Coal mine equipment fault emergency maintenance scheduling method and system based on dynamic workflow

Abstract

The invention provides a coal mine equipment fault emergency maintenance scheduling method and system based on a dynamic workflow, and the method comprises the steps: collecting equipment operation state data in real time through a sensor disposed on coal mine equipment, and receiving manually inputted fault information through a mobile terminal, then, whether a fault electronic work order is generated or not is judged based on the equipment operation state data and the input fault information; when the fault electronic work order is generated, obtaining a resource state matrix of each department, and performing similar historical case matching of the fault electronic work order to obtain a spare part model, a maintenance scheme and an emergency degree corresponding to the fault electronic work order; and determining a resource allocation scheme and a shortest transportation path according to the spare part model, the maintenance scheme, the emergency degree and the resource state matrix of each department, and executing the resource allocation scheme and the shortest transportation path to realize coal mine equipment fault emergency maintenance scheduling. According to the technical scheme provided by the invention, rapid, accurate and transparent emergency maintenance scheduling of coal mine equipment faults is realized.

Description

Technical Field

[0001] This application relates to the field of industrial equipment operation and maintenance technology, and in particular to a method and system for emergency repair and scheduling of coal mine equipment faults based on dynamic workflow. Background Technology

[0002] The underground production environment in coal mines is complex, and equipment operates under high loads for extended periods, making it prone to various malfunctions. The timeliness and effectiveness of emergency repairs directly impact the safety and continuous operation of coal mines. Currently, emergency repair scheduling for coal mine equipment malfunctions largely relies on traditional manual communication and coordination models, which present the following prominent problems: First, low efficiency in cross-departmental collaboration. Emergency repairs involve multiple departments, including the dispatch center, storage center, transportation team, and electromechanical department. Under the existing model, the transmission path for fault information is lengthy, and communication methods are outdated. From the discovery of a fault on-site to its final resolution, multiple phone calls, walkie-talkies, and even personnel exchanges are required, and information is prone to distortion and delays during transmission. Unclear boundaries of departmental responsibilities and the lack of a unified and transparent collaborative work platform easily lead to mutual buck-passing and slow responses, severely restricting the efficiency of fault handling. Second, dynamic resource allocation relies on manual methods and lacks intelligent decision support. Existing resource allocation (including spare parts, transportation vehicles, and technical personnel) heavily relies on the dispatcher's experience and personal judgment, lacking multi-dimensional comprehensive analysis of the urgency of the fault and the real-time status of resources. This allocation method is highly subjective, making it difficult to achieve optimal resource allocation. It often leads to problems such as unreasonable resource allocation, suboptimal transportation routes, and the inability of key technical personnel to arrive in a timely manner, thus prolonging the overall time for troubleshooting. Third, troubleshooting experience and knowledge are difficult to effectively accumulate and reuse. During long-term equipment operation and maintenance, a large number of troubleshooting cases and expert experience have been accumulated, but this knowledge exists mostly in the form of unstructured text records or scattered personal experiences, failing to undergo systematic digital organization and mining. When encountering similar faults, maintenance personnel cannot quickly and accurately retrieve historical solutions, resulting in the inability to effectively inherit and reuse valuable experience and knowledge. Similar faults may recur, and this is detrimental to the overall improvement of the maintenance team's technical level. In summary, existing technologies, due to inherent defects in process collaboration, resource scheduling, and knowledge management, can no longer meet the requirements of modern coal mines for efficient, intelligent, and transparent emergency repair of equipment faults. Therefore, there is an urgent need for a solution that can integrate IoT monitoring, intelligent decision-making, and automated processes. Summary of the Invention

[0003] This application provides a method and system for emergency maintenance and scheduling of coal mine equipment based on dynamic workflow, in order to at least solve the technical problems of low efficiency in cross-departmental collaboration, unintelligent dynamic allocation of resources, and difficulty in reusing experience and knowledge.

[0004] The first aspect of this application proposes a method for emergency maintenance and scheduling of coal mine equipment faults based on dynamic workflow, the method comprising: The system collects real-time equipment operation status data by using sensors deployed on coal mine equipment and receives manually entered fault information via mobile terminals. Then, it determines whether to generate an electronic fault work order based on the equipment operation status data and the entered fault information. When a fault electronic work order is generated, the resource status matrix of each department is obtained, and similar historical cases of the fault electronic work order are matched in a pre-established maintenance fault case library based on the fault electronic work order to obtain the spare parts model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. Based on the spare parts model and maintenance plan corresponding to the fault electronic work order, the urgency level, and the resource status matrix of each department, a resource allocation plan and the shortest transportation path are determined, and the resource allocation plan and the shortest transportation path are executed to realize emergency maintenance and scheduling of coal mine equipment failures.

[0005] Preferably, the equipment operating status data includes: vibration signal sequence and temperature data sequence; The step of determining whether to generate an electronic fault work order based on the equipment operating status data and the entered fault information includes: The vibration signal sequence is subjected to discrete Fourier transform to obtain a frequency domain signal. Then, the vibration energy index of the coal mine equipment is determined based on the frequency domain signal. When the vibration energy index is greater than a preset energy threshold, an electronic fault work order is generated. And / or, Based on the temperature data sequence, determine whether it satisfies... and / or and / or If the conditions are met, an electronic work order for the fault is generated. Let be the temperature at time t. This is the upper limit of temperature. The lower limit of temperature, For the rate of temperature change, This is an outlier in the rate of change; And / or, When manually entered fault information is received, an electronic fault work order is generated.

[0006] Furthermore, obtaining the resource status matrix of each department includes: Access the resource status database to obtain the resource availability status of each department; A resource status matrix is ​​constructed based on the resource availability status of each department.

[0007] Furthermore, the matching of similar historical cases of the electronic fault work order in a pre-established maintenance fault case database based on the electronic fault work order includes: A method combining TF-IDF weighting with BERT context embedding vectors is used to generate feature vectors for fault description text; Based on the feature vector, an improved Levenshtein distance algorithm with attention weights is used to calculate the similarity between the current fault description and historical case texts in the case library. The historical case with the highest similarity in the case library is used as the similar historical case of the fault electronic work order; The cases include: spare parts model and repair plan, fault type and urgency level.

[0008] Furthermore, the step of determining the resource allocation plan and shortest transportation route based on the spare parts model and repair plan corresponding to the fault electronic work order, the fault type, the urgency level, and the resource status matrix of each department includes: The spare parts model, repair plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order are input into the pre-established resource allocation model to obtain the optimal resource allocation plan, wherein the repair plan includes the repair time. Based on the optimal resource allocation scheme, and using the Dijkstra algorithm to perform path planning on the underground roadway network, the shortest transportation path is obtained.

[0009] Furthermore, the process of establishing the resource allocation model includes: Construct an objective function with the goal of minimizing the total fault handling time; A resource allocation model is constructed by using spare parts inventory constraints and transportation capacity constraints as constraints, and combining them with the objective function.

[0010] Furthermore, the objective function is calculated as follows:

[0011] In the formula, This represents the total time for troubleshooting. As the first weight, As the second weight, As the third weight, For the time of material allocation, For transportation time, The repair time is defined as follows, where the values ​​of the first weight and the second weight are proportional to the degree of urgency. The formula for calculating the spare parts inventory constraint is as follows:

[0012] In the formula, For department i, the amount of material j requested. The current inventory of material j owned by department i; The formula for calculating the transport capacity constraint is as follows:

[0013] In the formula, For supplies weight, For the department The transport weight limit is specified, where m represents the type of goods.

[0014] Preferably, the method further includes: A fault knowledge graph is constructed based on a historical fault information image library. The nodes of the fault knowledge graph include: equipment model, fault phenomenon, solution, spare parts list and technical personnel, and the edges represent the relationship between nodes. When performing case retrieval, the knowledge graph is traversed and matched in multiple dimensions by combining cosine similarity calculation with breadth-first search (BFS) algorithm.

[0015] Furthermore, the method also includes: After the fault repair is completed, extract the work order information, fault symptoms, handling process and solution data of the current fault handling process; Natural language processing technology is used to parse, classify and label unstructured text data, and the processed structured and unstructured data are stored in the maintenance fault case library and the fault knowledge graph is updated.

[0016] The second aspect of this application proposes a coal mine equipment fault emergency repair and scheduling system based on dynamic workflow, including: The monitoring and reporting module is used to collect equipment operating status data in real time through sensors deployed on coal mine equipment and receive manually entered fault information through a mobile terminal. Then, based on the equipment operating status data and the entered fault information, it determines whether to generate an electronic fault work order. The analysis module is used to obtain the resource status matrix of each department when a fault electronic work order is generated, and to perform similar historical cases matching of the fault electronic work order in a pre-established maintenance fault case library based on the fault electronic work order, so as to obtain the spare parts model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. The workflow engine module is used to determine the resource allocation plan and the shortest transportation path based on the spare parts model and maintenance plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order, and to execute the resource allocation plan and the shortest transportation path to realize emergency maintenance scheduling of coal mine equipment failure.

[0017] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects: This application proposes a method and system for emergency maintenance and scheduling of coal mine equipment failures based on dynamic workflow. The method includes: collecting real-time equipment operating status data through sensors deployed on the coal mine equipment and receiving manually entered fault information through a mobile terminal; then determining whether to generate an electronic work order based on the equipment operating status data and the entered fault information; when an electronic work order is generated, obtaining the resource status matrix of each department, and matching similar historical cases of the electronic work order with a pre-established maintenance fault case library to obtain the spare parts model, maintenance plan, and urgency level corresponding to the electronic work order, wherein the case library is updated by continuously learning from historical work orders and maintenance reports; determining a resource allocation plan and the shortest transportation path based on the spare parts model, maintenance plan, urgency level, and resource status matrix of each department corresponding to the electronic work order, and executing the resource allocation plan and the shortest transportation path to realize emergency maintenance and scheduling of coal mine equipment failures. The technical solution proposed in this application solves the problems of low efficiency in cross-departmental collaboration, unintelligent resource allocation, and difficulty in reusing experience and knowledge in the existing technology, and realizes rapid, accurate and transparent emergency maintenance and scheduling for coal mine equipment failures.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of an emergency maintenance and scheduling method for coal mine equipment failure based on dynamic workflow, according to an embodiment of this application; Figure 2 This is a structural diagram of a coal mine equipment fault emergency repair and scheduling system based on dynamic workflow provided according to an embodiment of this application. Detailed Implementation

[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0021] This application proposes a method and system for emergency maintenance and scheduling of coal mine equipment faults based on dynamic workflow. The method includes: collecting real-time equipment operating status data through sensors deployed on the coal mine equipment and receiving manually entered fault information through a mobile terminal; then determining whether to generate an electronic work order based on the equipment operating status data and the entered fault information; when an electronic work order is generated, obtaining the resource status matrix of each department, and matching similar historical cases of the electronic work order with a pre-established maintenance fault case library to obtain the spare parts model, maintenance plan, and urgency level corresponding to the electronic work order, wherein the case library is updated by continuously learning from historical work orders and maintenance reports; determining a resource allocation plan and the shortest transportation path based on the spare parts model, maintenance plan, urgency level, and resource status matrix of each department corresponding to the electronic work order, and executing the resource allocation plan and the shortest transportation path to realize emergency maintenance and scheduling of coal mine equipment faults. The technical solution proposed in this application solves the problems of low efficiency in cross-departmental collaboration, unintelligent resource allocation, and difficulty in reusing experience and knowledge in the existing technology, and realizes rapid, accurate and transparent emergency maintenance and scheduling for coal mine equipment failures.

[0022] The following describes, with reference to the accompanying drawings, an embodiment of the coal mine equipment fault emergency repair and scheduling method and system based on dynamic workflow.

[0023] Example 1 Figure 1 The flowchart below shows a method for emergency maintenance and scheduling of coal mine equipment faults based on dynamic workflow, according to an embodiment of this application. Figure 1 As shown, the method includes: Step 1: Collect equipment operating status data in real time through sensors deployed on coal mine equipment and receive manually entered fault information through a mobile terminal. Then, determine whether to generate an electronic fault work order based on the equipment operating status data and the entered fault information. It should be noted that the equipment operating status data includes: vibration signal sequence and temperature data sequence.

[0024] In this embodiment of the disclosure, the step of determining whether to generate an electronic fault work order based on the equipment operating status data and the entered fault information includes: The vibration signal sequence is subjected to discrete Fourier transform to obtain a frequency domain signal. Then, the vibration energy index of the coal mine equipment is determined based on the frequency domain signal. When the vibration energy index is greater than a preset energy threshold, an electronic fault work order is generated. And / or, Based on the temperature data sequence, determine whether it satisfies... and / or and / or If the conditions are met, an electronic work order for the fault is generated. Let be the temperature at time t. This is the upper limit of temperature. The lower limit of temperature, For the rate of temperature change, This is an outlier in the rate of change; And / or, When manually entered fault information is received, an electronic fault work order is generated.

[0025] It should be noted that manually entered fault information includes: description of the fault phenomenon, on-site photos, voice notes, etc.

[0026] Specifically, vibration sensors, temperature sensors, and other IoT devices are deployed on coal mining equipment to collect real-time data on equipment operation status. Simultaneously, data acquisition terminals are provided to receive fault information entered by operators via mobile applications, including descriptions of fault phenomena, on-site photos, and voice annotations.

[0027] Data processing: Real-time analysis of data collected by sensors, using the following formulas and methods to identify equipment malfunctions.

[0028] Vibration spectrum analysis: Let the time-domain sequence of the vibration signal be... Through Discrete Fourier Transform Convert to frequency domain signal ; Define vibration energy index ,when When the preset energy threshold is reached, an abnormal vibration warning is triggered, and an electronic fault work order is generated.

[0029] Temperature threshold detection: Let the real-time temperature sequence be... The rate of temperature change The anomaly criteria are: When any of the conditions are met, a temperature abnormality work order, i.e., a fault electronic work order, is generated.

[0030] Step 2: When a fault electronic work order is generated, obtain the resource status matrix of each department, and match similar historical cases of the fault electronic work order in the pre-established maintenance fault case library based on the fault electronic work order to obtain the spare part model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. In this embodiment of the disclosure, obtaining the resource status matrix of each department includes: Access the resource status database to obtain the resource availability status of each department; A resource status matrix is ​​constructed based on the resource availability status of each department.

[0031] It should be noted that in the automatic trigger mode: when the IoT monitoring module detects a device malfunction and generates an electronic work order, the work order is automatically routed to the dispatch room, warehousing center, transportation team, electromechanical department, and other relevant departments in sequence. After receiving the work order, the dispatch room automatically accesses the resource status database to obtain information such as the spare parts inventory of the warehousing center, the vehicle scheduling status of the transportation team, and the availability status of technical personnel in the electromechanical department.

[0032] Manual Intervention Mode: In cases where the system cannot monitor in real time due to special environmental or technical limitations, a manual intervention mechanism is activated. On-site material handlers assess the availability of spare parts. If spare parts are sufficient, they immediately replace them and report to the dispatch center via a mobile application. The report information is synchronized to the system backend in real time. If spare parts are in short supply, the dispatch center, acting as the central hub, uses a workflow engine to digitally model the business processes of each department, breaking down departmental information system barriers and achieving cross-system resource coordination and scheduling.

[0033] The specific steps are as follows: trigger a spare parts allocation work order, determine the required departments and quantities of materials based on spare parts demand; and link the interface mechanism between the warehousing and transportation departments to determine the transportation vehicles, drivers, and transportation routes.

[0034] Dynamic resource allocation and progress tracking: The dispatching room uses big data analysis technology to construct a dynamic resource allocation model (which refers to the objective function and corresponding constraints described below) based on the fault types (determined by NLP semantic classification), urgency levels (divided into three levels: urgent, important, and general) and resource status matrices of each department, according to the fault types reported on site.

[0035] Resource Status Matrix: Let the department set be... Resource type set The resource state matrix S is represented as:

[0036] in Indicates department Resources The availability status (such as the quantity of spare parts in stock, the availability of technicians, etc.).

[0037] Furthermore, the matching of similar historical cases of the electronic fault work order in a pre-established maintenance fault case database based on the electronic fault work order includes: A method combining TF-IDF weighting with BERT context embedding vectors is used to generate feature vectors for fault description text; Based on the feature vector, an improved Levenshtein distance algorithm with attention weights is used to calculate the similarity between the current fault description and historical case texts in the case library. The historical case with the highest similarity in the case library is used as the similar historical case of the fault electronic work order; The cases include: spare parts model and repair plan, fault type and urgency level.

[0038] It should be noted that a troubleshooting case library is established based on artificial intelligence-based Natural Language Processing (NLP) and machine learning technologies. After each troubleshooting operation, the system automatically processes data such as work order information, fault symptoms, handling process, technical measures, problems, and solutions. Structured data extraction: Using regular expressions and OCR recognition technology, structured data such as equipment model, fault code, spare parts model, and maintenance time are extracted from work orders and maintenance reports.

[0039] Unstructured data processing: The BERT model is used to perform semantic parsing on maintenance record texts, extract unstructured data such as key technical points and fault cause analysis, and then classify and label them.

[0040] Case matching engine: When a new fault is encountered, the maintenance personnel input relevant information, and the system uses deep learning algorithms to perform semantic matching and similarity analysis in the case library.

[0041] Feature vector generation: TF-IDF weighted + context embedding representation method is adopted.

[0042] Term Frequency-Inverse Document Frequency (TF-IDF): Term In the document Weights in:

[0043] TF-IDF weights are fused with context embedding vectors to generate feature vectors for fault descriptions and case texts, providing a foundation for subsequent similarity calculations. Based on the generated feature vectors, a similarity calculation algorithm quickly finds the most similar historical cases to the current fault, providing maintenance personnel with reference solutions.

[0044] in For terms In the document The number of times it appears in This represents the total number of documents in the case study library.

[0045] Contextual embedding: Generating word vectors using the BERT model It integrates semantic information at the character and word levels.

[0046] Similarity calculation: Combining an improved Levenshtein distance algorithm with an attention mechanism. Let the fault description text... Case text Improved Levenshtein distance matrix Defined as: Introducing attention weights Prioritize matching of key terms (such as equipment model and fault codes), and calculate the final similarity score: .

[0047] Step 3: Based on the spare parts model and maintenance plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order, determine the resource allocation plan and the shortest transportation route, and execute the resource allocation plan and the shortest transportation route to realize emergency maintenance and scheduling of coal mine equipment faults.

[0048] In this embodiment of the disclosure, determining the resource allocation plan and the shortest transportation route based on the spare parts model and repair plan corresponding to the fault electronic work order, the fault type, the urgency level, and the resource status matrix of each department includes: The spare parts model, repair plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order are input into the pre-established resource allocation model to obtain the optimal resource allocation plan, wherein the repair plan includes the repair time. Based on the optimal resource allocation scheme, and using the Dijkstra algorithm to perform path planning on the underground roadway network, the shortest transportation path is obtained.

[0049] It should be noted that the process of establishing the resource allocation model includes: Construct an objective function with the goal of minimizing the total fault handling time; A resource allocation model is constructed by using spare parts inventory constraints and transportation capacity constraints as constraints, and combining them with the objective function.

[0050] The objective function is calculated as follows:

[0051] In the formula, This represents the total time for troubleshooting. As the first weight, As the second weight, As the third weight, For the time of material allocation, For transportation time, The repair time is defined as follows, where the values ​​of the first weight and the second weight are proportional to the degree of urgency. The formula for calculating the spare parts inventory constraint is as follows:

[0052] In the formula, For department i, the amount of material j requested. The current inventory of material j owned by department i; The formula for calculating the transport capacity constraint is as follows:

[0053] In the formula, For supplies weight, For the department The transport weight limit is specified, where m represents the type of goods.

[0054] It should be noted that the resource allocation priority optimization (linear programming model) is as follows: Objective function: Minimize the total fault handling time (including material allocation time) Transportation time Repair time ) ; The weight values ​​are not fixed. On the one hand, w1 (material allocation time weight) and w2 (transportation time weight) in the objective function will be dynamically adjusted according to the urgency of the fault. The higher the urgency (when Ek=1), the greater the weights of w1 and w2. On the other hand, the weight w(e) (travel time) of edge E in the transportation path optimization needs to be corrected in combination with the slope and width constraints of the underground roadway. It will change according to the actual roadway environment and is not a fixed value. The core of the objective function and constraints is to solve for the relevant data of the optimal resource allocation scheme. Constraints: Spare parts inventory constraints: (department For supplies (Inventory quantity) x ij This indicates the amount of material j requested by department i. This indicates the current inventory of material j held by department i. The overall meaning is that "department i's use of material j shall not exceed its current inventory." This is to avoid material use exceeding actual reserve capacity and to ensure the feasibility of resource allocation.

[0055] Transportation capacity constraints: ; Urgency Priority: (The higher the urgency, the more important the objective function is) (The larger the weight) Transportation route optimization (considering underground roadway constraints): using a weighted graph Modeling the tunnel network, nodes The intersection of the alleyways, side weight This includes travel time (including corrections for slope and width constraints). The shortest path problem is transformed into a Dijkstra's algorithm for solution: ,in From the starting point to the node The shortest time, It is the set of predecessor nodes.

[0056] Transportation route optimization (using Dijkstra's algorithm): The goal is to calculate the shortest transportation time t. t , is the t in the objective function t The value of is provided with precise basis.

[0057] This model employs linear programming algorithms to optimize material allocation priorities, transportation route optimization schemes, and technical personnel scheduling strategies. Simultaneously, a monitoring system built with a microservice architecture continuously tracks the progress of equipment fault repair processes in real time, accurately recording timestamps and operational information for each step through distributed tracing technology. Once delays or obstacles are detected, the system automatically generates early warning information, quickly locating the problem node through intelligent algorithms, allowing the dispatch center to promptly coordinate with relevant departments for resolution.

[0058] In this embodiment of the disclosure, the method further includes: A fault knowledge graph is constructed based on a historical fault information image library. The nodes of the fault knowledge graph include: equipment model, fault phenomenon, solution, spare parts list and technical personnel, and the edges represent the relationship between nodes. When performing case retrieval, the knowledge graph is traversed and matched in multiple dimensions by combining cosine similarity calculation with breadth-first search (BFS) algorithm.

[0059] It should be noted that the knowledge graph is constructed based on a graph database. Nodes include equipment model, fault symptoms, solutions, spare parts lists, and technical personnel. Relationships between nodes include "equipment model - fault symptoms," "fault symptoms - solutions," "solutions - spare parts lists," and "technical personnel - solutions," etc. Multi-dimensional retrieval of similar cases is supported, and retrieval efficiency is improved through cosine similarity and graph traversal algorithms.

[0060] Cosine similarity: node feature vectors Cosine similarity: ; Combining breadth-first search (BFS) to traverse the knowledge graph, matching cases are returned in order of similarity, resulting in a retrieval time complexity of O(n log n). ( For the number of nodes, (Number of edges). The knowledge graph update mechanism is as follows: when a new case is added, information such as equipment model, fault phenomenon, solution, spare parts list, and technical personnel is extracted and used as a new node or to update the information of existing nodes; based on the logical relationship in the case, edges between nodes are established or updated to ensure that the knowledge graph can reflect the latest fault handling experience in a timely manner.

[0061] Furthermore, the method also includes: After the fault repair is completed, extract the work order information, fault symptoms, handling process and solution data of the current fault handling process; Natural language processing technology is used to parse, classify and label unstructured text data, and the processed structured and unstructured data are stored in the maintenance fault case library and the fault knowledge graph is updated.

[0062] Furthermore, the method also includes: In the manual intervention mode, on-site personnel assess the availability of spare parts via mobile terminals. If spare parts are sufficient, they are replaced directly and reported. If spare parts are in short supply, a spare parts allocation request is triggered via mobile terminals. In response to the request, the dispatch center generates a spare parts allocation work order and drives the workflow engine to coordinate the warehousing and transportation departments to execute the allocation of materials and transportation capacity.

[0063] Furthermore, the method also includes: During the execution of dynamic workflows, the progress and status of each stage of fault handling are monitored in real time through microservice architecture and distributed tracing technology. When a delay or resource conflict is detected, an early warning message is automatically generated and sent to the dispatch room for manual intervention and coordination.

[0064] The emergency maintenance and scheduling method for coal mine equipment failure based on dynamic workflow proposed in this embodiment has the following advantages: 1. Cross-departmental collaboration efficiency has been significantly improved. By constructing a dynamic workflow system, the responsibilities and tasks of each department in the emergency repair and scheduling of equipment failures were clearly defined, and every step from failure discovery and reporting to handling was standardized and regulated. In areas such as spare parts assessment, scheduling coordination, and material supply, each department has clear work standards and process guidelines, effectively eliminating the previous phenomena of mutual buck-passing and work delays caused by unclear responsibilities. Actual test data shows that the average failure response time was reduced from 4.2 hours in the traditional method to 1.8 hours, the number of cross-departmental communications decreased by 60%, and the accuracy and reliability of the work were significantly improved.

[0065] 2. Resource dynamic allocation is more intelligent and efficient. By leveraging big data analytics and dynamic resource allocation models, real-time dynamic optimization and allocation of resources such as warehousing materials, transportation capacity, and electromechanical technical support were achieved. Through multi-dimensional analysis of fault type, urgency, and resource status, resource priority and allocation plans were rationally determined, resulting in a 39% reduction in resource idle rate and a 28% increase in warehouse turnover rate, effectively reducing resource waste and improving resource utilization efficiency. Simultaneously, optimized transportation routes and rational scheduling of technical personnel shortened spare parts delivery time and fault repair time, further improving fault handling efficiency.

[0066] 3. The ability to reuse experiential knowledge has been greatly enhanced. A comprehensive troubleshooting case library has been established, enabling the automatic extraction, classification, and annotation of troubleshooting experience through natural language processing and machine learning technologies. As the case library continues to accumulate, a rich system of experiential knowledge has been formed. When encountering similar faults, the system can quickly retrieve relevant solutions, reducing the average troubleshooting time by 40% and increasing the repair success rate by 25%. Furthermore, the case library provides vivid and practical training materials for new employees, promoting knowledge transfer and improving the overall technical level of the team.

[0067] In summary, the emergency maintenance and scheduling method for coal mine equipment failure based on dynamic workflow proposed in this embodiment realizes rapid, accurate, and transparent emergency maintenance and scheduling for coal mine equipment failure.

[0068] Example 2 Figure 2 This is a structural diagram of a coal mine equipment fault emergency repair and scheduling system based on dynamic workflow according to an embodiment of this application, as shown below. Figure 2 As shown, the system includes: The monitoring and reporting module 100 is used to collect equipment operating status data in real time through sensors deployed on coal mine equipment and receive manually entered fault information through a mobile terminal. Then, based on the equipment operating status data and the entered fault information, it determines whether to generate an electronic fault work order. The equipment operating status data includes: vibration signal sequence and temperature data sequence.

[0069] The analysis module 200 is used to obtain the resource status matrix of each department when a fault electronic work order is generated, and to perform similar historical cases matching of the fault electronic work order in a pre-established maintenance fault case library based on the fault electronic work order, so as to obtain the spare parts model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. The workflow engine module 300 is used to determine the resource allocation plan and the shortest transportation path based on the spare parts model and maintenance plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order, and execute the resource allocation plan and the shortest transportation path to realize emergency maintenance scheduling of coal mine equipment failure.

[0070] In this embodiment of the disclosure, the monitoring and reporting module 100 is further configured to: The vibration signal sequence is subjected to discrete Fourier transform to obtain a frequency domain signal. Then, the vibration energy index of the coal mine equipment is determined based on the frequency domain signal. When the vibration energy index is greater than a preset energy threshold, an electronic fault work order is generated. And / or, Based on the temperature data sequence, determine whether it satisfies... and / or and / or If the conditions are met, an electronic work order for the fault is generated. Let be the temperature at time t. This is the upper limit of temperature. The lower limit of temperature, For the rate of temperature change, This is an outlier in the rate of change; And / or, When manually entered fault information is received, an electronic fault work order is generated.

[0071] In this embodiment of the disclosure, the analysis module 200 is further configured to: Access the resource status database to obtain the resource availability status of each department; A resource status matrix is ​​constructed based on the resource availability status of each department.

[0072] In this embodiment of the disclosure, the analysis module 200 is further configured to: A method combining TF-IDF weighting with BERT context embedding vectors is used to generate feature vectors for fault description text; Based on the feature vector, an improved Levenshtein distance algorithm with attention weights is used to calculate the similarity between the current fault description and historical case texts in the case library. The historical case with the highest similarity in the case library is used as the similar historical case of the fault electronic work order; The cases include: spare parts model and repair plan, fault type and urgency level.

[0073] In this embodiment of the disclosure, the workflow engine module 300 is further configured to: The spare parts model, repair plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order are input into the pre-established resource allocation model to obtain the optimal resource allocation plan, wherein the repair plan includes the repair time. Based on the optimal resource allocation scheme, and using the Dijkstra algorithm to perform path planning on the underground roadway network, the shortest transportation path is obtained.

[0074] It should be noted that the process of establishing the resource allocation model includes: Construct an objective function with the goal of minimizing the total fault handling time; A resource allocation model is constructed by using spare parts inventory constraints and transportation capacity constraints as constraints, and combining them with the objective function.

[0075] The objective function is calculated as follows:

[0076] In the formula, This represents the total time for troubleshooting. As the first weight, As the second weight, As the third weight, For the time of material allocation, For transportation time, The repair time is defined as follows, where the values ​​of the first weight and the second weight are proportional to the degree of urgency. The formula for calculating the spare parts inventory constraint is as follows:

[0077] In the formula, For department i, the amount of material j requested. The current inventory of material j owned by department i; The formula for calculating the transport capacity constraint is as follows:

[0078] In the formula, For supplies weight, For the department The transport weight limit is specified, where m represents the type of goods.

[0079] In this embodiment of the disclosure, the analysis module 200 is further configured to: A fault knowledge graph is constructed based on a historical fault information image library. The nodes of the fault knowledge graph include: equipment model, fault phenomenon, solution, spare parts list and technical personnel, and the edges represent the relationship between nodes. When performing case retrieval, the knowledge graph is traversed and matched in multiple dimensions by combining cosine similarity calculation with breadth-first search (BFS) algorithm.

[0080] In this embodiment of the disclosure, the analysis module 200 is further configured to: After the fault repair is completed, extract the work order information, fault symptoms, handling process and solution data of the current fault handling process; Natural language processing technology is used to parse, classify and label unstructured text data, and the processed structured and unstructured data are stored in the maintenance fault case library and the fault knowledge graph is updated.

[0081] In summary, the coal mine equipment fault emergency repair and scheduling system based on dynamic workflow proposed in this embodiment realizes rapid, accurate, and transparent emergency repair and scheduling for coal mine equipment faults.

[0082] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0083] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0084] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for emergency maintenance and scheduling of coal mine equipment based on dynamic workflow, characterized in that, The method includes: The system collects real-time equipment operation status data by using sensors deployed on coal mine equipment and receives manually entered fault information via mobile terminals. Then, it determines whether to generate an electronic fault work order based on the equipment operation status data and the entered fault information. When a fault electronic work order is generated, the resource status matrix of each department is obtained, and similar historical cases of the fault electronic work order are matched in a pre-established maintenance fault case library based on the fault electronic work order to obtain the spare parts model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. Based on the spare parts model and maintenance plan corresponding to the fault electronic work order, the urgency level, and the resource status matrix of each department, a resource allocation plan and the shortest transportation path are determined, and the resource allocation plan and the shortest transportation path are executed to realize emergency maintenance and scheduling of coal mine equipment failures.

2. The method as described in claim 1, characterized in that, The equipment operating status data includes: vibration signal sequence and temperature data sequence; The step of determining whether to generate an electronic fault work order based on the equipment operating status data and the entered fault information includes: The vibration signal sequence is subjected to discrete Fourier transform to obtain a frequency domain signal. Then, the vibration energy index of the coal mine equipment is determined based on the frequency domain signal. When the vibration energy index is greater than a preset energy threshold, an electronic fault work order is generated. And / or, Based on the temperature data sequence, determine whether it satisfies... and / or and / or If the conditions are met, an electronic work order for the fault is generated. Let be the temperature at time t. This is the upper limit of temperature. The lower limit of temperature, For the rate of temperature change, This is an outlier in the rate of change; And / or, When manually entered fault information is received, an electronic fault work order is generated.

3. The method as described in claim 2, characterized in that, The process of obtaining the resource status matrix of each department includes: Access the resource status database to obtain the resource availability status of each department; A resource status matrix is ​​constructed based on the resource availability status of each department.

4. The method as described in claim 3, characterized in that, The process of matching similar historical cases of the electronic fault work order in a pre-established maintenance fault case database includes: A method combining TF-IDF weighting with BERT context embedding vectors is used to generate feature vectors for fault description text; Based on the feature vector, an improved Levenshtein distance algorithm with attention weights is used to calculate the similarity between the current fault description and historical case texts in the case library. The historical case with the highest similarity in the case library is used as the similar historical case of the fault electronic work order; The cases include: spare parts model and repair plan, fault type and urgency level.

5. The method as described in claim 4, characterized in that, The process of determining the resource allocation plan and shortest transportation route based on the spare parts model and repair plan corresponding to the fault electronic work order, the fault type, the urgency level, and the resource status matrix of each department includes: The spare parts model, repair plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order are input into the pre-established resource allocation model to obtain the optimal resource allocation plan, wherein the repair plan includes the repair time. Based on the optimal resource allocation scheme, and using the Dijkstra algorithm to perform path planning on the underground roadway network, the shortest transportation path is obtained.

6. The method as described in claim 5, characterized in that, The process of establishing the resource allocation model includes: Construct an objective function with the goal of minimizing the total fault handling time; A resource allocation model is constructed by using spare parts inventory constraints and transportation capacity constraints as constraints, and combining them with the objective function.

7. The method as described in claim 6, characterized in that, The objective function is calculated as follows: In the formula, This represents the total time for troubleshooting. As the first weight, As the second weight, As the third weight, For the time of material allocation, For transportation time, The repair time is defined as follows, where the values ​​of the first weight and the second weight are proportional to the degree of urgency. The formula for calculating the spare parts inventory constraint is as follows: In the formula, For department i, the amount of material j requested. The current inventory of material j owned by department i; The formula for calculating the transport capacity constraint is as follows: In the formula, For supplies weight, For the department The transport weight limit is specified, where m represents the type of goods.

8. The method as described in claim 1, characterized in that, The method further includes: A fault knowledge graph is constructed based on a historical fault information image library. The nodes of the fault knowledge graph include: equipment model, fault phenomenon, solution, spare parts list and technical personnel, and the edges represent the relationship between nodes. When performing case retrieval, the knowledge graph is traversed and matched in multiple dimensions by combining cosine similarity calculation with breadth-first search (BFS) algorithm.

9. The method as described in claim 8, characterized in that, The method further includes: After the fault repair is completed, extract the work order information, fault symptoms, handling process and solution data of the current fault handling process; Natural language processing technology is used to parse, classify and label unstructured text data, and the processed structured and unstructured data are stored in the maintenance fault case library and the fault knowledge graph is updated.

10. A coal mine equipment fault emergency repair and scheduling system based on dynamic workflow, characterized in that, The system includes: The monitoring and reporting module is used to collect equipment operating status data in real time through sensors deployed on coal mine equipment and receive manually entered fault information through a mobile terminal. Then, based on the equipment operating status data and the entered fault information, it determines whether to generate an electronic fault work order. The analysis module is used to obtain the resource status matrix of each department when a fault electronic work order is generated, and to perform similar historical cases matching of the fault electronic work order in a pre-established maintenance fault case library based on the fault electronic work order, so as to obtain the spare parts model, maintenance plan and urgency level corresponding to the fault electronic work order. The case library is updated by continuously learning historical work orders and maintenance reports. The workflow engine module is used to determine the resource allocation plan and the shortest transportation path based on the spare parts model and maintenance plan, urgency level, and resource status matrix of each department corresponding to the fault electronic work order, and to execute the resource allocation plan and the shortest transportation path to realize emergency maintenance scheduling of coal mine equipment failure.

Images