metal detection on the power equipment to be detected according to the output metal detection information.

[0037] In addition, the data management module includes: Data acquisition interface, call pre-stored JSON Schema to define detection data form structure according to extracted metal detection information, and dynamically generate input interface; Cloud synchronization unit, use MQTT protocol to upload collected detection data to cloud database in real time; Local storage unit, configured to cache data when the network is disconnected, and automatically synchronize to the cloud database after the network is restored.

[0038] In the detection application program of the mobile terminal, pre-store JSON Schema templates corresponding to various metal detection items (such as conductivity, plating thickness, and crack detection). When the nameplate recognition module outputs metal detection information (such as "switch cabinet copper bar conductivity detection"), the data acquisition interface automatically matches the corresponding JSON Schema, and dynamically generates an input interface that meets the detection requirements: Dynamic form field generation: Generate corresponding input components according to the detection item type (such as numerical, text, file type). For example, generate a digital input box for "conductivity detection" and automatically associate the unit "% IACS (International Annealed Copper Standard)". For "crack detection", generate a picture upload portal and text description box.

[0039] Field verification rule integration: Embed data verification logic in the form, such as limiting the conductivity value range to 0-100 and requiring the device number format to contain a "-" separator. If the input is not legal, a real-time pop-up prompt will be given (such as "The value is out of the qualified range, please re-enter").

[0040] Default value filling: Automatically fill in fixed information such as device model and detection instrument in the nameplate recognition results (such as device model "KYN28-12" and instrument "eddy current conductivity meter"), reducing the workload of manual input by detection personnel.

[0041] After the detection personnel completes data entry and submits, the cloud synchronization unit packages the data into lightweight messages through the MQTT (Message Queuing Telemetry Transport) protocol and transmits them to the cloud database: Message topic definition: Construct the message topic with "device type / device number / detection item" (such as "switchgear / 20250601-KYN28-03 / conductivity"), which facilitates cloud storage and retrieval by device dimension.

[0042] Real-time synchronization mechanism: Trigger a synchronization request immediately after data submission, transmit through an SSL (Secure Sockets Layer) encrypted channel (to ensure data transmission security), and then parse the message and write it to the "detection_records" table in the MySQL (Structured Query Language) database, including device ID, detection time, detection value, and operator.

[0043] Status feedback: The terminal interface displays the synchronization status in real time (such as "Data uploading... ", "Upload successful", "Upload failed, please try again"), allowing detection personnel to visually confirm the data processing result.

[0044] When the mobile terminal is in a network-free environment (such as the underground layer of a substation), detection data is temporarily stored in the terminal local database (using SQLite lightweight database): Cache strategy: the local database table structure is completely consistent with the cloud, and each record is attached with a "synchronization state" field (the values are "not synchronized", "synchronizing", "synchronized"). When the network is disconnected, the data is automatically marked as "not synchronized" and stored locally.

[0045] After the network is restored, the local storage unit periodically scans the unsynchronized records (default scanning every 5 minutes), and re-sends them to the cloud in chronological order through the MQTT protocol. If synchronization fails for 3 consecutive times, the terminal is triggered to remind (such as "partial data synchronization failure, please check the network").

[0046] When the cloud receives the data, it checks the uniqueness through the combination of "device ID + detection time + detection item" to avoid repeated storage (such as multiple transmissions caused by network fluctuations).

[0047] The invention based on JSON Schema dynamic form ensures that different detection items adopt a unified data structure (such as all conductivity detection includes "value-unit-time" field), solves the problem of chaotic format of traditional manual recording, and improves the data specification.

[0048] In addition, the determination module is configured to: compare each item in the real-time detection data set If it meets The detection data is determined to be qualified, otherwise it is marked as abnormal; wherein S is a pre-stored threshold value corresponding to the industry standard, is the measured value detected according to the metal detection information of the ith item, and the industry standard threshold S can be dynamically adjusted according to the equipment type.

[0049] After the determination module receives the detection data synchronized by the cloud (such as the copper bar conductivity detection value 97.5% IACS of device number "20250529-KYN28-01"), first extract the detection rule corresponding to the device from the cloud database: Device type identification: according to the device model (such as "KYN28-12") in the nameplate information, the device type is matched as "high-voltage switch cabinet"; Industry standard matching: according to the device type and detection item (such as "copper bar conductivity"), the pre-stored industry standard threshold is called, such as conductivity ≥ 96.6% IACS; Dynamic adjustment mechanism: if the device is a special model or a modified device (such as a switch cabinet after corrosion treatment), the detection personnel are allowed to manually adjust the threshold through the terminal (permission verification is required), and the adjusted threshold is automatically synchronized to the cloud and recorded in the log.

[0050] For the real-time detection data set (such as the conductivity values of 3 detection points of the same copper bar: 97.5%, 96.8%, 96.2%), the following operations are performed: Value comparison: compare each measured value Compared with the standard threshold S (such as 96.6% IACS) item by item: 97.5% ≥ 96.6%, output qualified; 96.8% ≥ 96.6%, output qualified; 96.2% < 96.6%, output unqualified.

[0051] Abnormal marking: generate an abnormal record for unqualified data (such as 96.2%), including: equipment number, detection item, measurement point position (such as "busbar room A phase copper bar middle section"); measured value, standard value, deviation amplitude (such as "-0.4% IACS"); marking time and judge personnel (the system automatically fills in the current detection personnel information).

[0052] Historical data association: if the historical detection data of the same item of the same equipment fluctuates (such as the previous detection value is 97.0%), the judgment module automatically triggers trend analysis, and if the detection value decreases by more than 5% for two consecutive times, even if the current value is qualified, it is marked as "trend abnormality" and prompted for review.

[0053] Combined condition judgment: for detection items that need to combine multiple parameters (such as metal material detection needs to meet hardness and tensile strength at the same time), the judgment module uses the "logical and" rule (all parameters are qualified, then the whole is qualified), and any parameter abnormality triggers the whole abnormality marking.

[0054] The present application dynamically matches industry standards based on equipment type (such as different conductivity thresholds for transformers and circuit breakers), which improves the accuracy of actual measurement and judgment. The item-by-item comparison mechanism ensures that each measurement point data can be traced back, and the abnormal marking is directly related to the specific detection part (such as "switch cabinet B phase copper bar connection"), which reduces the detection personnel's troubleshooting time.

[0055] Further, the report automatic generation module comprises: A template matching unit retrieves corresponding PDF and Word dual-format templates from a standard template library according to the equipment model in the nameplate information; A data filling engine dynamically fills the nameplate information, detection data and metal detection judgment results to the specified fields of the template; A format conversion unit is configured to generate a standardized metal detection report conforming to the power industry specifications from the filled template, and supports local download and cloud storage.

[0056] A template library is pre-established in the cloud server according to equipment models (such as KYN28-12 and S11-M-100 / 10) and detection items (such as conductivity and plating thickness), and each template contains both PDF and Word formats. For example, the template for "KYN28-12 switch cabinet conductivity detection" is pre-set with table structure, detection clauses (such as DL / T 438-2016), and conclusion expression paradigm.

[0057] The template that completely corresponds to the "equipment model + detection item" is matched preferentially (such as inputting the equipment model "KYN28-12" and the item "copper bar conductivity" to directly call the exclusive template); if no completely matched item is found, the general template is matched according to "equipment type + detection item" (such as the general template for "high-voltage switch cabinet + conductivity"); manual selection of the template by the detection personnel is supported (such as when special detection requirements exist), and the selection record is automatically synchronized to the template library for subsequent optimization.

[0058] Field mapping rules: Placard information: the equipment model, manufacturer, number, etc. are filled in the "basic information" column of the template (such as the "equipment model" text box of the Word template is automatically associated with the "device_model" field in the detection data); Detection data: the measured values (such as the conductivity data 97.5%, 96.8%, and 96.2% of 3 measuring points) are batch imported according to the pre-set table structure of the template (such as "measuring point number - detection value - unit"), and the unit is automatically matched ("% IACS"); Determination result: if all detection values are qualified, the "detection result is qualified" conclusion is automatically generated; if there is an abnormal value, the "abnormal situation explanation" paragraph in the template is triggered to be filled (such as "the conductivity of measuring point 3 is 96.2%, which is lower than the standard value 96.6%"), and the abnormal value is marked in red font.

[0059] Dynamic chart generation: for continuous detection data (such as 5 times of conductivity detection of the same equipment within one year), a line chart is automatically inserted in the Word template (X axis for detection time and Y axis for conductivity value), and a static chart is embedded in the PDF template to intuitively show the data trend.

[0060] Standardized layout example: Word format: the standard font of the electric power industry (such as FangSong_GB2312, four-point font) is adopted, the paragraph spacing is 1.5 times, the table of contents and page numbers are automatically generated; PDF format: the Word document is converted into PDF through a virtual printer, the "printing quality" mode is enabled to ensure the legibility of the text, and an electronic watermark (such as "XX Electric Power Company Detection Report") is added to prevent tampering.

[0061] The generated report is stored in the terminal local folder (such as " / reports / 202505") and the cloud server (stored in the path of "device number + detection date"), and the detection personnel can download and print at any time through the terminal APP or Web.

[0062] As shown in Figure 2 The power equipment metal detection method based on the power equipment metal detection auxiliary system provided by the application comprises the following steps: Step S1, acquiring the equipment nameplate image of the power equipment to be detected through the mobile terminal, identifying the nameplate information containing the equipment model based on the acquired equipment nameplate image, and matching the identified nameplate information with the preset equipment standard library to obtain corresponding metal detection information; the metal detection information comprises the metal detection item, the instrument used for detection, the detection position and the detection method; Step S2, acquiring the detection data obtained by performing metal detection according to the metal detection information, and uploading the detection data to the cloud; Step S3, determining whether the acquired detection data is qualified and marking the abnormal detection data; Step S4, retrieving the pre-stored corresponding metal detection report template from the standard template library according to the nameplate information, and dynamically filling the nameplate information, the detection data and the metal detection determination result of the equipment to be detected into the retrieved metal detection report template to generate a corresponding standardized metal detection report.

[0063] In order to facilitate the understanding of the application, the principle of the power equipment metal detection method of the application is described further in combination with the process of detecting the metal of the power equipment based on the power equipment metal detection auxiliary system in the embodiment.

[0064] Specifically, step S1 comprises: Step S11, preprocessing the acquired equipment nameplate image, comprising: adopting a pre-stored histogram equalization algorithm to perform illumination equalization processing on the acquired equipment nameplate image, and adjusting the contrast of the image; processing the equipment nameplate image after adjusting the contrast of the image by a pre-stored Wiener filter algorithm to eliminate image blur; detecting the edge of the equipment nameplate image based on Hough transformation, and applying a pre-stored perspective transformation algorithm to correct geometric distortion; Step S12, performing text region positioning on the preprocessed equipment nameplate image based on a pre-stored YOLOv8 model, and outputting text bounding box coordinates; Step S13, calling a pre-stored OCR engine to recognize characters in the positioned text region, and extracting the nameplate information containing the equipment model; Step S14, the device model is matched with the device type recorded in the preset device standard library, and the corresponding metal detection information is output after successful matching.

[0065] In addition, step S2 includes: According to the detection items in the metal detection information, the preset JSON Schema definition data form structure is called to dynamically generate an input interface matched with the detection items; According to the detection items in the metal detection information, the preset JSON Schema definition data form structure is called to dynamically generate an input interface matched with the detection items; After verification, the detection data that passes the verification is uploaded to the cloud through the MQTT protocol.

[0066] In addition, step S3 includes: Step S31, the detection rule corresponding to the current device is extracted from the cloud database, and the preset standard value S is obtained; Step S32, the real-time detection data set is compared item by item to determine whether the detection data is qualified, and the determination formula is: ; Wherein, S is the pre-stored corresponding industry standard threshold, is the i-th measured value detected according to the metal detection information, and the industry standard threshold S can be dynamically adjusted according to the device type; Step S33, the detection points corresponding to the unqualified detection data are visually marked, and the marking method is to highlight the abnormal value in the generated standardized metal detection report and generate a warning log.

[0067] Further, step S4 includes: Step S41, based on the device model in the nameplate information and the metal detection item type, the corresponding metal detection report template is retrieved from the standard template library through the pre-stored keyword matching algorithm; Step S42, the recognized nameplate information is automatically filled into the basic information column in the template; the nameplate information includes the device manufacturer, the device model, and the device number; the received detection values are batch imported according to the template preset table structure; according to the received determination result, the generation of the corresponding conclusion paragraph in the template is automatically triggered, and if the determination is unqualified, the abnormal data mark is added; Step S42, the pre-stored DOCX document object model or PDF generation tool is used to automatically layout according to the template preset format; and the detection part schematic diagram is automatically embedded when generating the report, and the schematic diagram sources include the built-in graphics in the template or the corresponding device structure diagram retrieved from the preset graphics library according to the nameplate information.

[0068] The application avoids the cumbersome process of manual checking and recording by collecting the equipment nameplate image through the mobile terminal and automatically identifying the equipment model and nameplate information, significantly improving the detection efficiency. The automatic matching of nameplate information and pre-set equipment standard library ensures the accuracy and consistency of metal detection information, reducing errors caused by human negligence. The real-time uploading of the collected detection data to the cloud ensures the timeliness and accessibility of the data, facilitating subsequent data analysis and processing. Cloud storage provides a safe backup of data, preventing data loss due to local device failure. The application automatically compares the collected detection data with the pre-stored detection standard data to objectively determine the eligibility of metal detection, avoiding the subjectivity and errors of manual judgment. Labeling of abnormal detection data helps quickly locate problems and improves the relevance and effectiveness of detection work. According to the nameplate information, the corresponding pre-stored metal detection report template is automatically retrieved, ensuring the uniformity and standardization of the report format.

[0069] Although the application has been described in detail with reference to the preferred embodiments, the application is not limited thereto. Those skilled in the art can make various equivalent modifications or replacements to the embodiments of the application without departing from the spirit and essence of the application, and these modifications or replacements shall be within the scope of the application. Any person skilled in the art within the technical scope disclosed by the application can easily think of changes or replacements, which shall be covered within the protection scope of the application.

Claims

1. A metal detection auxiliary system for power equipment, characterized in that, include: The nameplate recognition module is used to receive and process the nameplate image of the power equipment to be tested, and output the nameplate information and corresponding metal detection information of the power equipment to be tested. Metal detection information includes the metal detection items, the instruments used for detection, the detection sites, and the detection methods. The data management module is used to receive and store detection data in real time, and synchronize the detection data with the cloud database. The judgment module is used to compare real-time detection data with pre-stored detection standard data to determine the pass rate of metal detection; The automatic report generation module is used to retrieve the corresponding pre-stored metal testing report template from the standard template library based on the nameplate information of the device under test, and dynamically fill the nameplate information, test data and metal testing judgment results of the device under test into the retrieved metal testing report template to generate the corresponding standardized metal testing report.

2. The auxiliary system for metal detection of power equipment according to claim 1, characterized in that, The nameplate recognition module includes: An image acquisition unit is used to capture images of the device nameplate using a configured camera; The image preprocessing unit is configured to preprocess the device nameplate image acquired by the mobile terminal. The preprocessing includes illumination equalization, deblurring, and distortion correction. The text recognition unit uses the YOLOv8 model to locate the text region in the pre-processed device nameplate image; and calls the pre-stored OCR engine to perform character recognition on the located text region to extract the device nameplate information. The data matching unit is used to associate the extracted equipment nameplate information with the preset equipment standard library and output the metal detection information corresponding to the power equipment to be tested.

3. The auxiliary system for metal detection of power equipment according to claim 1, characterized in that, The data management module includes: The data acquisition interface calls a pre-stored JSON Schema to define the structure of the detection data form based on the extracted metal detection information and dynamically generates the input interface; The cloud synchronization unit uses the MQTT protocol to upload the collected detection data to the cloud database in real time; The local storage unit is configured to cache data when the network is down and automatically synchronize it to the cloud database when the network is restored.

4. The auxiliary system for metal detection of power equipment according to claim 1, characterized in that, The determination module is configured to: for the real-time detection data set perform item-by-item comparison. If it meets then it is determined that the detection data is qualified; otherwise, the detection data is marked as abnormal. Among them, S is the pre-stored corresponding industry standard threshold, is the measured value detected for the i-th item according to the metal detection information, and the industry standard threshold S can be dynamically adjusted according to the equipment type.

5. The auxiliary system for metal detection of power equipment according to claim 1, characterized in that, The automatic report generation module includes: The template matching unit retrieves the corresponding PDF and Word dual-format templates from the standard template library based on the device model information in the nameplate. The data population engine dynamically populates the nameplate information, test data, and metal detection results into the specified fields of the template; The format conversion unit is configured to generate a standardized metal testing report that conforms to power industry standards from the completed template, and supports local download and cloud storage.

6. A method for detecting metal in power equipment based on the auxiliary system for metal detection of power equipment according to any one of claims 1-5, characterized in that, include: Step S1: Collect the equipment nameplate image of the power equipment to be tested through a mobile terminal, identify the nameplate information containing the equipment model based on the collected equipment nameplate image, and match it with the preset equipment standard library to obtain the corresponding metal detection information; the metal detection information includes metal detection items, detection instruments, detection parts and detection methods; Step S2: Collect the detection data obtained from metal detection based on the metal detection information and upload it to the cloud; Step S3: Determine whether the collected detection data is qualified and mark abnormal detection data; Step S4: Retrieve the pre-stored corresponding metal test report template from the standard template library based on the nameplate information, and dynamically fill the nameplate information, test data and metal test judgment results of the equipment to be tested into the retrieved metal test report template to generate the corresponding standardized metal test report.

7. The method for metal detection in power equipment according to claim 6, characterized in that, Step S1 includes: Step S11: Preprocess the acquired equipment nameplate image, including: The acquired equipment nameplate image is processed for illumination equalization using a pre-stored histogram equalization algorithm to adjust the image contrast. The image of the device nameplate after image contrast adjustment is processed by a pre-stored Wiener filtering algorithm to eliminate image blur. Based on the Hough transform, the edges of the equipment nameplate image are detected, and a pre-stored perspective transformation algorithm is applied to correct geometric distortion. Step S12: Based on the pre-stored YOLOv8 model, perform text region localization on the pre-processed device nameplate image and output the text bounding box coordinates; Step S13: Call the pre-stored OCR engine to perform character recognition on the located text area and extract the nameplate information containing the device model. Step S14: Perform a fuzzy match between the equipment model and the equipment type recorded in the preset equipment standard library, and output the corresponding metal detection information after a successful match.

8. The method for metal detection in power equipment according to claim 6, characterized in that, Step S2 includes: Based on the detection items in the metal detection information, the system calls the preset JSON Schema to define the data form structure and dynamically generates an input interface that matches the detection items. The system collects detection data obtained from metal detection based on metal detection information, enters the collected detection data into the generated input interface, and performs real-time verification on the input detection data according to preset verification rules. If the verification fails, an error message is triggered and the submission of input data is prohibited. After verification, the verified test data is uploaded to the cloud via the MQTT protocol.

9. The method for metal detection in power equipment according to claim 6, characterized in that, Step S3 includes: Step S31: Extract the detection rules corresponding to the current device from the cloud database and obtain the preset standard value S; Step S32: Process the real-time detection data set A step-by-step comparison is performed to determine whether the test data is qualified. The determination formula is as follows: ； Where S is the pre-stored corresponding industry standard threshold. The industry standard threshold S is the measured value detected based on the metal detection information for the i-th item; the threshold value S can be dynamically adjusted according to the equipment type. Step S33: Visually mark the detection points corresponding to the non-conforming detection data. The marking method is to highlight the abnormal values ​​in the generated standardized metal detection report and generate an early warning log.

10. The method for metal detection in power equipment according to claim 9, characterized in that, Step S4 includes: Step S41: Based on the equipment model and metal detection item type in the nameplate information, retrieve the corresponding metal detection report template from the standard template library using a pre-stored keyword matching algorithm; Step S42: Automatically fill the identified nameplate information into the basic information field of the template; the nameplate information includes the equipment manufacturer, equipment model, and equipment number; import the received test values ​​in batches according to the preset table structure of the template; based on the received judgment results, automatically trigger the generation of the corresponding conclusion paragraph in the template, and add an abnormal data mark if the judgment is unqualified. Step S42: Use a pre-stored DOCX document object model or PDF generation tool to automatically format the document according to the template's preset format; and automatically embed a schematic diagram of the detected part when generating the report. The schematic diagram can be sourced from the template's built-in graphics or the corresponding equipment structure diagram retrieved from the preset image library based on the nameplate information.

Images