Power grid tool conflict-free identification method and system based on multiple modes

Abstract

The invention discloses a power grid tool conflict-free identification method and system based on multiple modes, and relates to the field of power grid tool management.The power grid tool conflict-free identification method comprises the steps that RFID tags are set and installed according to the requirements of power grid tools, and an RFID tag reader-writer is arranged; processing response signal collision of a plurality of tags by using a fusion algorithm, and finding a matched tag; extracting a face feature vector through a convolutional neural network model, and comparing the face feature vector with information in a database to check an identity; performing permission judgment in combination with the position information, and generating operation record data; and the information is recorded in the database and used for data analysis and safety management and optimization of personnel management and energy distribution strategies. According to the invention, the phenomenon of signal collision when a plurality of RFID tags enter a reading range at the same time can be solved, and the operation safety and compliance of power grid tools are improved.

Description

Technical Field

[0001] The present invention relates to the field of power grid tool management, and in particular to a multi-modal-based power grid tool conflict-free identification method and system. Background Art

[0002] In the current field of power grid tool management, traditional management methods such as manual recording and barcode scanning, although they support daily operational needs to a certain extent, their limitations are becoming increasingly significant. Manual recording is not only inefficient, but also requires a lot of manpower to enter, check and update data, and data errors and omissions caused by human factors are difficult to avoid; although barcode scanning technology has improved the automation of data entry to a certain extent, its recognition speed is limited by the performance of the scanning equipment and the clarity of the barcode, and the recognition efficiency is significantly reduced in multi-target or long-distance scenarios, making it difficult to meet the large-scale and efficient management needs of power grid tools.

[0003] RFID (Radio Frequency Identification) technology, with its unique advantages, provides innovative solutions for the management of power grid tools. RFID technology uses radio waves for contactless automatic identification, and can quickly read tag information without human intervention, greatly improving identification efficiency and accuracy. At the same time, RFID technology has the ability to simultaneously identify multiple targets and read at long distances, and can easily cope with challenges such as the large number of power grid tools, wide distribution, and complex environments, and achieve rapid inventory and real-time tracking of power grid tools. In addition, the large-capacity storage capacity of RFID tags also makes it possible to manage detailed information on power grid tools, including usage records, maintenance history, safety status, etc., providing strong support for the safe operation of the power grid.

[0004] However, in actual applications, when multiple RFID tags enter the reading range at the same time, signal collision may occur, resulting in some tag information not being correctly read or the reading time being prolonged, which in turn affects management efficiency and accuracy. To address this problem, advanced anti-collision algorithms and optimized reading strategies are needed to ensure that the RFID system can operate stably and efficiently in a multi-tag environment and give full play to its potential in power grid tool management. Summary of the invention

[0005] The present invention provides a multi-modal-based conflict-free identification method and system for power grid tools, which can solve the problem that when multiple RFID tags enter the reading range at the same time, signal collision occurs, resulting in that part of the tag information cannot be correctly read or the reading time is prolonged.

[0006] In order to solve the above technical problems, a technical solution adopted by the present invention is to provide a method and system for identifying conflict-free power grid tools based on multi-modality, the method comprising: S100: Set and install RFID tags according to the usage requirements of power grid tools; S200: deploying an RFID tag reader at a key location, sending a radio frequency signal to the RFID tag, and the RFID tag receives the radio frequency signal and sends a response signal to the RFID tag reader; S300: using a fusion algorithm combining a slotted-ALOHA protocol and a query tree to process collisions caused by response signals sent simultaneously by a plurality of RFID tags, and finding an adapted RFID tag; S400: extracting the location information of the RFID tag reader from the database, and obtaining a facial feature vector using a convolutional neural network model; S500: extracting staff information from the database and comparing it with the facial feature vector, verifying the identity of the staff, and combining the location information of the RFID tag reader to make authority judgment, generate operation record data and record it in the database; S600: recording the operation record data in the database for subsequent data analysis and safety management and in-depth data mining to optimize personnel management and energy allocation strategies; The database refers to a database used to store power grid tool information, RFID tag information, RFID tag reader location information, staff information and operation record data; The power grid tool information includes: the power grid tool's UID (Unique Identifier), name, type, specification and model, production date, inspection date, usage status, storage location, staff authority and usage location authority; The key locations include: power grid tool storage entrances and exits, power grid tool operation locations, and operation location entrances and exits.

[0007] Furthermore, the setting and installing of RFID tags according to the use requirements of power grid tools refers to analyzing the use requirements of power grid tools, setting RFID tags for power grid tools based on the reading distance, durability, and type and quantity of power grid tools, and installing them on the power grid tools, and the RFID tag information includes the UID, type, specification and basic information of the manufacturer of the power grid tools; The usage requirements refer to the usage methods and usage scenarios in power grid tool management; The setting of the RFID tag refers to setting the type, size and shape of the RFID tag.

[0008] Furthermore, the sending of a radio frequency signal to the RFID tag, the RFID tag receiving the radio frequency signal and sending a response signal to the RFID tag reader / writer, means that the RFID tag reader / writer sends a radio frequency signal to the surroundings, the RFID tag receives the radio frequency signal and sends a response signal to the RFID tag reader / writer, and the RFID tag reader / writer captures the response signal and uses an EPC (Electronic Product Code) filtering algorithm to eliminate incomplete and interfered response signals to obtain a tag signal; The RFID tag reader / writer refers to an identification device that uses wireless radio frequency signals to transmit data and is used to communicate with RFID tags and read and write data.

[0009] Furthermore, the S300 includes: S310: Initialize the time extension length and number of time slots of the slotted ALOHA algorithm; S320: Divide a time period into time slots having the same number as the number of time slots, and form a time slot set; S330: All the RFID tags randomly select a time slot in the time slot set to send a response signal, and put all the RFID tags into the tag set; S340: Determine whether only one RFID tag responds in the same time slot. If so, it indicates that the tag is successfully identified and an adapted RFID tag is obtained. If not, it indicates that a tag conflict occurs and the process goes to S350. S350: Using a query tree algorithm to process the tag conflict and obtain an adapted RFID tag; The one time period refers to a time window allocated to the plurality of RFID tags to send response signals in the slotted ALOHA algorithm; The time slot set refers to a set consisting of all available time slots in the time period; The tag set refers to a set of all the RFID tags participating in the identification tag.

[0010] Furthermore, the S350 includes: S351: treating all the RFID tags that have the tag conflict collision as a conflict tag set and initializing them as the root node of the query tree; S352: Starting from the root node, dividing the nodes according to the common prefix of the UID of the RFID tag to obtain a query tree; S353: The RFID tag reader sends a query command, starts from the root node, recursively performs a binary query on the query tree, gradually queries and identifies the RFID tags corresponding to the leaf nodes, determines whether the query command matches the corresponding RFID tag, and obtains the adapted RFID tag; The node division according to the common prefix of the UID of the RFID tag means that each non-leaf node in the query tree represents the common prefix of the UID of the RFID tag corresponding to the child node under the non-leaf node, and further child node division is recursively performed according to the next UID prefix code of the UID of the RFID tag until each leaf node in the query tree represents a separate RFID tag; The query command refers to a command for searching for an adapted RFID tag, including the UID prefix represented by the current node.

[0011] Furthermore, the determination of whether the query command matches the corresponding RFID tag means that the RFID tag corresponding to each leaf node in the query tree responds according to the matching degree between the received query command and its own UID. If only one tag responds, the tag is successfully identified and removed from the query tree. If multiple RFID tags respond, it means that they share the same UID prefix and need to further refine the query. The further refined query means that the RFID tag reader updates the query command according to the next UID prefix code of the UID of the responding RFID tag, and sends it to the next level of the query tree until the maximum depth of the query tree is reached.

[0012] Further, the extracting the location information of the RFID tag reader from the database and obtaining the facial feature vector using the convolutional neural network model refers to extracting the location information of the RFID tag reader from the database, using a face recognition camera to capture the facial image of the staff member at the current moment and performing a preprocessing operation to obtain a key image, and extracting a feature vector from the key image using a convolutional neural network model to obtain a facial feature vector; The extracting the location information of the RFID tag reader from the database refers to extracting the location information of the installation location of the RFID tag reader based on the location information of the RFID tag reader entered into the database by the RFID tag reader.

[0013] Furthermore, the S500 includes: S510: extracting a face image from the staff information in the database, and extracting a feature vector using the convolutional neural network model; S520: Calculate the Euclidean distance between the feature vector and the face feature vector. If the Euclidean distance is less than or equal to the distance threshold, it means that the verification is passed and the process goes to S530. Otherwise, an alarm is triggered. S530: Check whether the staff is in a reasonable working place at the current moment. If so, go to S540, otherwise, trigger an alarm; S540: Check the current staff's authority to use the power grid tools, generate operation record data and record it in the database; The reasonable work location refers to the work location specified by the staff at the current moment according to the use location authority requirements; The triggering alarm refers to sending a violation to the system and recording the violation of the staff at the current moment in the operation record data in the database; The checking of the current staff's authority to use the power grid tools refers to checking whether the current staff has the authority to use the current power grid tools according to the staff information in the database, if yes, the check is passed, otherwise, an alarm is triggered; The generating of the operation record data refers to recording the entire operation process of the current staff, including compliant operations and illegal operations.

[0014] Furthermore, the system includes: a label face acquisition module, a data storage module, an application processing module and a user interaction module; The tag face acquisition module includes: an RFID tag reader / writer, an RFID tag, and a face recognition camera, which is used for signal interaction between the RFID tag reader / writer and the RFID tag, and can acquire the face image of the staff member, and transmit the signal and the face image to the application processing module for processing; The data storage module is used to store power grid tool information, RFID tag information, RFID tag reader location information, staff information and operation record data; The application processing module is used to process the signal data uploaded by the RFID tag reader and writer, and perform the logic processing of the anti-collision algorithm, extract the face image uploaded by the tag face acquisition module, extract the face feature vector, and perform face recognition judgment with the feature vector in the staff information in the data storage module; The user interaction module is used to provide a human-computer interaction interface to facilitate staff to query operation information and alarm information.

[0015] The beneficial effects of the present invention are:

[0016] 1. The present invention provides a multi-modal-based conflict-free identification method for power grid tools, which effectively reduces the errors caused by RFID tag collisions and ensures a conflict-free identification process through a fusion algorithm of slotted ALOHA and query tree.

[0017] 2. The present invention provides a multi-modal-based conflict-free identification method for power grid tools, which ensures that only authorized personnel can operate relevant power grid tools through the combination of face recognition and RFID location information, thereby reducing the risk of misoperation and unauthorized use. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the framework flow of a multi-modal-based power grid tool conflict-free identification method provided by the present invention.

[0019] Figure 2 The present invention provides a multi-modal power grid tool non-conflict identification method for processing conflict collisions using a fusion algorithm.

[0020] Figure 3 The present invention provides a schematic diagram of a flow chart of obtaining an adapted RFID for a method for conflict-free identification of power grid tools based on multi-modality.

[0021] Figure 4 It is a schematic diagram of power grid tool use authority review based on a multi-modal power grid tool conflict-free identification method provided by the present invention.

[0022] Figure 5 It is a schematic diagram of the framework flow of a multi-modal-based power grid tool conflict-free identification system provided by the present invention. DETAILED DESCRIPTION

[0023] The exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the invention and to fully convey the scope of the present invention to those skilled in the art.

[0024] Please refer to Figure 1 , which is a flow chart of a multi-modal-based conflict-free identification method for power grid tools provided by an embodiment of the present invention, the method includes steps S100 to S600.

[0025] S100: Set and install RFID tags according to the usage requirements of power grid tools.

[0026] Furthermore, the setting and installing of RFID tags according to the use requirements of power grid tools refers to analyzing the use requirements of power grid tools, setting RFID tags for power grid tools based on the reading distance, durability, and type and quantity of power grid tools, and installing them on the power grid tools, and the RFID tag information includes the UID, type, specification and basic information of the manufacturer of the power grid tools; The usage requirements refer to the usage methods and usage scenarios in power grid tool management; The setting of the RFID tag refers to setting the type, size and shape of the RFID tag.

[0027] For example, there are many types of power grid tools, including high-voltage pliers, insulating gloves, grounding wires, measuring instruments, ladders, pedals, foot buckles, etc. The frequency of use, storage location and working environment of each power grid tool are different. Requirements for the use of power grid tools: Some are high-frequency working power grid tools, such as measuring instruments and insulating gloves, which are taken out many times a day; while some power grid tools, such as ladders, pedals, and foot buckles, are used relatively less frequently and are only used in specific operations. RFID tags of different specifications are set for power grid tools used at different frequencies to facilitate accurate identification. Power grid tools may be used in a variety of environments, including warehouses, outdoor construction sites, high-voltage work areas, high-altitude work areas, etc. RFID tags also have good durability and can withstand the influence of harsh environments such as high temperature, low temperature, humidity, and dust.

[0028] For example, the setting and installation of RFID tags, according to the management requirements of different power grid tools, set the reading distance of RFID tags; RFID tags must have high wear resistance and waterproof and dustproof functions; for larger power grid tools, such as grounding wires or high-voltage clamps, larger hard tags can be used to facilitate installation on power grid tools and not easily damaged; and for small power grid tools such as insulating gloves that need to be used flexibly, small-sized, flexible tags can be selected to ensure that the tags do not affect the normal operation of power grid tools. For example: a flexible RFID tag is installed on the inside of a glove through a special pasting technology to ensure that the tag will not wear or fall off during use, and at the same time, it does not affect the comfort of use. For each power grid tool, the UID of the power grid tool is entered into the RFID tag, including basic information such as the name, type, specification, production date and manufacturer of the power grid tool. This information will be recorded by the system for subsequent automatic identification, management and maintenance.

[0029] It can be understood that through the above implementation methods, RFID tags can be set and installed for power grid tools to ensure that the power grid tools can be efficiently managed in different operating scenarios.

[0030] S200: Deploy RFID tag readers at key locations, send radio frequency signals to the RFID tags, and the RFID tags receive the radio frequency signals and send response signals to the RFID tag readers.

[0031] Furthermore, the key locations include: power grid tool storage entrances and exits, power grid tool operation locations, and operation location entrances and exits.

[0032] Furthermore, the sending of a radio frequency signal to the RFID tag, the RFID tag receiving the radio frequency signal and sending a response signal to the RFID tag reader / writer, means that the RFID tag reader / writer sends a radio frequency signal to the surroundings, the RFID tag receives the radio frequency signal and sends a response signal to the RFID tag reader / writer, and the RFID tag reader / writer captures the response signal and uses the EPC filtering algorithm to eliminate incomplete and interfered response signals to obtain a tag signal; The RFID tag reader / writer refers to an identification device that uses wireless radio frequency signals to transmit data and is used to communicate with RFID tags and read and write data.

[0033] For example, a high-frequency RFID tag reader is deployed at the entrance and exit of the power grid tool warehouse. The high-frequency RFID tag reader is responsible for reading the RFID tag information of all power grid tools entering and leaving the warehouse, ensuring that the entry and exit time, user and purpose of each power grid tool are recorded in real time. At the power construction site, such as the substation operation area or the high-voltage line maintenance area, a portable RFID tag reader is deployed to read and track the use of each power grid tool in real time to prevent the loss of power grid tools or left behind after the operation is completed; at the entrance and exit of the operation site, such as the entrance and exit of the safety isolation zone of the construction area, an RFID tag reader is installed. The reader can automatically scan the RFID tag when the power grid tools are taken out of or into the operation area, ensuring that the use and transfer of all power grid tools in the area are effectively monitored.

[0034] When the carrier of power grid tools passes through key locations, such as the entrances and exits of warehouses where power grid tools are stored, the RFID tag readers deployed at these locations will automatically send wireless radio frequency signals to the surrounding area. After receiving the signal, the RFID tag carried on the power grid tools will send a response signal to the reader. The response signal contains the UID and other basic information of the power grid tools. After the reader captures the response signal sent by the RFID tag, it uses the EPC filtering algorithm to filter and clean the captured signal. The filtering algorithm will eliminate incomplete response signals and interfered response signals to ensure that the RFID tag reader only receives valid response signals sent by the RFID tag.

[0035] It can be understood that the above implementation methods ensure efficient management of power grid tools during storage, operation and transfer, and greatly reduce the risk of loss and management confusion of power grid tools.

[0036] S300: Using a fusion algorithm combining slotted ALOHA and query tree to process collisions caused by response signals sent simultaneously by a plurality of RFID tags, and searching for an adapted RFID tag.

[0037] Please refer to Figure 2 , which is a schematic diagram of a conflict collision process using a fusion algorithm for processing a conflict collision in a multi-modal power grid tool non-conflict identification method provided by the present invention, wherein S300 includes steps S310 to S350.

[0038] S310: Initialize the time extension length and number of time slots of the slotted ALOHA algorithm; S320: Divide a time period into time slots having the same number as the number of time slots, and form a time slot set; S330: All the RFID tags randomly select a time slot in the time slot set to send a response signal, and put all the RFID tags into the tag set; S340: Determine whether only one RFID tag responds in the same time slot. If so, it indicates that the tag is successfully identified and an adapted RFID tag is obtained. If not, it indicates that a tag conflict occurs and the process goes to S350. S350: Using a query tree algorithm to process the tag conflict and obtain an adapted RFID tag; The one time period refers to a time window allocated to the plurality of RFID tags to send response signals in the slotted ALOHA algorithm; The time slot set refers to a set consisting of all available time slots in the time period; The tag set refers to a set of all the RFID tags participating in the identification tag.

[0039] Please refer to Figure 3 , which is a schematic diagram of the process of obtaining an adapted RFID in a multi-modal power grid tool conflict-free identification method provided by the present invention, wherein S350 includes steps S351 to S353.

[0040] S351: treating all the RFID tags that have the tag conflict collision as a conflict tag set and initializing them as the root node of the query tree; S352: Starting from the root node, dividing the nodes according to the common prefix of the UID of the RFID tag to obtain a query tree; S353: The RFID tag reader sends a query command, starts from the root node, recursively performs a binary query on the query tree, gradually queries and identifies the RFID tags corresponding to the leaf nodes, determines whether the query command matches the corresponding RFID tag, and obtains the adapted RFID tag; The node division according to the common prefix of the UID of the RFID tag means that each non-leaf node in the query tree represents the common prefix of the UID of the RFID tag corresponding to the child node under the non-leaf node, and further child node division is recursively performed according to the next UID prefix code of the UID of the RFID tag until each leaf node in the query tree represents a separate RFID tag; The query command refers to a command for searching for an adapted RFID tag, including the UID prefix represented by the current node.

[0041] Furthermore, the determination of whether the query command matches the corresponding RFID tag means that the RFID tag corresponding to each leaf node in the query tree responds according to the matching degree between the received query command and its own UID. If only one tag responds, the tag is successfully identified and removed from the query tree. If multiple RFID tags respond, it means that they share the same UID prefix and need to further refine the query. The further refined query means that the RFID tag reader updates the query command according to the next UID prefix code of the UID of the responding RFID tag, and sends it to the next level of the query tree until the maximum depth of the query tree is reached.

[0042] S400: extracting the location information of the RFID tag reader from the database, and obtaining a facial feature vector using a convolutional neural network model.

[0043] Furthermore, the database refers to a database for storing power grid tool information, RFID tag information, location information of RFID tag readers, staff information, and operation record data; The power grid tool information includes: the UID, name, type, specification and model, production date, inspection date, usage status, storage location, staff authority and usage location authority of the power grid tool.

[0044] Further, the extracting the location information of the RFID tag reader from the database and obtaining the facial feature vector using the convolutional neural network model refers to extracting the location information of the RFID tag reader from the database, using a face recognition camera to capture the facial image of the staff member at the current moment and performing a preprocessing operation to obtain a key image, and extracting a feature vector from the key image using a convolutional neural network model to obtain a facial feature vector; The extracting the location information of the RFID tag reader from the database refers to extracting the location information of the installation location of the RFID tag reader based on the location information of the RFID tag reader entered into the database by the RFID tag reader.

[0045] Exemplarily, first, the location information of the RFID tag reader is extracted from the database. The database records the installation locations of all RFID tag readers. In the current scenario, the RFID tag reader is installed at the entrance of the power grid maintenance warehouse. The specific location of this reader is determined by querying the corresponding location information in the database. With the cooperation of the face recognition camera, the facial image of the current staff member is captured. For example, when the staff enters the warehouse to prepare to use power grid tools, the camera will automatically capture their front face image. The captured image will undergo a series of preprocessing operations, such as image denoising, brightness adjustment, and image cropping, to ensure that the obtained image is clear and focused on the key parts of the face. After the image preprocessing is completed, the convolutional neural network model is used to extract features from the preprocessed key image to obtain the face feature vector. The feature extraction is calculated as follows: ; in, For the Tier feature map, is the feature map, is the feature map index, is the number of layers, is the horizontal axis coordinate of the image, is the vertical coordinate of the image, is the activation function, is the weight of the convolution kernel, is the index of the convolution kernel in the horizontal axis direction, is the index of the convolution kernel in the vertical axis direction, For the Tier The input value of the feature map, is the feature map index, For the Tier The bias term, is the feature map index, is the width of the convolution kernel, is the height of the convolution kernel.

[0046] S500: extracting the staff information in the database and comparing it with the facial feature vector, verifying the identity of the staff, and combining the location information of the RFID tag reader to make an authority judgment, generate operation record data and record it in the database.

[0047] Please refer to Figure 4 , which is a schematic diagram of a power grid tool use authority review based on a multi-modal power grid tool conflict-free identification method provided by the present invention, wherein S500 includes steps S510 to S540.

[0048] S510: extracting a face image from the staff information in the database, and extracting a feature vector using the convolutional neural network model; S520: Calculate the Euclidean distance between the feature vector and the face feature vector. If the Euclidean distance is less than or equal to the distance threshold, it means that the verification is passed and the process goes to S530. Otherwise, an alarm is triggered. S530: Check whether the staff is in a reasonable working place at the current moment. If so, go to S540, otherwise, trigger an alarm; S540: Check the current staff's authority to use the power grid tools, generate operation record data and record it in the database; The reasonable work location refers to the work location specified by the staff at the current moment according to the use location authority requirements; The triggering alarm refers to sending a violation to the system and recording the violation of the staff at the current moment in the operation record data in the database; The checking of the current staff's authority to use the power grid tools refers to checking whether the current staff has the authority to use the current power grid tools according to the staff information in the database, if yes, the check is passed, otherwise an alarm is triggered; The generating of the operation record data refers to recording the entire operation process of the current staff, including compliant operations and illegal operations.

[0049] Exemplarily, the basic information of the current staff is extracted from the database, including their name, unique identification code, photo, feature vector, and operation authority, where the feature vector is extracted from the photo of the staff using a convolutional neural network model. The Euclidean distance between the real-time facial feature vector of the staff and the feature vector extracted from the database is calculated. If the Euclidean distance is less than or equal to the distance threshold, it means that the identity verification is passed and the identity of the person is confirmed. The Euclidean distance calculation formula is: ,in, is the Euclidean distance, is the dimension of the feature vector, They are the real-time facial feature vector of the staff and the feature vector extracted from the database, Real-time facial feature vector for staff No. eigenvalues, The feature vector extracted from the database No. feature values.

[0050] Further check whether the staff is in a reasonable working place, for example: through the RFID tag reader, the staff is identified at the warehouse entrance to take the power grid tools, and the location information authority of the staff is queried in the database. If the authority indicates that the staff does not have the authority to use the power grid tools here, an alarm will be triggered. If so, the check will be passed, and further check whether the current staff has the authority to use the power grid tools. The database records the use authority information of each power grid tool in detail, including which personnel have the right to use it and where it can be used. For example: a power grid tool is a high-voltage cable, which can only be operated by specific senior maintenance personnel. If the staff does not have the relevant authority, an alarm will be issued immediately to prevent them from using it; if the authority meets the requirements, the check will pass and the operation will continue.

[0051] Furthermore, once the authority check is completed, the record data of the current operation is automatically generated. The record includes: the identity of the staff, the power grid tools used, the operation time, location, authority check results, and whether the operation is compliant. If any non-compliant behavior occurs in any step, such as identity verification, location verification, and authority check, an alarm is immediately triggered, and the violation is recorded in the operation record data for the administrator or superior to review and handle.

[0052] It can be understood that through the above implementation steps, a comprehensive review of the staff's identity, work location and power grid tool use rights is achieved, and a complete operation record is generated, which not only ensures the safe use of power grid tools, but also can promptly detect and prevent potential illegal operations, thereby ensuring the smooth progress of power grid maintenance work.

[0053] S600: The operation record data is recorded in the database for subsequent data analysis and safety management and in-depth data mining to optimize personnel management and energy allocation strategies.

[0054] Please refer to Figure 5Based on the above-mentioned multi-modal power grid tool conflict-free identification method, the present invention also provides a multi-modal power grid tool conflict-free identification system, the system comprising: a tag face acquisition module, a data storage module, an application processing module and a user interaction module.

[0055] Furthermore, the system includes: a label face acquisition module, a data storage module, an application processing module and a user interaction module; The tag face acquisition module includes: an RFID tag reader / writer, an RFID tag, and a face recognition camera, which is used for signal interaction between the RFID tag reader / writer and the RFID tag, and can acquire the face image of the staff member, and transmit the signal and the face image to the application processing module for processing; The data storage module is used to store power grid tool information, RFID tag information, RFID tag reader location information, staff information and operation record data; The application processing module is used to process the signal data uploaded by the RFID tag reader and writer, and perform the logic processing of the anti-collision algorithm, extract the face image uploaded by the tag face acquisition module, extract the face feature vector, and perform face recognition judgment with the feature vector in the staff information in the data storage module; The user interaction module is used to provide a human-computer interaction interface to facilitate staff to query operation information and alarm information.

[0056] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A multi-modal power grid tool conflict-free identification method, characterized in that: The method comprises: S100: Set and install RFID tags according to the usage requirements of power grid tools; S200: deploying an RFID tag reader at a key location, sending a radio frequency signal to the RFID tag, and the RFID tag receives the radio frequency signal and sends a response signal to the RFID tag reader; S300: using a fusion algorithm combining slotted ALOHA and query tree to process collisions caused by response signals sent simultaneously by multiple RFID tags, and finding an adapted RFID tag; S400: extracting the location information of the RFID tag reader from the database, and obtaining a facial feature vector using a convolutional neural network model; S500: extracting staff information from the database and comparing it with the facial feature vector, verifying the identity of the staff, and combining the location information of the RFID tag reader to make authority judgment, generate operation record data and record it in the database; S600: recording the operation record data in the database for subsequent data analysis and safety management and in-depth data mining to optimize personnel management and energy allocation strategies; The database refers to a database used to store power grid tool information, RFID tag information, RFID tag reader location information, staff information and operation record data; The power grid tool information includes: the UID, name, type, specification and model, production date, inspection date, usage status, storage location, staff authority and usage location authority of the power grid tool; The key locations include: power grid tool storage entrances and exits, power grid tool operation locations, and operation location entrances and exits.

2. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 1, characterized in that: The setting and installation of RFID tags according to the use requirements of power grid tools refers to analyzing the use requirements of power grid tools, setting RFID tags for power grid tools based on the reading distance, durability, and type and quantity of power grid tools, and installing them on the power grid tools, and the RFID tag information includes the UID, type, specification and basic information of the manufacturer of the power grid tools; The usage requirements refer to the usage methods and usage scenarios in power grid tool management; The setting of the RFID tag refers to setting the type, size and shape of the RFID tag.

3. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 1, characterized in that: The sending of radio frequency signals to the RFID tag, the RFID tag receiving the radio frequency signals and sending a response signal to the RFID tag reader-writer, means that the RFID tag reader-writer sends radio frequency signals to the surroundings, the RFID tag receives the radio frequency signals and sends a response signal to the RFID tag reader-writer, and the RFID tag reader-writer captures the response signals and uses the EPC filtering algorithm to eliminate incomplete and interfered response signals to obtain the tag signal; The RFID tag reader / writer refers to an identification device that uses wireless radio frequency signals to transmit data and is used to communicate with RFID tags and read and write data.

4. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 1, characterized in that: The S300 includes: S310: Initialize the time extension length and number of time slots of the slotted ALOHA algorithm; S320: Divide a time period into time slots having the same number as the number of time slots, and form a time slot set; S330: All the RFID tags randomly select a time slot in the time slot set to send a response signal, and put all the RFID tags into the tag set; S340: Determine whether only one RFID tag responds in the same time slot. If so, it indicates that the tag is successfully identified and an adapted RFID tag is obtained. If not, it indicates that a tag conflict occurs and the process goes to S350. S350: Using a query tree algorithm to process the tag conflict and obtain an adapted RFID tag; The one time period refers to a time window allocated to the plurality of RFID tags to send response signals in the slotted ALOHA algorithm; The time slot set refers to a set consisting of all available time slots in the time period; The tag set refers to a set of all the RFID tags participating in the identification tag.

5. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 4, characterized in that: The S350 includes: S351: treating all the RFID tags that have the tag conflict collision as a conflict tag set and initializing them as the root node of the query tree; S352: Starting from the root node, dividing the nodes according to the common prefix of the UID of the RFID tag to obtain a query tree; S353: The RFID tag reader sends a query command, starts from the root node, recursively performs a binary query on the query tree, gradually queries and identifies the RFID tags corresponding to the leaf nodes, determines whether the query command matches the corresponding RFID tag, and obtains the adapted RFID tag; The node division according to the common prefix of the UID of the RFID tag means that each non-leaf node in the query tree represents the common prefix of the UID of the RFID tag corresponding to the child node under the non-leaf node, and further child node division is recursively performed according to the next UID prefix code of the UID of the RFID tag until each leaf node in the query tree represents a separate RFID tag; The query command refers to a command for searching for an adapted RFID tag, including the UID prefix represented by the current node.

6. A method for identifying non-conflict of power grid tools based on multi-modality according to claim 5, characterized in that: The determination of whether the query command matches the corresponding RFID tag means that the RFID tag corresponding to each leaf node in the query tree responds according to the matching degree between the received query command and its own UID. If only one tag responds, the tag is successfully identified and removed from the query tree. If multiple RFID tags respond, it means that they share the same UID prefix and need to further refine the query. The further refined query means that the RFID tag reader updates the query command according to the next UID prefix code of the UID of the responding RFID tag, and sends it to the next level of the query tree until the maximum depth of the query tree is reached.

7. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 1, characterized in that: The extracting the location information of the RFID tag reader from the database and obtaining the face feature vector using the convolutional neural network model refers to extracting the location information of the RFID tag reader from the database, using a face recognition camera to capture the face image of the staff member at the current moment and performing a preprocessing operation to obtain a key image, and extracting a feature vector from the key image using a convolutional neural network model to obtain a face feature vector; The extracting the location information of the RFID tag reader from the database refers to extracting the location information of the installation location of the RFID tag reader based on the location information of the RFID tag reader entered into the database by the RFID tag reader.

8. The method for identifying non-conflict of power grid tools based on multi-modality according to claim 1, characterized in that: The S500 comprises: S510: extracting a face image from the staff information in the database, and extracting a feature vector using the convolutional neural network model; S520: Calculate the Euclidean distance between the feature vector and the face feature vector. If the Euclidean distance is less than or equal to the distance threshold, it means that the verification is passed and the process goes to S530. Otherwise, an alarm is triggered. S530: Check whether the staff is in a reasonable working place at the current moment. If so, go to S540, otherwise, trigger an alarm; S540: Check the current staff's authority to use the power grid tools, generate operation record data and record it in the database; The reasonable work location refers to the work location specified by the staff at the current moment according to the use location authority requirements; The triggering alarm refers to sending a violation to the system and recording the violation of the staff at the current moment in the operation record data in the database; The checking of the current staff's authority to use the power grid tools refers to checking whether the current staff has the authority to use the current power grid tools according to the staff information in the database, if yes, the check is passed, otherwise, an alarm is triggered; The generating of the operation record data refers to recording the entire operation process of the current staff, including compliant operations and illegal operations.

9. A multi-modal power grid tool conflict-free identification system, characterized in that: The system includes: a label face collection module, a data storage module, an application processing module and a user interaction module; The tag face acquisition module includes: an RFID tag reader / writer, an RFID tag, and a face recognition camera, which is used for signal interaction between the RFID tag reader / writer and the RFID tag, and can acquire the face image of the staff member, and transmit the signal and the face image to the application processing module for processing; The data storage module is used to store power grid tool information, RFID tag information, RFID tag reader location information, staff information and operation record data; The application processing module is used to process the signal data uploaded by the RFID tag reader and writer, and perform the logic processing of the anti-collision algorithm, extract the face image uploaded by the tag face acquisition module, extract the face feature vector, and perform face recognition judgment with the feature vector in the staff information in the data storage module; The user interaction module is used to provide a human-computer interaction interface to facilitate staff to query operation information and alarm information.

Images