A power tool lending management method and system

Abstract

This invention discloses a method and system for managing the requisition of electrical appliances, relating to the field of appliance management technology. It includes constructing a historical knowledge base, standardizing fault text descriptions into multiple fault types based on historical fault work order data using natural language processing technology, and establishing a fault-tool mapping set for each fault type. Tools are dynamically marked as frequently used or occasionally used tools based on their probability of occurrence in historical work orders corresponding to the fault type. This application can improve operational safety and prevent the omission of core tools: by establishing a "fault-tool" mapping model and dynamically marking the core "frequently used tools set," the system can perform mandatory verification during the requisition process. Once a core tool is detected as unclaimed, an audible and visual alarm is triggered, and the process is intercepted, fundamentally eliminating the problem of safety tool omissions due to human negligence and greatly reducing the risk of serious accidents such as electric shock and falls from heights.

Description

Technical Field

[0001] This invention belongs to the field of appliance management technology, specifically a method for managing the requisition of electrical appliances. Background Technology

[0002] In the daily operation and maintenance and emergency repair of power systems, the standardized requisition and management of electrical safety tools and specialized maintenance equipment are crucial for ensuring operational safety and improving repair efficiency. Currently, the requisition and management of electrical equipment largely relies on manual experience. When a fault occurs, repair personnel typically determine which tools need to be carried based on the personal experience of the person in charge. This model has significant drawbacks: First, its heavy reliance on personal experience makes it prone to overlooking essential tools (such as voltage detectors and personal safety grounding wires) due to varying skill levels or negligence, creating serious safety hazards. Second, for infrequent fault types or newly hired employees, insufficient experience may lead to inappropriate tool preparation, affecting the repair progress.

[0003] Existing intelligent tool management systems, such as smart tool cabinets using RFID technology for tool entry and exit management, while enabling digital recording and anti-theft measures, primarily focus on asset management and entry / exit control, lacking intelligent integration with specific operational scenarios. They cannot proactively and intelligently determine the completeness and appropriateness of the required toolset based on the specific type of fault, failing to fundamentally address the core safety issue of "bringing the wrong tools due to inexperience or negligence." Therefore, there is an urgent need in this field for a management method and system that can intelligently link fault information with tool requisition needs, enabling real-time verification and proactive early warning, to overcome the shortcomings of existing technologies. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a method and system for the management of the requisition of electrical appliances.

[0005] A method for managing the requisition of electrical appliances, comprising: A historical knowledge base is constructed. Based on historical fault work order data, the fault text description standard is standardized into multiple fault types through natural language processing technology. A fault-tool mapping set is established for each fault type, where tools are dynamically marked as frequently used tools or occasionally used tools based on their probability of occurrence in the historical work orders of the corresponding fault type. Real-time fault triggering and toolset matching: When a new fault occurs, the fault description is analyzed and matched with the standard fault type, and the corresponding commonly used toolset and occasionally used toolset are retrieved from the fault-tool mapping set. Intelligent requisition monitoring and differentiated response: During the tool requisition process, the actual toolset requisitioned is monitored in real time and compared with the frequently used toolset and occasionally used toolset. Based on the comparison results, differentiated response operations are performed, including triggering an alarm and blocking the outbound process when frequently used tools are missing, and providing prompt information when occasionally used tools are retrieved.

[0006] Furthermore, in the step of constructing the historical knowledge base, the condition for marking a tool as a frequently used tool is that the tool appears in the historical work orders of the corresponding fault type with a probability greater than 80%; the condition for marking it as an occasionally used tool is that the probability of its appearance is between 25% and 80%.

[0007] Furthermore, in the differentiated response operation, when the actual toolset does not contain all commonly used tools, visual and auditory alarms are triggered, alarm information is sent to the mobile terminal of the person in charge of the work, and the tool cabinet outbound confirmation function is disabled.

[0008] Furthermore, in the differentiated response operation, when the actual toolset being retrieved includes occasional tools, a prompt message is displayed but the retrieval process is not interrupted.

[0009] Furthermore, it also includes: The global data analysis process involves calculating the universality index of each tool and building a universal tool library based on the universality index. In the real-time fault triggering and toolset matching step, a general toolset is also invoked; In the intelligent requisition monitoring process, a non-blocking reminder is provided when the actual requisition toolset does not include common tools.

[0010] Furthermore, the formula for calculating the generality index Ui is: Ui = (Number of fault types encountered by this tool / Total number of fault types M) * α + (Total number of work orders encountered by this tool / Total number of work orders) * β; Where α and β are adjustable parameters.

[0011] Furthermore, in the real-time fault triggering and toolset matching step, a comprehensive recommendation list is generated, including core tools, general tools, and contextual tools.

[0012] An electrical appliance requisition management system is configured to execute the aforementioned electrical appliance requisition management method.

[0013] Furthermore, it includes a smart tool cabinet, which has a built-in RFID reader for real-time identification of tool retrieval status.

[0014] Furthermore, it includes a mobile terminal for receiving alarm information and for the person in charge of the work to confirm the operation.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This application can improve operational safety and prevent the omission of core tools: by establishing a "fault-tool" mapping model and dynamically marking the core "common toolset", the system can perform mandatory verification during the requisition process. Once it is detected that a core tool (such as an electric detector) has not been requisitioned, an audible and visual alarm is triggered and the process is blocked, fundamentally eliminating the problem of safety tool omissions due to human negligence and greatly reducing the risk of serious accidents such as electric shock and falls from heights.

[0016] Simultaneously, it enables intelligent and precise tool allocation, reducing reliance on individual experience: This invention transforms the tool preparation process, which depends on experienced technicians' personal experience, into a standardized and automated recommendation process based on historical big data analysis. The system can accurately match and recommend specialized and general tools based on real-time fault information, providing clear and reliable guidance for new employees or teams handling unfamiliar fault types, lowering the personnel threshold, and improving the overall standardization of work. Attached Figure Description

[0017] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Please see Figure 1 This application provides a method for the management of the requisition of electrical appliances; As an embodiment of this application, the method specifically includes the following steps: A historical knowledge base is constructed. Based on historical fault work order data, the fault text description standard is standardized into multiple fault types through natural language processing technology. A fault-tool mapping set is established for each fault type, where tools are dynamically marked as frequently used tools or occasionally used tools based on their probability of occurrence in the historical work orders of the corresponding fault type. Real-time fault triggering and toolset matching: When a new fault occurs, the fault description is analyzed and matched with the standard fault type, and the corresponding commonly used toolset and occasionally used toolset are retrieved from the fault-tool mapping set. Intelligent requisition monitoring and differential response. During the tool requisition process, the actual tool set being received is monitored in real time and compared with the commonly used tool set and occasionally used tool set retrieved. Based on the comparison results, differential response operations are executed, including triggering an alarm and intercepting the outbound process when a commonly used tool is missing, and providing a prompt message when an occasionally used tool is retrieved.

[0020] Embodiment 2: As Embodiment 2 of this application, the method in this embodiment specifically includes: Step S101, construct a historical knowledge base; The system extracts the fault handling work order data for the past five years from the production management system (PMS). Each piece of data includes: fault code, fault text description, processing result, and the list of tools used.

[0021] S1011, standardize the fault type: Use natural language processing technology to extract keywords and perform semantic analysis on the fault text description, and merge the faults into several standard types.

[0022] For example, merge all work orders described as overheating of the transformer, alarm of the main transformer oil temperature, etc. into the transformer overheating fault category; the existing technology is used for this processing here, so no specific details will be elaborated; S1012, establish a fault-tool mapping set: For each standard fault type, count the tools that appear in all its historical work orders to form an initial tool pool.

[0023] S1013, dynamic tool marking: In this step, a core algorithm is introduced to classify the importance of tools: Calculate the appearance probability P of each tool in each sample fault type. The calculation formula is: P = the number of times the tool appears / the total number of work orders of this type of fault; Here, the number of times the tool appears refers to the number of times the tool is used in the corresponding sample fault type; Mark as a commonly used tool: If P > 80%, it is determined that this tool is a core essential tool for handling such faults.

[0024] Mark as an occasionally used tool: If 25% < P ≤ 80%, it is determined that this tool is a situational tool, and its use is strongly is strongly related to the specific details of the fault.

[0025] Associate the marking results with the corresponding fault types and store them in the fault type-tool set mapping model.

[0026] Example: For the 10kV overhead line lightning strike and wire break fault type, there are a total of 100 historical work orders.

[0027] Insulating gloves, voltage detectors, personal safety grounding wires, and wire cutters appeared 98 times (P-98%), and are marked as commonly used tools. Foot straps, safety belts, and ropes appeared 45 times (P=45%), and are marked as occasionally used tools (because some wire breakage locations can be handled without climbing).

[0028] Step S102: Real-time fault triggering and toolset matching; S1021: The dispatch center received a new fault report: the porcelain insulator of phase B on pole No. 28 of the 35kV Zhuyuan line has been broken down. The system obtains this information in real time through the interface. S1022: The system performs real-time analysis of the fault description and matches it to the standard fault type, assuming it is an overhead line insulator breakdown fault. S1023: The system immediately retrieves the commonly used toolset TC and occasionally used toolset To corresponding to the fault type from the fault type-toolset mapping model, and generates a recommendation list.

[0029] Step S103: Intelligent requisition monitoring and differentiated response; Repair team member Zhang San swiped his card to open the smart tool cabinet and prepared to collect tools; S1031. Real-time monitoring: Every time Zhang San takes a tool, the RFID reader in the cabinet automatically identifies the tool's ID, and the system adds it to the actual tools collection set Ta.

[0030] S1032, Logical Judgment and Response: The system compares Ta with Tc and To in real time and executes the following differentiated response strategy: Scenario 1: Core tools omitted; Judgment condition: There exists an instrument t such that t∈Tc and t∉Ta; Response action: 1. The tool cabinet display screen is flashing red and a warning box pops up: Alarm: Core tool voltage detector not retrieved! Please check immediately, otherwise it may lead to a serious safety accident! 2. Triggers a continuous buzzing alarm.

[0031] 3. At the same time, an alarm message was pushed to the mobile terminal of Li Si, the person in charge of the work: Zhang San of your team forgot to use the core tool, the voltage detector, when preparing tools for insulator breakdown faults; 4. Process Interception: The tool cabinet confirmation button is disabled until Zhang San puts the missing tool into the cabinet, or the person in charge, Li Si, forces confirmation on the mobile terminal.

[0032] Scenario 2: Using situational tools Judgment condition: There exists an instrument k such that k∈To and k∈Ta; Response action: 1. The tool cabinet display screen shows a yellow warning message: Reminder: You have received your foot straps and safety belt. This operation may require climbing. Please double-check that your safety measures are in place. 2. This is an informative reminder that does not interrupt the requisition process, but it serves as a crucial secondary safety reminder. Scenario 3: The requisition and use are compliant; Judgment condition: Tc⊆Ta, meaning the actual tools used have fully covered the commonly used toolset. Response action: 1. The tool cabinet display screen shows a green checkmark and the text: Tools are ready, please be careful! 2. Zhang San clicks to confirm the outbound shipment. The system records the complete requisition list, personnel, and timestamp, and the process ends.

[0033] Example 3: As a third embodiment of this application, this embodiment is implemented based on embodiment two, except that the method in this embodiment specifically includes: Step S201: Global data analysis and generality calculation; The system performs a full analysis of all historical fault work orders. All historical fault work orders are analyzed here without distinguishing fault type. The specific analysis method is as follows: Assume the system has M different standard fault types; Assume there are N different types of tools in the tool library; For each tool ti (i=1 to N), calculate its generality index Ui; The formula for calculating the generality index Ui is: Ui = (Number of fault types encountered by this tool / Total number of fault types M) * α + (Total number of work orders encountered by this tool / Total number of work orders) * β; Among them, α and β are adjustable parameters used to balance the importance of coverage breadth and usage frequency; for example, α=0.6 and β=0.4 can be set to place more emphasis on the tool's ability to be applicable across scenarios.

[0034] Step S202: Establish a general tool library; Set a universality threshold Ut; Mark all tools that satisfy Ui>Ut as general-purpose tools and store them in the general-purpose tool library list Tu.

[0035] Exemplary analysis results: The tool's personal safety line, also known as the grounding line, may appear in 95% of fault types and has a very high frequency of occurrence in the overall work order. It has a very high Ui value and is marked as a general-purpose tool.

[0036] Tool insulating gloves: They may appear in 80% of fault types, meaning they are required for almost all live-line work. They appear frequently in work orders, have high Ui values, and are marked as general-purpose tools.

[0037] Multimeter: It covers a wide range of fault types, which are electrical measurements, but it is not used for every job. Its UI value may still reach the threshold, so it is marked as a general-purpose tool.

[0038] Transformer oil chromatography analyzer: It only appears in a very few types of faults, such as internal transformer faults. Although it is important in its specific field, its Ui value is extremely low and it is not marked as a general-purpose tool.

[0039] Step S203: Integrate general tools into the requisition process. It is recommended that this step be seamlessly integrated with the process of the first embodiment to form a complete requisition verification logic.

[0040] 1. Triggering and Matching: When the system receives a new fault alarm and matches the dedicated commonly used toolset Tc and the dedicated occasional toolset To, it automatically calls the general toolset Tu from the general toolset library.

[0041] 2. Generate a comprehensive recommendation list: The system generates a three-column visual list for emergency repair personnel: Core tool: from Tc.

[0042] General tools: from Tu.

[0043] Contextual tools: from To.

[0044] 3. Intelligent Monitoring and Response: In the requisition monitoring phase, in addition to executing the rules of the first embodiment, the system adds a monitoring strategy for general tools. Judgment condition: If there exists a tool g such that g∈Tu and g∈Ta, then the general tool has not been claimed.

[0045] Response action: On the tool cabinet display screen, the general tool item is shown in blue with a prompt icon.

[0046] Text prompt: Friendly reminder: The general-purpose multimeter has not been collected. It is recommended to bring it with you in case of emergency. This prompt is a non-blocking reminder, will not trigger alarms, and will not interrupt the process; it is only a friendly suggestion. Personnel can decide whether to collect it based on the specific circumstances of this task.

[0047] Example 4: Example 4 is based on Example 2, except that the specific method of describing the fault text using natural language processing technology in step S1011 is as follows: SS01: Use Chinese word segmentation tools such as jieba. During this process, load the administrator's preset power field dictionary. Through the field dictionary, ensure that professional terms (such as single-phase grounding, SF6 leakage) are not incorrectly segmented. SS02: Keyword Extraction and Expansion The TF-IDF (Term Frequency-Inverse Document Frequency) statistical method was used to identify the characteristic words in each type of fault text; The TextRank algorithm is used to extract key phrases from the text; Semantic expansion can be performed using the HIT IR-Lab Tongyici Cilin or a thesaurus in the power industry; for example, associating main transformers with transformers, and porcelain insulators with insulators. SS03: Vectorized Representation: Method A (classic machine learning): Use TF-IDF Vectorizer to convert text into numerical vectors; this method is simple, effective, and highly interpretable.

[0048] Method B (Deep Learning / Semantic Understanding): Use pre-trained language models, such as BERT and ERNIE (a knowledge-enhancing model launched by Baidu), to obtain deep semantic vectors (embeddings) of the text. This method can better understand the context, such as distinguishing between switch refusal and switch malfunction. SS04: Using a hierarchical classification strategy to address the complex tree-like structure of power faults: 1. First-level classification: Identification of major equipment categories; Objective: To determine which master device the fault occurred on.

[0049] Categories: Transformers, circuit breakers, disconnect switches, transmission lines, capacitors, relay protection devices, etc.

[0050] Features: Use equipment keywords as strong features (e.g., if the text contains the main transformer, it is very likely that the transformer is faulty).

[0051] Model: Naive Bayes or SVM (Support Vector Machine) can be used because the number of categories is small and the features are obvious, and simple models already perform well.

[0052] 2. Second-level classification: Specific fault mode identification Objective: After determining the major categories of equipment, further subdivide the specific phenomena or causes of the faults.

[0053] Example: For transformers, the subcategories are: excessively high oil temperature, winding fault, bushing damage, light gas activation, cooler failure, etc.

[0054] For transmission lines, the subcategories are: lightning strikes causing line breaks, wind-induced discharges, ice-induced collapses, bird-induced short circuits, and tree-related grounding.

[0055] Model: Train a separate secondary classifier for each device category.

[0056] Because the semantics of subcategories are more complex, it is recommended to use deep learning models such as BERT or TextCNN (Text Convolutional Neural Network) to capture more subtle semantic differences.

[0057] SS05: Model Training and Evaluation; Data annotation: A batch of historical work order data that has been annotated by experts is needed as the training set and test set.

[0058] Training process: First, train the first-level model, and then use the prediction results of the first level to split the data into the corresponding second-level models for training.

[0059] Evaluation metrics: Accuracy, Precision, Recall, and F1 score are used to evaluate model performance.

[0060] Example 5: Example 5: This example provides a power appliance requisition and management system, configured to implement a power appliance requisition and management method mentioned in any one of Examples 1 to 4; specifically including: The intelligent tool cabinet has a built-in RFID reader for real-time identification of tool retrieval status; The mobile terminal is used to receive alarm information and for the person in charge of the work to confirm the operation.

[0061] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for managing the requisition of electrical appliances, characterized in that, include: A historical knowledge base is built. Based on historical fault work order data, the fault text description standard is standardized into multiple fault types through natural language processing technology. A fault-tool mapping set is established for each fault type. Tools are dynamically marked as frequently used tools or occasionally used tools based on their probability of appearing in historical work orders of the corresponding fault type. Real-time fault triggering and toolset matching: When a new fault occurs, the fault description is analyzed and matched with the standard fault type, and the corresponding commonly used toolset and occasionally used toolset are retrieved from the fault-tool mapping set. Intelligent requisition monitoring and differentiated response: During the tool requisition process, the actual toolset requisitioned is monitored in real time and compared with the frequently used toolset and occasionally used toolset. Based on the comparison results, differentiated response operations are performed, including triggering an alarm and blocking the outbound process when frequently used tools are missing, and providing prompt information when occasionally used tools are retrieved.

2. The method for managing the requisition of electrical appliances according to claim 1, characterized in that, In the step of building the historical knowledge base, the condition for marking a tool as a frequently used tool is that the tool appears in the historical work orders of the corresponding fault type with a probability greater than 80%; the condition for marking it as an occasionally used tool is that the probability of its appearance is between 25% and 80%.

3. The method for managing the requisition of electrical appliances according to claim 1, characterized in that, In the differentiated response operation, when the actual toolset does not contain all commonly used tools, visual and auditory alarms are triggered, alarm information is sent to the mobile terminal of the person in charge of the work, and the tool cabinet outbound confirmation function is disabled.

4. The method for managing the requisition of electrical appliances according to claim 1, characterized in that, In the differentiated response operation, when the actual toolset received includes occasional tools, a prompt message is displayed but the receiving process is not interrupted.

5. The method for managing the requisition of electrical appliances according to claim 1, characterized in that, Also includes: The global data analysis process involves calculating the universality index of each tool and building a universal tool library based on the universality index. In the real-time fault triggering and toolset matching step, a general toolset is also invoked; In the intelligent requisition monitoring process, a non-blocking reminder is provided when the actual requisition toolset does not include common tools.

6. The method for managing the requisition of electrical appliances according to claim 5, characterized in that, The formula for calculating the universality index Ui is: Ui = (Number of fault types encountered by this tool / Total number of fault types M) * α + (Total number of work orders encountered by this tool / Total number of work orders) * β; Where α and β are adjustable parameters.

7. The method for managing the requisition of electrical appliances according to claim 5, characterized in that, In the real-time fault triggering and toolset matching step, a comprehensive recommendation list is generated, including core tools, general tools, and contextual tools.

8. A power appliance requisition and management system, characterized in that, It is configured to perform the electrical appliance requisition management method as described in any one of claims 1 to 7.

9. A power appliance requisition and management system according to claim 8, characterized in that, This includes a smart tool cabinet, which has a built-in RFID reader for real-time identification of tool retrieval status.

10. A power appliance requisition and management system according to claim 8, characterized in that, This includes mobile terminals used to receive alarm information and for the person in charge to confirm the operation.

Images