CN117808450A - Visual guidance system and visual guidance method for field operation - Google Patents

Abstract

The invention discloses a visual guidance system and a visual guidance method for field operation, which belong to the field of maintenance of electric power systems, in particular to a visual guidance system and a visual guidance method for field operation, comprising the following steps: the mobile terminal can upload and receive information such as instrument images, instrument descriptions and the like, the server receives service requests of the mobile terminal, recognizes the received instrument images, searches the intelligent city diagrams of the received instrument images, retrieves the instrument images and the instrument descriptions from the database and sends the instrument images and the instrument descriptions to the mobile terminal, and receives images and descriptions after the last field operation for the same instrument every time of field operation, and the server is used for storing all instrument information. Through communication connection among the mobile terminal, the server and the database, the operation standard and the previous operation image for the instrument can be issued to the field personnel according to the service requirement, and effective guidance is provided for the operators.

Description

Visual guidance system and visual guidance method for field operation

Technical Field

The invention relates to the field of maintenance of power systems, in particular to a visual guidance system and method for on-site operation.

Background

The instrument is an important component of the power system, not only relates to various electric instruments, but also comprises a large number of oil gas instruments, the working principle, the connecting method and the daily operation and maintenance method of the instruments are completely different, and the maintenance priority and the frequency of the instruments are different from each other, and if the instruments are only inspected by guiding professionals according to the regulation and the records of paper, the instruments are easy to cause faults, and even risk hidden dangers existing after the technical personnel are inspected cannot be recognized in time. In the current guidance of field operation, some schemes have been developed, and in the whole, one is to rely on manual overhaul and acceptance mode to deal with, and for any field operation, a plurality of groups are configured, wherein no less than two persons in each group overhaul one person, and the other person accepts acceptance, and only two persons confirm that the field operation result is free of problems, the task can be considered to be completed, and the paper surface is recorded. The other is to use an information system to perform work order distribution and field operation result collection, and then to perform image detection to determine whether a problem exists in the field operation.

In the field operation of the current oil gas instrument, the main problems include: 1. the job problem is difficult to identify. The cause of the problem is that the oil gas meters are numerous and are coupled with other types of meters or equipment, especially power equipment, for example, some oil gas meters comprise various sensors, controllers and meters, other electronic devices with communication or power connection relations are arranged around the oil gas meters, and the connection relations among the equipment are complex, so that the existing field operation problems are difficult to identify. 2. Management complexity increases abruptly. The reason for this problem is that, based on the scheme of paper surface data recording, even if it is recorded completely in accordance with specifications after field operation, the field operator lacks image-based guidance, and the joint processing cannot be completed for some joint areas, so that the joint areas are gradually confused, and the management complexity is increased suddenly. 3. The skill training difficulty is high. In the traditional field operation, only the traditional teacher and apprentice system can be adopted to train the personnel, and the training mode only depends on the experience of the personnel with abundant experience, so that the training effect is general. 4. The data management is heavy. On-site operations in power systems can generate large amounts of data, which can be difficult to update immediately into a database if conventional paper records are used, as well as some simple worksheet dispatch systems, requiring significant effort to sort the data.

Disclosure of Invention

The invention discloses a visual guidance system and a visual guidance method for on-site operation, which are used for solving the problems of difficulty in identifying and managing work complexity suddenly and great skill training difficulty in primary data management work heavy in the prior art, and specifically comprise the following steps:

a visual guidance system for field operations, comprising:

the mobile terminal, the server and the database;

the mobile terminal is used for shooting and uploading instrument images and also used for receiving the instrument information issued by the server;

the server is used for receiving the instrument image uploaded by the mobile terminal and comparing the instrument image with the instrument information stored in the database;

the database is used for storing the instrument information;

the mobile terminal is in communication connection with the server, and the server is in communication connection with the database.

Optionally, the mobile terminal comprises an image display window, a photographing button, an image uploading button, a form display page, a work order display window, a search frame and a scanning frame;

the image display window is used for displaying images shot by the mobile terminal and received instrument images;

the photographing button is used for calling a mobile terminal camera, and when the photographing button is clicked, an image display window or a scanning frame is called;

the image uploading button is used for uploading the instrument image to the server;

the table display page displays a view layer table, lines and columns are divided through GridLayout, text view display data are filled, and ScrollView rolling data are used;

the work order display window is used for displaying field work order information issued to the mobile terminal by the server;

the scanning frame is used for scanning the number nameplate or the two-dimensional code information of the instrument;

the search box is used for inputting keywords.

Optionally, the search box stores historical search keywords, and scrolls and displays the historical search keywords.

Optionally, the server sets an intelligent recognition algorithm, including:

acquiring the instrument image data set and classifying and grouping the image data;

setting a color threshold value, a saturation selection threshold value, a tone threshold value and a contrast brightness threshold value for the new instrument image, selecting at least 2 sets of threshold values, and executing enhancement processing to obtain a new instrument image with enhanced data;

and constructing an instrument YOLOV5 model based on the new instrument image.

Optionally, after the server acquires the field operation instrument image uploaded by the mobile terminal, the field operation instrument image is enhanced, and the enhanced field operation instrument image is compared with the YOLOV5 model.

Optionally, the intelligent recognition algorithm further includes an improved recognition algorithm, including:

based on the new instrument image, combining a neural network to obtain a ncnn model;

adjusting the integral space domain weight, the positive side space domain weight and the step-by-step domain threshold value screening weight, and performing marginal reduction optimization;

and carrying out formation pooling concurrent classification deduction on the plurality of ncnn models, predicting the sufficiency degree of the candidate YOLOV5 models, and selecting the YOLOV5 model with the highest sufficiency degree.

Optionally, the database is an embedded database sqlite.

Optionally, a DAO interface is also included to perform data manipulation tasks.

A visual guidance method for field operation comprises the following steps:

step one, importing the instrument images into the database, classifying and grouping the instrument images by the server, and carrying out enhancement processing on the instrument images to obtain new instrument images;

training the new instrument image by using a tuning device to obtain the YOLOV5 model;

step three, after receiving the field operation instrument image uploaded by the mobile terminal, the server identifies the instrument based on the YOLOV5 model;

and step four, after the instrument is identified, the server sends the instrument information to the mobile terminal and synchronously sends the field operation specification of the instrument.

Optionally, after the meter is identified, the server sends the meter information to the mobile terminal, and synchronously sends a field operation specification of the meter, including:

when the server obtains the field operation instrument image and stores the field operation instrument image into the database, a time stamp is set;

the instrument image sent to the mobile terminal by the server is the field operation image corresponding to the timestamp closest to the field operation time.

The invention has the beneficial effects that:

1. and acquiring the guiding data in real time. When the on-site operation is started, based on the number of the oil gas instrument or other indication beacons, the mobile terminal directly acquires specific data of the corresponding oil gas instrument from the database, including types, heavy operation difficulties, operation requirements and the like, and simultaneously provides prefabricated images for the oil gas instrument after the on-site operation is finished so as to guide operators. In addition, for some oil and gas meters with significantly higher complexity and importance, the field operation is performed in steps, and the invention can provide guiding data for each step.

2. The real-time monitoring of the operation quality is realized. In the traditional oil gas instrument field operation, the quality monitoring of the field operation is delayed due to the difficulty in data transmission and the monitoring quality of the operation quality, and the defects existing in the field operation are easy to exist for a long time in the field operation of the oil gas instrument, so that the operation hidden trouble of the whole power system is formed. The invention uses the mobile terminal to support shooting and uploading in each stage of flow, and the database automatically analyzes whether the current field operation has operation problems or not and feeds back the operation problems in real time, thereby realizing real-time monitoring of the field operation quality.

3. And the operation and training efficiency is improved. In the field operation, technicians with insufficient working experience acquire working tasks required to be completed, acquire initial data of an oil-gas instrument and data after the completion of the work on the one hand, and are required to shoot operation photos according to the working of the department in each stage, send the operation photos to a server, evaluate the operation quality by the server, and on the other hand, check and accept the field operation by a system, other personnel in the same operation group are responsible for field safety protection without secondary check by other personnel, so that the field operation efficiency is fully improved.

4. The data management is simple and efficient. In the technical scheme of the invention, all generated data can be received by the server, including the field operation time of the oil gas instrument, whether the oil gas instrument is replaced or not, the parameters of the new instrument after the replacement, the field operation times, the new parameters of the oil gas instrument after the field operation and the like, are updated in the server, the data issued to the field operation personnel are the data newly generated after the last field operation, and meanwhile, a database specially used for storing historical data is also arranged for the call of a background manager or the field operation personnel, the aspects of data management are more ordered and standardized, all the data are combined with the corresponding images, the problem of separation of the images and the data is avoided, and the simplicity and the high efficiency of the data management are realized.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.

FIG. 1 is a schematic diagram of a visual guidance system for field operation according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of the operation of a visual guidance system for field operations according to an embodiment of the present invention;

FIG. 3 is a flow chart of feature value extraction of a visual guidance system for field operation according to an embodiment of the present invention;

fig. 4 is a flow chart of operation of a visual guidance system for field operation according to an embodiment of the present invention.

Detailed Description

The invention is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the structures related to the present invention are shown in the drawings.

Example 1

The embodiment of the invention discloses a field operation visual guidance system schematic diagram, and shows the connection mode and the operation principle of all subsystems, so as to explain the technical scheme about the system part in the invention.

Referring to fig. 1, a schematic diagram of a visual guidance system for field operation is provided for explaining all subsystems and connection logic relations in the whole system. As shown in fig. 1, includes: mobile terminals, servers, databases, oil and gas meter systems, beacon facilities, etc.

The mobile terminal is a mobile communication device commonly used at present, and the mobile terminal system is mainly an android system.

Engineering software is installed on the mobile terminal, the software comprises a search frame, a shooting button, a scanning frame, an image display frame, an image sending button, a form page and the like, and the search frame is used for searching information such as the number and the position of an oil and gas instrument and supports accurate search and fuzzy search; the shooting button can call a camera at the mobile end to shoot an oil gas instrument code, an oil gas instrument itself and the like; the scanning frame can call the camera at the mobile end to scan codes or information two-dimensional codes arranged on the oil-gas instrument; the image display frame is used for displaying the real-time image acquired by the camera, completely displaying the real-time image in the image display frame after photographing, displaying the oil gas instrument image acquired based on the search frame, and updating the image after the last field operation; the image transmitting button is used for transmitting the shot image; the form page is a jump page, and when a plurality of oil gas meters with the same type are obtained by adopting a search frame, the information such as the number, the position, whether field operation is needed, the last field operation time and the like of the oil gas meters are displayed in a form mode;

in some embodiments, the engineering software may also call the NFC system in the mobile terminal, and display specific information of the oil gas meter in combination with the beacon on the oil gas meter by touching the mobile terminal device and the beacon.

The server is used for calling the information of the oil gas instrument, receiving a data acquisition request submitted by the engineering software of the mobile terminal, and comparing and updating the data.

The server is also used for identifying the information such as the model, the position, the number and the like of the oil gas instrument.

The server is also used for extracting the characteristics of the oil gas instrument images in the database, extracting the numerical values, and then obtaining the results of the extracted characteristic data.

In the invention, all the oil gas instrument joints and instruments can be managed, referring to fig. 2, a field operation visual guidance system operation schematic diagram is disclosed for the embodiment of the invention, as shown in fig. 2, the whole set of system automatically establishes the grouping by the system through the work of grouping, body and joint identification and management of the oil gas instruments and the like, based on an original image, and after the field operation is performed, sub-grouping can be also established, and simultaneously, the field operation personnel can be supported to use a mobile terminal for searching, the shooting searching and keyword searching are supported, and the searching result is fed back.

The server performs feature value extraction based on the neural network, see fig. 3, and is a flow chart for extracting feature values of a visual guidance system for field operation disclosed in the embodiment of the invention, as shown in fig. 3, specifically:

step one: and extracting key features of the sample image.

And acquiring images of various oil gas meters from a database, performing overturn and inclination, proportional molecular scaling and color gamut transformation on the images until characteristics of joints, types, models and the like of the oil gas meters in the images are extracted, determining based on boundary conditions and character conditions of various components in the oil gas meters in the extraction process, and obtaining a result based on an image processing algorithm. For example, a certain oil-gas instrument is provided with two joints, the diameter of a pipeline of an inlet joint is phi 10, the diameter of a pipeline of an outlet joint is phi 8, and the numerical areas indicated by dial pointers are green, white and red splices and contain unit numerical values which are all used as key characteristics. It should be noted that, as for the key features, it belongs to the features of the oil and gas meter in the normal state.

Step two: the image is classified, the model of the oil gas instrument is determined based on the key characteristics of the sample graph, and the oil gas instruments of the same model are included in the same classification.

The images in the database can have various oil-gas meters, and simultaneously can have various expression forms for the same oil-gas meter, the former is easy to understand, the latter comprises an explosion diagram of the oil-gas meter, a feedback diagram of field operation, a typical damage diagram and an atypical damage diagram of the oil-gas meter and the like, each image acquires key features based on the mode of the step one, performs key feature comparison, and is included in the same image set for the same key features;

in some embodiments, the image set for the same class may be further subdivided, such as classified as an explosion map image set, a typical damage image set, an atypical damage image set, etc., for the same class of hydrocarbon meters.

Step three: and comparing the field diagram, shooting an oil-gas instrument image of the operation field by the mobile terminal, uploading the image to a server, and comparing the field image with key characteristics of the oil-gas instrument by the server.

The field diagram comparison can be used for searching information of the oil-gas instrument in the operation site, and after the field diagram is acquired, the image is turned up and down, scaled, cut and the like left and right, and compared with the obtained key characteristics of the oil-gas instrument, and a feedback result is given. If the angle of the connecting flange at two sides and the deviation between the key features are too large, and the distance between the connecting flange and the oil gas instrument body and the deviation between the key features are too large, warning information is sent out, so that an operator can further check the information, and if the information displayed in the key features cannot be followed due to special conditions, the operator at the upper stage uses another device to photograph, upload and confirm.

Step four: and updating data, and setting an image tag for the field operation image by the server.

After receiving the field operation image, the server automatically generates an image tag, including data such as field operation time, place, oil and gas instrument number and the like.

Step five: and storing and calling the data, and storing the live operation image with the image tag as a database.

And (3) based on the data obtained in the third step and the fourth step, sending the field map and the image tag data to a database and storing the field map and the image tag data. When the staff needs to call the information of the oil gas instrument in the subsequent field operation after the field operation of the oil gas instrument, the called oil gas instrument image and data are updated field images and data after the last field operation.

And when the data is stored and called, the normal operation image and data of the oil gas instrument are also supported for the on-site operation staff to call.

The server is in communication connection with the engineering software of the mobile terminal, and the server is also in communication connection with the database.

The server also supports responding to keyword search requests in the engineering software of the mobile terminal, and specifically:

and identifying the number of the mobile terminal equipment and the signature of the field operator, after receiving the keywords input in the search box, searching all images and data corresponding to the keywords from the database, feeding back to the mobile terminal engineering software, and displaying in a graph mode. In addition, the historical search keywords are displayed in a scrolling mode in the search frame, and 16 keywords are supported at maximum.

Wherein, for the identification of the image, enhancement algorithm processing is adopted, and the specific steps are as follows:

according to schemes such as threshold adjustment, color threshold adjustment, contrast brightness threshold adjustment and the like of the image, two sets of thresholds and two sets of random enhancement strategies are adopted for processing, contrast enhancement is carried out on data of the image, the enhanced image is input into an improved YOLOV5 model for training, an optimal solution model is obtained through an adjuster, and the improved YOLOV5 model is obtained.

Wherein, the improved YOLOV5 model comprises:

performing random left-right, lower-upper-turn tilting, proportional molecular scaling and color gamut transformation on the model, and performing stitching line image feature extraction after cutting according to a preset template;

and carrying out data feature specification extraction according to 2 sets of threshold values, and randomly comparing feature data in advance according to random enhancement.

Performing inversion residual error on the feature extraction part to extract image features;

designing multiple scales for feature fusion, and generating a plurality of detection heads with different degrees of visibility; and optimizing a regression loss function of the prediction frame.

And simultaneously, performing image recognition by utilizing an improved neural network forward calculation ncnn model, wherein the improvement method comprises the following steps:

the method comprises the steps of threshold value digitizing the spatial domain weight, positive side converting the spatial domain weight, step-by-step domain threshold value screening the weight, performing marginal reduction optimization, performing layering pooling concurrent classification deduction on a plurality of ncnn models, predicting candidate sufficiency, and multiplying the candidate model characteristics by a factor.

The database is built based on a light embedded database sqlite to store all data information, the database occupies low resources, the data operation adopts a DAO (DataAccess Object ) design mode, the service logic of the data access and app program is separated, and the maintainability and the flexibility of a data layer are improved;

in some embodiments, the databases are functionally divided into two categories, one being a fixed database and the other being an adjustable database, wherein:

the images and data stored in the fixed database can not be modified, the stored images comprise the model numbers of various oil and gas meters, new oil and gas meter images, images after the new oil and gas meters are installed, oil and gas meter explosion diagrams, oil and gas meter joint processing standards and the like, and the information is used as the characteristic of the oil and gas meters, namely the standard characteristic.

The fixed database is in communication connection with the server, and only when the server sends out a standard feature calling instruction, standard feature data is sent, and a data modification instruction sent by the server is not received.

The adjustable database is a database with adjustable information such as images and data and is used for receiving the information such as the images and the data after the oil gas instrument performs on-site operation, and specifically:

after field operation is performed on a certain oil and gas meter, the server receives new meter image, data and other information and transmits the information into the adjustable database.

After receiving an information calling instruction sent by a server and aiming at a certain oil gas instrument, the adjustable database defaults to send oil gas instrument images and data after the last field operation and also supports the sending of historical data.

The database needs to define DAO interface standard operation functions, such as powerstationIF, define DAO implementation class PowerstationSqliteDao, leave voiles for expanding different databases later, define entity class powerstationentity, and store object data attributes.

The oil gas instrument system is all oil gas instrument assemblies in the current power system, and can acquire data for any one of the oil gas instruments.

The beacon facility is the position and number evidence of the oil gas instrument, and after the mobile end engineering software obtains the position and number evidence, the information of the oil gas instrument can be accurately described;

in some embodiments, the beacon facility is an RFID chip, and the information of the oil gas instrument is directly displayed after the supporting mobile terminal is attached;

in some embodiments, the beacon facility is a coded nameplate or two-dimensional code, and supports scanning or photographing of the mobile terminal equipment to acquire oil and gas instrument information.

The oil gas instrument system is in communication connection or physical connection with the beacon facilities, and in the physical connection, the beacon facilities are directly and fixedly arranged on or around the oil gas instrument.

The oil gas instrument system and the beacon facility are in communication connection with the mobile terminal.

Example two

The embodiment is used for explaining the operation method of the whole set of field operation visual guidance system, and enables the technical implementation mode to be better understood by the person skilled in the art according to each working stage in the field operation.

Referring to fig. 4, a flow chart of operation of a visual guidance system for field operation disclosed in an embodiment of the present invention is shown in fig. 4, and a flow chart of a visual guidance method for field operation provided in an embodiment of the present invention is shown, wherein all implementation matters involved in the operation of the whole system are described. Specifically, the method comprises the following steps:

step one, information preparation work, normal oil gas instrument images are injected into a database, and feature values are obtained.

The aim of the work is to acquire the standard characteristics of various oil gas instrument images based on machine learning by inputting a large number of oil gas instrument images into a database of a visual guidance system for field operation.

The standard features are a joint mode, an information display mode, a fault expression mode and the like of the oil gas instrument in normal operation.

The input oil-gas meters are two kinds, one is a normal oil-gas meter, namely a new meter or a meter which is used for a period of time and operates normally, and the other is an abnormal oil-gas meter, namely an oil-gas meter with fault expression characteristics.

For a normal oil gas meter, the input image comprises a joint, a flange of the oil gas meter, a meter flange connection mode, an oil gas meter flange connection angle, an oil gas meter flange and meter body distance, oil gas meter display characteristics and the like, all meter special diagnosis is extracted by an algorithm in a server, a data label is generated, and the data label and the normal oil gas meter image are combined to generate normal oil gas meter data.

For an abnormal oil gas meter, an input image comprises a joint, a flange of the oil gas meter, a meter flange connection mode, an oil gas meter flange connection angle, an oil gas meter flange and meter body distance, oil gas meter display characteristics and the like, all meter characteristic special diagnosis is extracted by an algorithm in a server, a data tag is generated, and the data tag is combined with a normal oil gas meter image to generate abnormal oil gas meter data.

And the generated normal oil gas instrument data and the generated abnormal oil gas instrument data are stored in a grouping way for the normal oil gas instrument and the abnormal oil gas instrument to form two data groups.

Step two, calling the data before the field operation to work, injecting a normal oil gas instrument image into the database, and obtaining the characteristic value.

The purpose of this work is, when sending work order to on-the-spot operation personnel, the oil gas instrument data such as the type of direct display oil gas instrument, position, on-the-spot operation detail let on-the-spot operation personnel have certain understanding to oil gas instrument itself.

The data call instruction is derived from two aspects, namely a work order instruction and a field operator instruction, and the specific steps are as follows:

the server directly generates a work order based on the field operation requirement of the oil-gas instrument, the work order is provided with model information of a field operation object, the model information of the operation object is sent to a database by the server in a data calling instruction mode, and two data groups in the database receive the calling instruction. The normal oil gas instrument data group feeds back the normal oil gas instrument data of the oil gas instrument of the model according to the instrument model information; and the abnormal oil gas instrument data group feeds back abnormal oil gas instrument data of the oil gas instrument of the model according to instrument model information.

The oil gas instrument data fed back by the two data sets are temporarily stored in the server in a data caching or linking mode.

The server generates a worker and transmits the worker to mobile terminals of all on-site operators.

The on-site operator opens the engineering software interface by clicking pushing information of the mobile terminal engineering software based on the work order information to rob the work order, and clicks the confirmation button to consider the work order.

The server sets the first field operator participating in the order taking as the operator of the work order according to the order taking time, and the group where the field operator is located is the execution group.

After confirming the execution group, the server sends the temporary stored cache data or links in the server to the mobile terminal equipment of the field operator in the execution group based on the personnel information and the signature in the execution group, and the field operator instruction is obtained. And on the basis of obtaining the two instructions simultaneously, the server considers that the dispatch is completed.

In some embodiments, particularly when the number of persons in the field work group is not less than 3, it is determined whether to dispatch a order to the group of persons based on the number of persons, following "minority-compliance-majority" judgment logic, namely: after the first on-site operation personnel participating in the robbery order is determined, judging whether the personnel participating in the robbery order in the execution group are more than the personnel not participating in the robbery order, if so, confirming the personnel group as the execution group, and if not, checking the conditions of the personnel group of other on-site operation personnel participating in the robbery order, until the execution group is judged.

After the execution group is determined, normal oil gas instrument data and abnormal oil gas instrument data of the oil gas instrument are sent to mobile terminal equipment of field operators in the execution group, indexes are established according to a chart mode, and picture consulting links are arranged in the charts, so that the field operators can acquire corresponding images and detailed descriptions after clicking connection.

And thirdly, performing field operation, injecting normal oil gas instrument images into the database, and obtaining characteristic values.

After the site operation personnel reach the position of the oil-gas instrument in the work order, shooting an image of the oil-gas instrument by using the mobile terminal, wherein the image comprises a joint, a component and an oil-gas instrument body, extracting information of the site operation image by a server, and acquiring characteristic data of the site operation image by adopting an intelligent recognition algorithm.

And comparing the characteristic data of the field operation image with the normal oil gas instrument data, and if the deviation of the characteristic data of the field operation image and the normal oil gas instrument data is confirmed to be in the allowable range, considering that the oil gas instrument of the operation field operates normally, only reading the instrument display value by field operators, uploading the instrument display value, and performing simple oil gas instrument appearance inspection and cleaning.

In the comparison between the characteristic data of the field operation image and the normal meter data, all data types need to be traversed, for example: the normal oil gas meter data of a certain oil gas meter comprises joint pipeline diameter, flange type, flange diameter, flange angle, external flange, oil gas meter body distance and the like. After the server acquires the field operation image, the server also acquires the characteristic data and performs data comparison.

In the comparison of the characteristic data of the field operation image and the normal instrument data, one or more data differences exceed the set normal operation range of the instrument, corresponding data in an abnormal instrument data set are called, abnormal instrument data with minimum special diagnosis data differences of the field operation image are judged, and the fault of the oil-gas instrument indicated by the field operation image is judged based on the concrete expression of the abnormal instrument data.

And if the oil gas instrument is judged to be abnormal based on the field operation image, overhauling is needed, and the field operation staff can finish the field operation task according to the corresponding maintenance instruction by determining the fault performance of the abnormal oil gas instrument and the corresponding maintenance instruction in the fault performance.

Under the conditions that abnormal oil gas instrument data and multi-step maintenance is required to be achieved during maintenance, after the maintenance work of the corresponding steps is completed, on-site operators shoot maintenance work results of the corresponding steps by using the mobile terminal and upload the maintenance work results, after the server receives the acquired on-site operation images, the server checks the completed maintenance results by means of an intelligent graph recognition algorithm, and if the maintenance quality is determined to meet the requirement, the mobile terminal automatically jumps to a maintenance description of the next maintenance step.

And for the final maintenance result of the oil-gas instrument, shooting on-site operation images after all maintenance tasks are completed by on-site operators, uploading the on-site operation images to a server, checking and accepting the on-site operation images by the server, extracting characteristic data of maintenance results by the server, setting data labels for the characteristic data, and combining the characteristic data with the on-site operation result images to generate maintenance result data.

After the final maintenance work of the oil gas instrument is finished, and after the server acquires the image and the characteristic data, the image and the data are transmitted to an adjustable database for storage, and the data are provided with time stamps, and in the next field operation of the oil gas instrument, the oil gas instrument data with the nearest time stamp are directly called and sent to a mobile terminal;

in some embodiments, the self-checking system of the oil gas instrument of the electric power system is also supported to check the working state of the oil gas instrument of the field operation, whether the oil gas instrument of the field operation can work normally is determined, if the self-checking result shows that the oil gas instrument can work normally, the field operation result is confirmed to be free of problems, the database of the field operation visual guidance system records the oil gas instrument image and the characteristic data returned by the movable end after the field operation, otherwise, the result with the overhaul problem is fed back, and the field operation personnel can check the work.

And step four, updating on-site operation information and recording information, injecting normal oil gas instrument images into a database, and obtaining characteristic values.

Considering that some oil and gas meters may change some information after a plurality of field operations, or the life cycle is shortened after the field operations, if such information is not updated and recorded, the guiding method provided by subsequent work is easily inconsistent with the actual situation, so that the information update and recording are necessary. The invention adopts the measure that the field operator actively submits the data to be updated and the field operator passively submits the data to be updated.

In a data mode of actively submitting to be updated aiming at field operators, setting endorsements in field operation images shot by a mobile terminal are supported, the endorsements are uploaded to a server along with the images, after the server receives the endorsements, intelligent image recognition is carried out on oil-gas meters in the field operation images, the quality of task completion is judged, if the requirements are not met, a feedback maintenance result is not met, if the field operators with the highest authority in an execution group confirm that the result meets the maintenance quality requirement of the oil-gas meters at the moment, after the personnel click confirmation, the server directly acquires the instruction, records signature information of the personnel in the execution group, records the images and the endorsements after the field operation tasks are completed in an adjustable database, and realizes the updating and recording of the field operation information.

In a data mode of passively submitting the amount of the field operation personnel to be updated, all data completed by the field operation personnel are directly received by a server and recorded by an adjustable database, so that the field operation information is updated and recorded.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.

The foregoing detailed description of the embodiments has further described the objects, technical solutions and advantageous effects of the present application, and it should be understood that the foregoing is only a detailed description of the present application and is not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present application should be included in the scope of protection of the present application.

Claims (10)

1. A visual guidance system for field operations, comprising:

the mobile terminal, the server and the database;

the mobile terminal is used for shooting and uploading instrument images and also used for receiving the instrument information issued by the server;

the server is used for receiving the instrument image uploaded by the mobile terminal and comparing the instrument image with the instrument information stored in the database;

the database is used for storing the instrument information;

the mobile terminal is in communication connection with the server, and the server is in communication connection with the database.

2. The field operation visualization guide system of claim 1, wherein:

the mobile terminal comprises an image display window, a photographing button, an image uploading button, a form display page, a work order display window, a search frame and a scanning frame;

the image display window is used for displaying images shot by the mobile terminal and received instrument images;

the photographing button is used for calling a mobile terminal camera, and when the photographing button is clicked, an image display window or a scanning frame is called;

the image uploading button is used for uploading the instrument image to the server;

the table display page displays a view layer table, lines and columns are divided through GridLayout, text view display data are filled, and ScrollView rolling data are used;

the work order display window is used for displaying field work order information issued to the mobile terminal by the server;

the scanning frame is used for scanning an instrument number nameplate or a two-dimensional code;

the search box is used for inputting keywords.

3. The field operation visualization guide system according to claim 2, wherein:

the search box stores historical search keywords and scrolls and displays the historical search keywords.

4. The field operation visualization guide system of claim 1, wherein:

the server sets up intelligent recognition algorithm, including:

acquiring the instrument image data set and classifying and grouping the image data;

setting a color threshold value, a saturation selection threshold value, a tone threshold value and a contrast brightness threshold value for the new instrument image, selecting at least 2 sets of threshold values, and executing enhancement processing to obtain a new instrument image with enhanced data;

and constructing an instrument YOLOV5 model based on the new instrument image.

5. The field operation visualization guide system of claim 4, wherein:

after the server acquires the field operation instrument image uploaded by the mobile terminal, the field operation instrument image is enhanced, and the enhanced field operation instrument image is compared with the YOLOV5 model.

6. The field operation visualization guide system of claim 4, wherein:

the intelligent recognition algorithm also comprises an improved recognition algorithm, comprising:

based on the new instrument image, combining a neural network to obtain a ncnn model;

adjusting the integral space domain weight, the positive side space domain weight and the step-by-step domain threshold value screening weight, and performing marginal reduction optimization;

and carrying out formation pooling concurrent classification deduction on the plurality of ncnn models, predicting the sufficiency degree of the candidate YOLOV5 models, and selecting the YOLOV5 model with the highest sufficiency degree.

7. The field operation visualization guide system of claim 1, wherein:

the database is an embedded database sqlite.

8. The field operation visualization guide system of claim 1, wherein:

a DAO interface is also included to perform data manipulation tasks.

9. A method of visual guidance of field operations for performing the system of any one of claims 1-8, comprising the steps of:

step one, importing the instrument images into the database, classifying and grouping the instrument images by the server, and carrying out enhancement processing on the instrument images to obtain new instrument images;

training the new instrument image by using a tuning device to obtain the YOLOV5 model;

step three, after receiving the field operation instrument image uploaded by the mobile terminal, the server identifies the instrument based on the YOLOV5 model;

and step four, after the instrument is identified, the server sends the instrument information to the mobile terminal and synchronously sends the field operation specification of the instrument.

10. The visual guidance method for field operations according to claim 9, wherein:

after the instrument is identified, the server sends the instrument information to the mobile terminal and synchronously sends the field operation specification of the instrument, and the method comprises the following steps:

when the server obtains the field operation instrument image and stores the field operation instrument image into the database, a time stamp is set;

the instrument image sent to the mobile terminal by the server is the field operation image corresponding to the timestamp closest to the field operation time.