Visual process management system and method based on primary equipment starting operation

By designing a visual process management system, using deep learning models to automatically identify equipment damage and warning slogans, and collecting electrical parameters, the problem of lack of automation and data sharing in the inspection process in the existing technology is solved, and the safety and efficiency of equipment startup operations are improved.

CN119067303APending Publication Date: 2024-12-03GUANGXI POWER GRID CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411137967.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In the prior art, the inspection process before the start of a device is lacking automation and integration, resulting in insufficient data recording, incomplete sharing mechanism, and lack of real-time monitoring and early warning functions, which affects the safety and efficiency of equipment operation.

Method used

Design a visual process management system based on a primary device startup operation, including hardware inspection unit, index inspection unit, record and storage unit, identify equipment damage and warning slogans through deep learning models, collect electrical parameters, and realize automated inspection and data sharing. And visual operations and alarms are performed through the terminal.

Benefits of technology

It improves the safety and efficiency of equipment startup operations, reduces human errors, realizes real-time monitoring and data sharing, and enhances the collaborative efficiency of different departments and teams.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067303A_ABST
    Figure CN119067303A_ABST
Patent Text Reader

Abstract

The invention discloses a visual process management system and method based on primary equipment starting operation, and solves the problems of insufficient data recording and imperfect sharing mechanism caused by lack of automation and integration of an inspection process in the prior art. According to the invention, the hardware checking unit can check the perfection degree of hardware setting in the starting process of the primary equipment, including whether the primary equipment is damaged or not, and whether warning slogans are set on the appearance and installation environment of the primary equipment according to requirements or not; the deviation value between each electrical parameter or mechanical parameter and the standard parameter can be checked through the index checking unit, and then whether an alarm signal is sent to the terminal is determined based on the output results of the hardware checking unit and the index checking unit, so that an operator can carry out corresponding processing conveniently. In addition, the recording and storing unit can record various process data and result data of the hardware checking unit and the index checking unit, so that the terminal can conveniently carry out corresponding processing based on the data in the operation process of overhauling and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment performance detection, and in particular to a visual process management system and method based on a one-time equipment startup operation. Background Art

[0002] In the field of power systems, the startup operation of primary equipment is a key step to ensure the safe and stable operation of the power grid. Primary equipment, including transformers, generators, circuit breakers, etc., are core components in the power system. Before starting these devices, it is crucial to conduct a detailed inspection, which not only involves the appearance damage inspection of the equipment to ensure that there is no visible damage or defects, but also includes the integrity inspection of the warning signs to ensure the safety of operators. In addition, the detection of electrical parameters such as circuit voltage and current is also essential, which are directly related to the operating status and energy efficiency performance of the equipment. Comprehensive inspection can prevent potential failures, reduce the risk of equipment damage, and ensure personnel safety and long-term stable operation of equipment.

[0003] Although the inspection before the start-up of a primary equipment is crucial, the existing inspection process often lacks comprehensiveness and systematicness. In many cases, the inspection work relies almost entirely on manual operations, which is not only inefficient but also easily affected by human factors, resulting in inaccurate or missed inspection results. In addition, the inspection data is often not effectively recorded and shared, making problem tracking and subsequent analysis difficult. The existing system lacks automated data collection and analysis tools, and cannot achieve real-time monitoring and early warning, which is an obvious shortcoming in a rapidly changing industrial environment. More importantly, the lack of standardized inspection processes and sharing mechanisms limits the collaborative efficiency between different departments and teams, affecting the response speed and decision-making quality of the entire organization. Therefore, there is an urgent need to develop an integrated, automated inspection and data sharing system to improve the safety and efficiency of primary equipment startup operations.

[0004] In view of this, a visual process management system and method based on a one-time device startup operation is needed. Summary of the invention

[0005] In view of the problems in the prior art that the inspection process lacks automation and integration, resulting in insufficient data recording, imperfect sharing mechanism, and lack of real-time monitoring and early warning functions, the present invention provides a visual process management system and method based on the primary equipment startup operation, which can automatically check the hardware facilities and electrical performance of the primary equipment, and visualize and share the data, thereby improving the safety and inspection efficiency of the primary equipment startup operation. The specific technical solution is as follows:

[0006] A visual process management system based on a one-time equipment startup operation, comprising:

[0007] The hardware inspection unit collects a picture of the appearance of a device, identifies whether the appearance of the device is obviously damaged through a pre-trained hardware damage inspection model, and decides whether to issue an alarm signal based on the identification result;

[0008] An index checking unit collects electrical parameters and mechanical parameters in a control loop, wherein the electrical parameters include current and voltage, and the mechanical parameters include vibration and temperature, and compares and analyzes the parameters with standard parameters or historical data. If the degree of deviation exceeds a threshold, an alarm signal is issued;

[0009] A recording and storage unit, connected to the hardware inspection unit and the index inspection unit, for collecting and storing the original data and output results of the operation process of the hardware inspection unit and the index inspection unit;

[0010] The terminal is equipped with a visualization unit, which is connected to the recording and storage unit, the hardware inspection unit and the indicator inspection unit. The visualization unit includes a display module and an alarm module, wherein the display module is used to display the storage data of the recording and storage unit, and the alarm module is used to receive an alarm signal and perform an alarm operation.

[0011] The terminals include several executor terminals and several manager terminals. The executor terminals and the manager terminals are provided with different data access rights, and the executor terminals can input data into the recording and storage unit for reading by the manager terminal or other executor terminals.

[0012] The training process of the hardware damage detection model is as follows:

[0013] Collect pictures of electrical equipment, including pictures of intact electrical equipment and pictures of damaged electrical equipment;

[0014] Resize all images to a uniform size, convert them into grayscale images, and normalize the image pixel values;

[0015] Select the deep learning model framework, design the number of model layers and the parameters of each layer;

[0016] Divide the data into training set, validation set and test set, use the training set data to train the model, use the validation set data to adjust the model parameters, and evaluate the performance of the model on the test set;

[0017] Adjust the learning rate according to the model performance and save the trained model after training is completed.

[0018] The hardware inspection unit is also provided with a pre-trained warning slogan inspection model, and also collects a picture of the surrounding environment of the primary device, identifies whether there is a warning slogan in the installation environment of the primary device through the pre-trained warning slogan inspection model, and decides whether to issue an alarm signal based on the identification result. The training process of the warning slogan inspection model is as follows:

[0019] Collect pictures that include warning slogans and pictures that do not contain warning slogans as negative samples;

[0020] Use image annotation tools to annotate the collected images;

[0021] Resize the image to meet the model input requirements and perform image enhancement operations including rotation, scaling, and color changes;

[0022] Select a deep learning model and use the labeled dataset to train the selected model and adjust the model parameters and hyperparameters;

[0023] Evaluate model performance on the validation set, use confusion matrix to analyze the model's performance on different categories, and optimize the model based on the evaluation results, which may include adjusting the network structure, using data enhancement, and changing the training strategy.

[0024] The hardware inspection unit is also provided with a warning slogan judging module, which performs the following operations:

[0025] Collect a picture of the device appearance and a picture of the device surrounding environment;

[0026] Preprocess the images, including resizing, denoising, and binarization, to improve recognition accuracy;

[0027] Select and use the OCR tool to perform text recognition on the image and extract the text in the image;

[0028] The text recognized by OCR is compared with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued.

[0029] The system also includes an equipment debugging module, which is mounted on the inspector terminal and is provided with an oscilloscope and a power analyzer. The oscilloscope is used to detect the signal waveform, and the power analyzer is used to test the power quality. The signal waveform and power quality status are uploaded through the inspector terminal for sharing with other inspector terminals and manager terminals. Moreover, when the operator determines that there is a problem with the signal waveform or power quality, an alarm signal can be directly sent to the manager terminal through the interface between the equipment debugging module and the inspector terminal.

[0030] A visual process management method based on a one-time equipment startup operation includes the following steps:

[0031] Collect pictures of intact and damaged electrical equipment, and train an electrical equipment recognition model based on the deep learning model framework to identify whether the electrical equipment is damaged;

[0032] Collect pictures containing warning slogans and pictures without warning slogans, and train a warning slogan detection model based on the deep learning model to identify whether the pictures contain warning slogans;

[0033] Collect pictures of the appearance of the primary equipment and the surrounding environment, and identify whether there is any damage to the electrical equipment and whether there is no warning sign based on the electrical equipment recognition model and the warning sign inspection model;

[0034] Send an alarm signal to the terminal when the electrical equipment is damaged or there is no warning sign;

[0035] Collect a picture of the device appearance and a picture of the device surrounding environment;

[0036] Preprocess the images, including resizing, denoising, and binarization, to improve recognition accuracy;

[0037] Select and use the OCR tool to perform text recognition on the image and extract the text in the image;

[0038] The text recognized by OCR is compared with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is sent.

[0039] A computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the visual process management method based on a one-time device startup operation as described above.

[0040] A processor is used to run a program, wherein when the program is running, the visual process management method based on a one-time device startup operation as described above is executed.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The present invention is provided with a hardware inspection unit, an index inspection unit, a recording and storage unit and a terminal, and the terminal is equipped with a visualization unit. The hardware inspection unit can be used to check the degree of perfection of the hardware settings during the startup of a primary device, including whether the primary device is damaged, whether the appearance of the primary device and its installation environment are set with warning signs as required, etc. The index inspection unit can be used to check the deviation values ​​of various electrical parameters or mechanical parameters from the standard parameters during or after the startup of the primary device, and then, based on the output results of the hardware inspection unit and the index inspection unit, it is decided whether to send an alarm signal to the terminal, thereby facilitating the operator to perform corresponding processing. In addition, the recording and storage unit will record various process data and result data of the hardware inspection unit and the index inspection unit, so as to facilitate the terminal to perform corresponding processing based on the data during maintenance and other operations.

[0043] 2. The hardware inspection unit is equipped with a pre-trained hardware damage inspection model and a warning slogan inspection model. The hardware damage inspection model can be used to check whether a device is damaged, and the warning slogan inspection model can be used to check whether a device and its surroundings are equipped with warning slogans. By deploying a camera at the device site and taking pictures of the device and its surroundings, the model can be used to directly identify whether there is an abnormality, without the need for manual identification, saving time and effort.

[0044] 3. The hardware inspection unit is also equipped with a warning slogan evaluation module, which collects a picture of the appearance of the device and a picture of the surrounding environment of the device, performs text recognition on the picture based on the OCR tool and extracts the text in the picture, and compares the text recognized by the OCR with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued. The existence of the warning slogan evaluation module enables the system to realize functions that are not only limited to identifying the presence of warning slogans in a device and its environment, but also further identifying the content of the warning slogan, and then comparing whether the content of the warning slogan is correct. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0046] Figure 1 It is the system principle diagram of the present invention;

[0047] Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0050] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0051] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0052] Combine the following Figure 1-Figure 2 , the implementation modes of the present invention are further described.

[0053] In one embodiment of the present invention, a visual process management system based on a one-time device startup operation is provided, comprising:

[0054] The hardware inspection unit collects a picture of the appearance of a primary device, identifies whether the appearance of the primary device is obviously damaged through a pre-trained hardware damage inspection model, and decides whether to issue an alarm signal based on the identification result; the training process of the hardware damage inspection model is as follows:

[0055] (1) Collect pictures of electrical equipment, including pictures of intact electrical equipment and pictures of damaged electrical equipment;

[0056] (2) Adjust all images to a uniform size, convert them into grayscale images, and normalize the image pixel values;

[0057] (3) Select the deep learning model framework and design the number of model layers and the parameters of each layer;

[0058] (4) Divide the data into training set, validation set and test set, use the training set data to train the model, use the validation set data to adjust the model parameters, and evaluate the performance of the model on the test set;

[0059] (5) Adjust the learning rate according to the model performance and save the trained model after training is completed.

[0060] The hardware inspection unit is also provided with a pre-trained warning slogan inspection model, and also collects a picture of the surrounding environment of the primary device, identifies whether there is a warning slogan in the installation environment of the primary device through the pre-trained warning slogan inspection model, and decides whether to issue an alarm signal based on the identification result. The training process of the warning slogan inspection model is as follows:

[0061] (1) Collect pictures that include warning slogans and pictures that do not contain warning slogans as negative samples;

[0062] (2) Use image annotation tools to annotate the collected images;

[0063] (3) Adjust the image size to meet the model input requirements and perform image enhancement operations including rotation, scaling, and color change;

[0064] (4) Select a deep learning model and use the annotated dataset to train the selected model and adjust the model parameters and hyperparameters;

[0065] (5) Evaluate model performance on the validation set, use the confusion matrix to analyze the model's performance on different categories, and optimize the model based on the evaluation results, which may include adjusting the network structure, using data augmentation, and changing the training strategy.

[0066] The hardware inspection unit is also provided with a warning slogan judging module, which performs the following operations:

[0067] (1) Collect a picture of the device appearance and a picture of the device surrounding environment;

[0068] (2) Preprocess the image, including resizing, denoising, and binarization, to improve recognition accuracy;

[0069] (3) Select and use OCR tools to perform text recognition on the image and extract the text in the image;

[0070] (4) Compare the text recognized by OCR with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued.

[0071] An index checking unit collects electrical parameters and mechanical parameters in a control loop, wherein the electrical parameters include current and voltage, and the mechanical parameters include vibration and temperature, and compares and analyzes the parameters with standard parameters or historical data. If the degree of deviation exceeds a threshold, an alarm signal is issued;

[0072] An equipment debugging module is mounted on the inspector terminal, and an oscilloscope and a power analyzer are provided on the equipment debugging module. The oscilloscope is used to detect the signal waveform, and the power analyzer is used to test the power quality. The signal waveform and the power quality are uploaded through the inspector terminal for sharing with other inspector terminals and manager terminals. Moreover, when the operator determines that there is a problem with the signal waveform or power quality, an alarm signal can be directly sent to the manager terminal through the interface between the equipment debugging module and the inspector terminal.

[0073] A recording and storage unit, connected to the hardware inspection unit and the index inspection unit, for collecting and storing the original data and output results of the operation process of the hardware inspection unit and the index inspection unit;

[0074] The terminal is equipped with a visualization unit, which is connected to the recording and storage unit, the hardware inspection unit and the indicator inspection unit. The visualization unit includes a display module and an alarm module, wherein the display module is used to display the storage data of the recording and storage unit, and the alarm module is used to receive the alarm signal and perform the alarm operation. The terminal includes a plurality of executor terminals and a plurality of manager terminals, wherein the executor terminals and the manager terminals are provided with different data access rights, and the executor terminals can enter data into the recording and storage unit for the manager terminal or other executor terminals to read.

[0075] In one embodiment of the present invention, a visual process management method based on a device startup operation is provided, comprising the following steps:

[0076] S1: Collect pictures of intact and damaged electrical equipment, and train an electrical equipment recognition model based on the deep learning model framework to identify whether the electrical equipment is damaged;

[0077] S2: Collect pictures containing warning slogans and pictures without warning slogans, and train a warning slogan detection model based on the deep learning model to identify whether the pictures contain warning slogans;

[0078] S3: Collect the appearance pictures of the primary equipment and the surrounding environment pictures, and identify whether there is any damage to the electrical equipment and whether there is no warning sign based on the electrical equipment recognition model and the warning sign inspection model;

[0079] S4: Send an alarm signal to the terminal when the electrical equipment is damaged or there is no warning sign;

[0080] S5: Select and use the OCR tool to perform text recognition on the image, extract the text in the image, and compare the text recognized by the OCR with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued.

[0081] The electrical equipment identification model described in step S1 is obtained by the following steps:

[0082] (1) Collect a large number of pictures of electrical equipment, including pictures of intact and damaged equipment.

[0083] (2) Annotate the collected images and distinguish between intact and damaged devices, which usually requires professional knowledge.

[0084] (3) Resize all images to a uniform size. If necessary, convert the images to a color space that is more suitable for processing, such as grayscale. The image pixel values ​​can also be normalized.

[0085] (4) Build an image recognition model: Select a deep learning model architecture suitable for image recognition, such as a convolutional neural network (CNN). Design the number of layers in the model and the parameters of each layer.

[0086] (5) Divide the data into training set, validation set and test set; use the training set data to train the model and use the validation set data to adjust the model parameters; evaluate the performance of the model on the test set.

[0087] (6) Adjust hyperparameters such as learning rate and batch size according to model performance; apply dropout or L2 regularization to prevent overfitting.

[0088] (7) Save the trained model and load the model in actual application.

[0089] In step S2, image recognition technology is used to automatically identify whether the photo contains warning signs and fences, which can be achieved through the following detailed steps:

[0090] (1) Data collection: Collect images containing warning signs and fences, as well as images without these elements as negative samples.

[0091] (2) Data labeling: Use image labeling tools (such as LabelImg, MakeSense, etc.) to label the collected images. Label the locations of warning signs and fences, usually using rectangular boxes, and classify them (e.g., warning signs, fences).

[0092] (3) Data preprocessing: Adjust the image size to meet the model input requirements. Perform image enhancement such as rotation, scaling, and color change to increase the generalization ability of the model.

[0093] (4) Select an image recognition model: Select a deep learning model suitable for target detection, such as Faster R-CNN, YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), etc.

[0094] (5) Model training: Use the labeled dataset to train the selected model and adjust the model parameters and hyperparameters, such as learning rate, batch size, number of iterations, etc.

[0095] (6) Model evaluation: Evaluate model performance on the validation set, focusing on indicators such as accuracy and recall. Use the confusion matrix to analyze the performance of the model on different categories.

[0096] (7) Model optimization: Optimize the model based on the evaluation results, which may include adjusting the network structure, using data augmentation, changing the training strategy, etc.

[0097] (8) Model testing: Test the model on an independent test set to ensure that the model has good generalization ability.

[0098] (9) Model deployment: Save and deploy the trained model to a server or cloud platform, and then develop an API interface so that the model can receive input images and return recognition results.

[0099] In step S5, automatically identifying whether the content of the warning slogan in the picture is correct can be achieved by the following steps:

[0100] (1) Data collection and annotation: Collect pictures containing warning slogans. These pictures should cover different warning slogan styles and backgrounds. Annotate the pictures and determine the correct content of the slogan text. This usually requires the use of image annotation tools.

[0101] (2) Image preprocessing: Perform preprocessing operations such as denoising, binarization, and automatic tilt correction on the image to improve the accuracy of subsequent recognition.

[0102] (3) Choose OCR technology: Choose the appropriate OCR technology, which can be a general OCR service, such as Alibaba Cloud OCR, Tencent Cloud Text Recognition OCR, etc. These services are usually based on deep learning technology and can handle multiple languages ​​and fonts. In addition, if the existing OCR service cannot meet the needs, it may be necessary to train a custom OCR model, which usually involves using a deep learning framework such as TensorFlow or PyTorch and preparing a labeled dataset to train the model.

[0103] (4) Call the OCR service: Call the OCR service through the API or SDK, upload the image and obtain the recognition result. Make sure that the selected OCR service supports the required language and character set.

[0104] (5) Result analysis and verification: Analyze the OCR recognition results and compare them with the correct annotated content to determine whether the slogan content is correct. In this step, a threshold can be set to determine when the recognition result is considered acceptable, for example, when the recognition accuracy exceeds a certain percentage.

[0105] In summary, the present invention is provided with a hardware inspection unit, an index inspection unit, a recording and storage unit and a terminal, and the terminal is equipped with a visualization unit. The hardware inspection unit can be used to check the degree of perfection of the hardware settings during the startup of a device, including whether the device is damaged, whether the appearance of the device and its installation environment are set with warning signs as required, etc. The index inspection unit can be used to check the deviation values ​​of various electrical parameters or mechanical parameters from the standard parameters during or after the startup of the device, and then, based on the output results of the hardware inspection unit and the index inspection unit, it is decided whether to send an alarm signal to the terminal, thereby facilitating the operator to perform corresponding processing. In addition, the recording and storage unit will record various process data and result data of the hardware inspection unit and the index inspection unit, so as to facilitate the terminal to perform corresponding processing based on the data during maintenance and other operations.

[0106] Among them, the hardware inspection unit is provided with a pre-trained hardware damage inspection model and a warning slogan inspection model. The hardware damage inspection model can be used to check whether the primary equipment is damaged, and the warning slogan inspection model can be used to check whether the primary equipment and its surrounding environment are equipped with warning slogans. By deploying a camera at the site of the primary equipment and taking pictures of the primary equipment and its surrounding environment, the model can be directly used to identify whether there is an abnormality, without the need for manual identification, saving time and energy. In addition, the hardware inspection unit is also provided with a warning slogan evaluation module. By collecting pictures of the appearance of the primary equipment and the surrounding environment of the primary equipment, the pictures are subjected to text recognition based on the OCR tool and the text in the pictures is extracted. The text recognized by the OCR is compared with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued. The existence of the warning slogan evaluation module enables the system to realize functions that are not only limited to identifying the presence of warning slogans in the primary equipment and its environment, but also further identifies the content of the warning slogan, and then compares whether the content of the warning slogan is correct.

[0107] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0108] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0109] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0110] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A visual process management system based on a one-time equipment startup operation, characterized in that: include: The hardware inspection unit collects a picture of the appearance of a device, identifies whether the appearance of the device is obviously damaged through a pre-trained hardware damage inspection model, and decides whether to issue an alarm signal based on the identification result; An index checking unit collects electrical parameters and mechanical parameters in a control loop, wherein the electrical parameters include current and voltage, and the mechanical parameters include vibration and temperature, and compares and analyzes the parameters with standard parameters or historical data. If the degree of deviation exceeds a threshold, an alarm signal is issued; A recording and storage unit, connected to the hardware inspection unit and the index inspection unit, for collecting and storing the original data and output results of the operation process of the hardware inspection unit and the index inspection unit; The terminal is equipped with a visualization unit, which is connected to the recording and storage unit, the hardware inspection unit and the indicator inspection unit. The visualization unit includes a display module and an alarm module, wherein the display module is used to display the storage data of the recording and storage unit, and the alarm module is used to receive an alarm signal and perform an alarm operation.

2. A visual process management system based on a one-time equipment startup operation according to claim 1, characterized in that: The terminals include several executor terminals and several manager terminals. The executor terminals and the manager terminals are provided with different data access rights, and the executor terminals can input data into the recording and storage unit for reading by the manager terminal or other executor terminals.

3. A visual process management system based on a one-time equipment startup operation according to claim 1, characterized in that: The training process of the hardware damage detection model is as follows: Collect pictures of electrical equipment, including pictures of intact electrical equipment and pictures of damaged electrical equipment; Resize all images to a uniform size, convert them into grayscale images, and normalize the image pixel values; Select the deep learning model framework, design the number of model layers and the parameters of each layer; Divide the data into training set, validation set and test set, use the training set data to train the model, use the validation set data to adjust the model parameters, and evaluate the performance of the model on the test set; Adjust the learning rate according to the model performance and save the trained model after training is completed.

4. A visual process management system based on a one-time equipment startup operation according to claim 1, characterized in that: The hardware inspection unit is also provided with a pre-trained warning slogan inspection model, and also collects a picture of the surrounding environment of the primary device, identifies whether there is a warning slogan in the installation environment of the primary device through the pre-trained warning slogan inspection model, and decides whether to issue an alarm signal based on the identification result. The training process of the warning slogan inspection model is as follows: Collect pictures that include warning slogans and pictures that do not contain warning slogans as negative samples; Use image annotation tools to annotate the collected images; Resize the image to meet the model input requirements and perform image enhancement operations including rotation, scaling, and color changes; Select a deep learning model and use the labeled dataset to train the selected model and adjust the model parameters and hyperparameters; Evaluate model performance on the validation set, use confusion matrix to analyze the model's performance on different categories, and optimize the model based on the evaluation results, which may include adjusting the network structure, using data enhancement, and changing the training strategy.

5. A visual process management system based on a one-time equipment startup operation according to claim 4, characterized in that: The hardware inspection unit is also provided with a warning slogan judging module, which performs the following operations: Collect a picture of the device appearance and a picture of the device surrounding environment; Preprocess the images, including resizing, denoising, and binarization, to improve recognition accuracy; Select and use the OCR tool to perform text recognition on the image and extract the text in the image; The text recognized by OCR is compared with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is issued.

6. A visual process management system based on a one-time equipment startup operation according to claim 2, characterized in that: It also includes an equipment debugging module, which is mounted on the inspector terminal and connected to the recording and storage unit. The equipment debugging module is provided with an oscilloscope and a power analyzer. The oscilloscope is used to detect the signal waveform, and the power analyzer is used to test the power quality. The signal waveform and the power quality status are uploaded through the inspector terminal for sharing with other inspector terminals and manager terminals. Moreover, when the operating personnel determine that there is a problem with the signal waveform or power quality, an alarm signal can be directly sent to the manager terminal through the interface between the equipment debugging module and the inspector terminal.

7. A visual process management method based on a one-time equipment startup operation, characterized in that: The following steps are involved: Collect pictures of intact and damaged electrical equipment, and train an electrical equipment recognition model based on the deep learning model framework to identify whether the electrical equipment is damaged; Collect pictures containing warning slogans and pictures without warning slogans, and train a warning slogan detection model based on the deep learning model to identify whether the pictures contain warning slogans; Collect pictures of the appearance of the primary equipment and the surrounding environment, and identify whether there is any damage to the electrical equipment and whether there is no warning sign based on the electrical equipment recognition model and the warning sign inspection model; When electrical equipment is damaged or has no warning signs, an alarm signal is sent to the terminal.

8. A visual process management method based on a one-time equipment startup operation according to claim 7, characterized in that: The following steps are also included: Collect a picture of the device appearance and a picture of the device surrounding environment; Preprocess the images, including resizing, denoising, and binarization, to improve recognition accuracy; Select and use the OCR tool to perform text recognition on the image and extract the text in the image; The text recognized by OCR is compared with the predefined correct warning slogan. If the recognized text is inconsistent with the correct slogan, it is marked as an error and an alarm signal is sent.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the visual process management method based on a one-time device startup operation as described in any one of claims 7 to 8.

10. A processor, characterized in that: The processor is used to run a program, wherein when the program is run, the visual process management method based on a one-time device startup operation as described in any one of claims 7 to 8 is executed.