Thermal power overhaul platform integrating equipment information, overhaul standard and GIS (Geographic Information System) dynamic grid

Abstract

The invention relates to the technical field of thermal power plant equipment maintenance, and discloses a thermal power overhaul platform fusing equipment information, overhaul standards and GIS dynamic grids, the platform is integrated with an equipment information acquisition module, and the equipment state, operation parameters and appearance images are acquired in real time; accurate positioning and space management of equipment are realized through a GIS dynamic grid construction module; the maintenance standard matching module automatically matches maintenance standards according to equipment characteristics and generates a task list; the data analysis module deeply analyzes data, predicts faults and preferentially sorts maintenance tasks; the maintenance plan generation module automatically generates a high-efficiency maintenance plan according to the information; a supervisory feedback module ensures plan execution and facilitates continuous improvements. According to the platform, through intelligent and refined management, the efficiency and quality of equipment maintenance of the thermal power plant are remarkably improved, the stability and safety of power supply are guaranteed, and the platform has high practical value and application prospects.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power plant equipment maintenance, and in particular to a thermal power plant overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid. Background Art

[0002] In the field of thermal power generation, thermal power plants are the key link in energy supply, and the stable operation of their equipment is crucial to ensuring power supply, improving economic benefits and ensuring safe production. However, with the continuous increase in the capacity of thermal power units and the increase in equipment complexity, the traditional equipment maintenance model has gradually exposed many shortcomings.

[0003] At present, equipment monitoring in thermal power plants mostly relies on regular inspections, which is not only time-consuming and labor-intensive, but also difficult to achieve real-time and comprehensive monitoring of the equipment's operating status. The omission or delayed recording of equipment status data, operating parameters, and appearance damage often leads to inaccurate assessments of the equipment's health status. Passive maintenance based on fixed time periods or after a failure lacks the flexibility to dynamically adjust according to the actual working status of the equipment. The formulation of maintenance standards fails to fully consider the individual differences of equipment (such as service life, operating environment, load characteristics, etc.), resulting in excessive or insufficient maintenance, which wastes resources and fails to effectively prevent failures.

[0004] Moreover, the equipment in thermal power plants is widely distributed, the spatial distribution of maintenance tasks is complex, and there is a lack of effective space management tools to optimize maintenance paths and resource allocation. This not only increases maintenance time, but may also result in key equipment failures not being handled in a timely manner, affecting the overall operating efficiency of the power plant. Summary of the invention

[0005] The purpose of the present invention is to provide a thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grids to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid, comprising:

[0007] The equipment information collection module includes an equipment status sensor unit, an operating parameter recording unit, and an image recognition unit, which is used to collect the working status data, operating parameters, and equipment appearance images of various equipment in the thermal power plant in real time;

[0008] GIS dynamic grid construction module is used to divide the thermal power plant area into multiple dynamic grids and integrate with the geographic information system to spatially locate the equipment. Each grid contains a corresponding set of equipment.

[0009] The maintenance standard matching module has a built-in maintenance standard database, which is used to automatically match the corresponding maintenance standards and cycles according to the equipment type, service life and operating status, and generate a maintenance task list;

[0010] The data analysis module includes an equipment operation status analysis unit, a fault prediction unit, and a maintenance priority sorting unit. This module uses the data from the equipment information acquisition module and combines it with GIS dynamic grid information to analyze abnormal changes in the equipment operation status, predict potential fault points, and prioritize maintenance tasks based on the output of the maintenance standard matching module.

[0011] The maintenance plan generation module automatically generates a maintenance plan based on the output results of the data analysis module, including maintenance time, required resources, personnel allocation and safety measures;

[0012] The supervision and feedback module is used to monitor the implementation of the maintenance plan, collect feedback data from on-site maintenance personnel, evaluate the maintenance quality, and adjust subsequent maintenance plans and standards accordingly to form a closed-loop continuous improvement mechanism.

[0013] Preferably, the GIS dynamic grid construction module dynamically adjusts the grid division and the positioning of the equipment by calculating the adjustment parameters of each grid unit, including:

[0014] Dynamically update the content of grid cells and device collections based on the addition, removal or location change of devices;

[0015] Dynamically adjust the density and size of the grid based on the urgency and priority of equipment maintenance.

[0016] Preferably, the step of generating a maintenance task list by the maintenance standard matching module includes:

[0017] Step 1: Obtain equipment characteristics from the equipment information acquisition module, wherein the equipment characteristics include equipment type, service life, operating parameter history records and current status data;

[0018] Step 2: Based on the acquired equipment features, a fuzzy matching algorithm is used to compare and analyze the preset standards in the maintenance standard database. The specific algorithm is:

[0019] Assume that the device feature vector D = (T, Y, P1, P2, ..., P n ), where T is the equipment type, Y is the service life, P1, P2, ..., P n Each standard record in the maintenance standard database is S i =(T i , Y i , R i1 , R i2 , …, Rin , C i , M i ) where T i is the specific device type, Y i R is the applicable working years range. i1 , R i2 , …, R in is the standard range or threshold of each operating parameter, C i is the maintenance cycle, M i is the corresponding maintenance method or standard; for each device D, calculate its comparison with each standard S in the database i The similarity Sim(D, S i ):

[0020]

[0021] Among them, α, β, ω j is the weight coefficient, satisfying

[0022] δ(T,T i ) is the matching degree of the device type. If T = T i Then δ=1, otherwise δ=0;

[0023] γ(Y,Y i ) is the matching function of working years. If Y is in Y i If it is within the range, then γ=1, otherwise γ=0;

[0024] η(P j , R ij ) is the matching function of the running parameters. If P j In R ij If the deviation is within the range, then η = 1, otherwise a value between 0 and 1 is assigned according to the degree of deviation;

[0025] Step 3: Select the similarity Sim(D, S i ) The highest standard record is taken as the matching result, and the maintenance cycle C is extracted from it. i and maintenance method M i ;

[0026] Step 4: Based on the matching results, a maintenance task list including equipment identification, recommended maintenance cycle, specific maintenance method and time is automatically generated.

[0027] Preferably, the equipment operation status analysis unit obtains equipment operation parameters and real-time working status data collected by the equipment information acquisition module, and compares the obtained data with the normal state range to monitor in real time whether the equipment is in an abnormal state.

[0028] Preferably, the fault prediction unit establishes a fault mode library in combination with equipment types and known fault cases, and compares real-time data with patterns in the fault mode library through a pattern matching algorithm to identify potential fault types and fault points.

[0029] Preferably, the pattern matching algorithm adopts a decision tree algorithm.

[0030] Preferably, the fault prediction unit compares the real-time data with the patterns in the fault pattern library through a decision tree algorithm to identify potential fault types and fault points, and the implementation steps include:

[0031] S1: Obtain the real-time operation data of the equipment from the equipment information collection module, and select the characteristic parameters closely related to the failure mode in combination with the equipment type and known failure cases to construct the failure feature vector;

[0032] S2: Use historical fault data and corresponding fault modes as training sample sets; use decision tree algorithm to learn the training sample sets and build a decision tree model; the decision tree model consists of nodes and edges, each node represents a test of a feature attribute, and each edge represents a branch corresponding to the test result;

[0033] S3: Input the fault feature vector constructed in real time into the decision tree model, starting from the root node, traverse the nodes and edges of the decision tree in sequence according to the test results of the feature attributes; at each node, select the corresponding edge according to the value of the feature attribute until reaching the leaf node; the category represented by the leaf node is the identified fault mode.

[0034] Preferably, when constructing a decision tree model, the optimal split feature is selected by a Gini index algorithm, and the calculation formula of the Gini index algorithm is:

[0035]

[0036] Among them, k represents the number of categories, p i Represents the probability of category i appearing in the data set Data.

[0037] Preferably, it also includes:

[0038] The mobile maintenance module can quickly obtain equipment information, maintenance standards and historical maintenance records by scanning the equipment QR code or GPS positioning, assist in completing maintenance tasks, and upload maintenance progress and problems found in real time.

[0039] Preferably, when formulating the maintenance plan, the maintenance plan generation module optimizes the sequencing and time scheduling of maintenance tasks with the goal of minimizing maintenance time and maximizing resource utilization.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] Through the equipment information collection module, the working status data, operating parameters and appearance images of the equipment are obtained in real time, which effectively avoids the omission and delay of information in traditional periodic inspections, ensures the accuracy and timeliness of equipment health assessment, and not only improves the level of refinement of equipment management, but also provides a solid foundation for subsequent data analysis and maintenance decisions. The maintenance standard matching module automatically matches the most appropriate maintenance standards and cycles from the built-in maintenance standard database according to the type, service life and operating status of the equipment, avoids excessive or insufficient maintenance, and realizes the optimal allocation of resources. This personalized maintenance strategy not only improves maintenance efficiency, but also reduces maintenance costs.

[0042] The GIS dynamic grid construction module divides the thermal power plant area into multiple dynamic grids and integrates with the geographic information system to achieve accurate spatial positioning of equipment, which helps optimize the maintenance path, quickly respond to equipment failures, reduce maintenance time, and improve the efficiency of resource (such as manpower and materials) deployment, ensuring the overall operation efficiency of the power plant. The data analysis module can deeply mine equipment data, analyze abnormal changes in equipment operating status, predict potential fault points, and prioritize maintenance tasks according to the output of the maintenance standard matching module, providing a basis for the formulation of maintenance plans and ensuring the pertinence and effectiveness of maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a working principle diagram of the present invention;

[0044] Figure 2 Step diagrams generated for maintenance task lists;

[0045] Figure 3 Flowchart for construction of a failure mode library using a decision algorithm. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] See also Figure 1-3 The present invention provides a technical solution: a thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid, including:

[0048] The equipment information acquisition module includes an equipment status sensor unit, an operation parameter recording unit and an image recognition unit.

[0049] Equipment status sensor unit: Various sensors, such as temperature sensors, pressure sensors, vibration sensors, etc., are installed on key equipment in thermal power plants. These sensors collect the working status data of the equipment in real time and transmit it to the data concentrator.

[0050] Operation parameter recording unit: connected to the DCS (distributed control system) or SCADA (supervisory control and data acquisition) system of the thermal power plant to obtain and record the operating parameters of the equipment in real time, including but not limited to current, voltage, speed, flow, etc.

[0051] Image recognition unit: Install high-definition cameras at key locations on the equipment, use image recognition technology to regularly take photos of the equipment's appearance, and automatically identify anomalies in the images, such as cracks, corrosion, oil leaks, etc.

[0052] The GIS dynamic grid construction module uses GIS technology to divide the thermal power plant area into multiple dynamic grids. The size and shape of each grid can be dynamically adjusted according to the equipment distribution density and importance. The equipment location information obtained by the equipment information collection module is integrated with the GIS system to achieve the spatial positioning of the equipment. Each grid contains a group of equipment, which are displayed as icons or labels on the GIS map. The grid information can be dynamically updated as the equipment moves or changes in status to ensure the accuracy and real-time nature of the equipment spatial information.

[0053] The maintenance standard matching module establishes a maintenance standard database, which contains maintenance standards and cycles for different equipment types, service life and operating conditions. When new equipment is added or the equipment status changes, the system automatically matches the corresponding maintenance standards and cycles in the database based on the equipment information and generates a maintenance task list. The maintenance task list includes maintenance items, recommended maintenance time, required maintenance tools and materials, etc., which facilitates maintenance personnel to prepare in advance.

[0054] The data analysis module includes an equipment operation status analysis unit, a fault prediction unit and a maintenance priority sorting unit.

[0055] Equipment operation status analysis unit: Use the data provided by the equipment information acquisition module to conduct real-time analysis of the equipment operation status and discover abnormal changes or trends.

[0056] Fault prediction unit: Based on historical data and machine learning algorithms, it predicts the potential failure points and failure times of equipment, providing a basis for preventive maintenance.

[0057] Maintenance Prioritization Unit: Prioritizes maintenance tasks based on the importance of the equipment, risk of failure, and maintenance cycle to ensure that critical equipment is handled first.

[0058] The maintenance plan generation module automatically generates a maintenance plan based on the output results of the data analysis module, including maintenance time, required resources (such as personnel, tools, materials), personnel allocation and safety measures, etc. The maintenance plan can be adjusted and optimized according to actual conditions to ensure the smooth progress of maintenance work.

[0059] The supervision feedback module monitors the execution of the maintenance plan in real time, including maintenance progress, personnel presence, implementation of safety measures, etc. It collects feedback data from on-site maintenance personnel, including problems found during the maintenance process, solutions, maintenance quality, etc. It evaluates the maintenance quality and adjusts subsequent maintenance plans and standards based on the evaluation results to form a closed-loop continuous improvement mechanism.

[0060] The mobile maintenance module can quickly obtain equipment information, maintenance standards and historical maintenance records by scanning the equipment QR code or GPS positioning, assist in completing maintenance tasks, and upload maintenance progress and problems found in real time.

[0061] The core function of the GIS dynamic grid construction module is to dynamically build and adjust the GIS grid according to the distribution, changes and maintenance requirements of the equipment. The specific implementation steps are as follows:

[0062] ①Dynamically update grid cells

[0063] The system first continuously monitors the addition, removal or location change of equipment through the equipment information collection module. For example, when a new transformer is added to an area, the system automatically captures the change. Based on the equipment change information, the GIS dynamic grid construction module instantly updates the affected grid cells and incorporates the new equipment into the equipment collection of the corresponding grid. If the equipment is removed, it is deleted from the equipment list of the grid to which it belongs. For minor adjustments to the location of the equipment, the system recalculates the grid to which it belongs and updates the grid boundaries to ensure that each device is accurately mapped to the corresponding grid cell.

[0064] The system dynamically adjusts the grid density according to the urgency of equipment maintenance and preset priorities (such as critical equipment, equipment with frequent failures, etc.). For example, if a grid contains multiple critical equipment that urgently needs maintenance, the system will increase the grid density in that area to manage resources more finely. Based on the evaluation results, the system automatically adjusts the grid size. For areas that require high attention, the grid division is more detailed to more accurately locate equipment, optimize maintenance paths, and reduce response time. For areas with sparse equipment or less important, the grid size is appropriately expanded to reduce management complexity.

[0065] The equipment operation status analysis unit is responsible for real-time analysis of the equipment operation status and timely detection of abnormalities. The specific process is as follows:

[0066] Data acquisition: The equipment operation status analysis unit obtains equipment operation parameters (such as voltage, current, temperature, etc.) and real-time working status data from the equipment information acquisition module.

[0067] Status comparison: Compare the collected data with the preset normal status range. These data ranges are determined based on the specifications and historical operating data provided by the equipment manufacturer to ensure the accuracy of the analysis.

[0068] Abnormal detection: If any parameter is found to be outside the normal range, the system will immediately mark the device as abnormal and trigger the alarm mechanism. For example, if the temperature of a transformer continues to rise above the warning value, the system will automatically identify and report this abnormal situation.

[0069] Fault diagnosis and prediction: In addition to simple anomaly detection, the system also uses machine learning algorithms to analyze historical fault data, predict potential faults, and provide decision support for preventive maintenance.

[0070] Assume that in an industrial park, the GIS dynamic grid construction module initially divides several grids according to the distribution of equipment. One day, due to production needs, several high-voltage distribution cabinets were added in area A. After the system detects this change, it automatically adjusts the grid division of area A to include these new devices in management, and adjusts the surrounding grid boundaries accordingly to ensure that all equipment is accurately covered. At the same time, the equipment operation status analysis unit continuously monitors the operating parameters of these new devices. A few weeks later, the system discovered that the temperature of one of the distribution cabinets rose abnormally, and immediately issued an alarm and recommended that maintenance be arranged as a priority. Based on the importance and urgency of the equipment, the system also dynamically adjusts the grid density of area A to more efficiently dispatch maintenance resources and ensure rapid response and processing.

[0071] The present invention will be further described below in conjunction with Examples 1 to 3:

[0072] Embodiment 1:

[0073] The maintenance standard matching module has a detailed maintenance standard database built in, and automatically matches the most appropriate maintenance standard and cycle for each device based on the type, service life and real-time operating status of the device, generating an accurate maintenance task list. The following are the detailed steps and calculation process:

[0074] Step 1: Obtain equipment characteristics from the equipment information acquisition module, wherein the equipment characteristics include the type of equipment, service life, operating parameter history records, and current status data.

[0075] Step 2: Based on the acquired equipment features, a fuzzy matching algorithm is used to compare and analyze the preset standards in the maintenance standard database. The specific algorithm is:

[0076] Assume that the device feature vector D = T, Y, P1, P2, ..., P n , where T is the equipment type, Y is the service life, P1, P2, ..., P n Each standard record in the maintenance standard database is S i =(T i , Y i , R i1 , R i2 , …, R in , C i , M i ) where T i is the specific device type, Y i R is the applicable working years range. i1 , R i2 , …, R in is the standard range or threshold of each operating parameter, C i is the maintenance cycle, M i is the corresponding maintenance method or standard; for each device D, calculate its comparison with each standard S in the database i The similarity Sim(D, S i ):

[0077]

[0078] Among them, α, β, ω j is the weight coefficient, satisfying

[0079] δ(T,T i ) is the matching degree of the device type. If T = T i Then δ=1, otherwise δ=0;

[0080] γ(Y,Y i ) is the matching function of working years. If Y is in Y i If it is within the range, then γ=1, otherwise γ=0;

[0081] η(P j , R ij ) is the matching function of the running parameters. If P j In R ij If it is within the range, then η=1, otherwise a value between 0 and 1 is assigned according to the degree of deviation.

[0082] Step 3: Select the similarity Sim(D, S i ) The highest standard record is taken as the matching result, and the maintenance cycle C is extracted from it. i and maintenance method M i .

[0083] Step 4: Based on the matching results, a maintenance task list including equipment identification, recommended maintenance cycle, specific maintenance method and time is automatically generated.

[0084] The following examples illustrate the implementation and calculation process of the above steps:

[0085] S1: Obtain detailed information of a specific equipment (e.g., a steam turbine) in a thermal power plant from the equipment information collection module, including equipment type T (e.g., "steam turbine"), service life Y (e.g., "10 years"), and recent operating parameter history records, such as vibration amplitude P1, temperature P2, pressure P3, etc.

[0086] S2: Using fuzzy matching algorithm, the extracted equipment features are compared with the standard records in the maintenance standard database one by one, and the similarity is calculated.

[0087] The maintenance standards database contains a standard record S for steam turbines. i , where T i = "steam turbine", Yi = 8 to 12 years, Ri1 = 0 to 0.05 mm (vibration amplitude), Ri2 = 200 to 250 degrees Celsius (temperature), Ri3 = 5 to 7 MPa (pressure), C i = "Once a year", M i = "Comprehensive inspection and maintenance".

[0088] Calculate the similarity Sim(D, S i ):

[0089] Device type matching degree δ(T, T i )=1, because T=T i = "steam turbine".

[0090] Working years matching function γ(Y, Y i )=1, because Y=10 years, in Y i = In the range of 8 to 12 years.

[0091] Run the parameter matching function η(P j , R ij ) are calculated as follows:

[0092] Vibration amplitude: Assuming P1 = 0.03 mm, at R i1 = 0 to 0.05 mm, so η(P1, R i1 )=1.

[0093] Temperature: Assume P2 = 220 degrees Celsius, at R i2 =200 to 250 degrees Celsius, so η(P2, R i1 )=1.

[0094] Pressure: Assume P3 = 6 MPa, at R i3 =5 to 7 MPa, so η(P3, R i3 )=1.

[0095] Set the weight coefficients α = 0.2, β = 0.2, ω1 = ω2 = ω3 = 0.2, then the similarity Sim (D, S i ) is calculated as:

[0096] Sim(D,S i )=0.2·1+0.2·1+0.2·1+0.2·1+0.2·1=1

[0097] This result indicates that the device characteristics exactly match the standard records.

[0098] S3: Due to Sim(D, S i )=1, directly select this standard record as the optimal matching result and extract the maintenance cycle C i = "Once a year" and maintenance method M i = "Comprehensive inspection and maintenance".

[0099] Based on the matching results, a maintenance task list for the steam turbine is automatically generated, which clearly marks the maintenance time as once a year, the specific maintenance method as comprehensive inspection and maintenance, and specifies the corresponding resource requirements and personnel allocation plan.

[0100] Embodiment 2:

[0101] The fault prediction unit combines the equipment type and known fault cases to establish a fault pattern library, and compares the real-time data with the patterns in the fault pattern library through a pattern matching algorithm to identify potential fault types and fault points. The fault prediction unit compares the real-time data with the patterns in the fault pattern library through a decision tree algorithm to identify potential fault types and fault points, and the implementation steps include:

[0102] S1: Obtain the real-time operation data of the equipment from the equipment information collection module, and select the characteristic parameters closely related to the failure mode in combination with the equipment type and known failure cases to construct the failure feature vector.

[0103] S2: Use historical fault data and corresponding fault modes as training sample sets; use decision tree algorithm to learn the training sample sets and build a decision tree model; the decision tree model consists of nodes and edges, each node represents a test of a feature attribute, and each edge represents a branch corresponding to the test result. When building the decision tree model, the Gini index algorithm is used to select the optimal split feature. The calculation formula of the Gini index algorithm is:

[0104]

[0105] Among them, k represents the number of categories, p i Represents the probability of category i appearing in the data set Data.

[0106] S3: Input the fault feature vector constructed in real time into the decision tree model, starting from the root node, traverse the nodes and edges of the decision tree in sequence according to the test results of the feature attributes; at each node, select the corresponding edge according to the value of the feature attribute until reaching the leaf node; the category represented by the leaf node is the identified fault mode.

[0107] In order to more clearly illustrate the implementation of the fault pattern recognition method of the fault prediction unit in the present invention using the decision tree algorithm, a detailed description is given below through specific examples.

[0108] Assuming that a steam turbine unit in a thermal power plant may have multiple failure modes during operation, such as bearing overheating, blade wear, shaft misalignment, etc. In order to accurately identify these failure modes, a decision tree algorithm is used for pattern matching. Specific embodiment:

[0110] Construct fault feature vector: Obtain key operating parameters of the steam turbine unit from the equipment information acquisition module, such as vibration amplitude, vibration frequency, bearing temperature, bearing temperature, oil pressure, etc. Preprocess these parameters, including removing outliers, data smoothing, normalization, etc., to eliminate the impact of noise and dimensional differences on subsequent analysis. Combined with the working principle of the steam turbine unit and known fault cases, select characteristic parameters closely related to the fault mode, such as vibration spectrum characteristics, temperature change trends, etc., to construct the fault feature vector. For example, for a bearing overheating fault, the bearing temperature, vibration amplitude, and vibration frequency may be selected as characteristic parameters.

[0111] Establish a decision tree model: Collect historical fault data, including operating parameter records, fault types, and corresponding treatment measures when the fault occurs. These data will be used as training sample sets to build a decision tree model. Use a decision tree algorithm (such as the C4.5 algorithm) to learn the training sample set. In the process of building a decision tree, first calculate the information gain of each feature attribute, and select the feature with the largest information gain as the split attribute of the current node. Then, divide the data set into several subsets according to the value range of the feature, and recursively build a subtree on each subset. Until the stopping condition is met (such as all samples in the subset belong to the same category, the number of samples in the subset is less than the preset threshold, etc.), the recursion stops, and the main category of the current subset is used as the category of the leaf node.

[0112] Through the above process, a decision tree model is constructed, which can reflect the mapping relationship between different fault modes and characteristic parameters.

[0113] Pattern matching and fault identification: When the steam turbine unit is running, its operating parameters are collected in real time and a fault feature vector is constructed. The fault feature vector is input into the decision tree model, starting from the root node, and the nodes and edges of the decision tree are traversed in sequence according to the values ​​of the feature attributes. At each node, the corresponding edge is selected according to the value of the current feature attribute until a leaf node is reached. The category represented by the leaf node is the identified fault mode. For example, if the category represented by the leaf node is "bearing overheating", it indicates that the steam turbine unit may have a bearing overheating fault.

[0114] In practical applications, the decision tree model constructed by the above method can be embedded in the fault prediction system of a thermal power plant. When the system detects abnormal operating parameters of the steam turbine unit, the fault prediction process is automatically triggered, and the decision tree model is used to perform pattern matching on real-time data, quickly identify potential fault modes, and issue early warning signals. Maintenance personnel can take timely measures to investigate and repair according to the early warning information, thereby avoiding further deterioration of the fault and causing equipment damage or shutdown accidents.

[0115] Embodiment 3:

[0116] The maintenance plan generation module achieves the goal of minimizing maintenance time and maximizing resource utilization by optimizing sorting and time scheduling. The task priority evaluation method is as follows:

[0117] Urgency and scope of impact: The system assigns an urgency score and a scope of impact score to each maintenance task based on the urgency of the equipment failure (such as equipment safety risks, production impact, etc.) and the potential scope of impact (such as the number of production lines affected, downtime, etc.).

[0118] Resource requirements and availability: Consider the type and quantity of resources (such as maintenance personnel, tools, spare parts, etc.) required for each task, as well as the availability of current resources, and evaluate the degree to which the resource requirements of the task are met.

[0119] The weighted summation method is used to weight the urgency score and the impact score according to the preset weights to obtain the priority index of each task. For example, let the urgency score be E, the impact score be I, and the weights be w E and w I , then the priority index P = w E ·E+w I I. Sort the tasks by priority index from high to low.

[0120] Sorting adjustment considering resource constraints: Based on the preliminary sorting, consider resource constraints. For tasks with similar resource requirements and resource competition, adopt staggered scheduling or resource optimization allocation strategies to avoid resource conflicts. For example, if two tasks require the use of the same professional equipment, arrange one of them during the equipment's idle time, or find alternative resources.

[0121] The scheduling algorithm is:

[0122] According to the maintenance personnel's schedule, equipment downtime and other constraints, a feasible time window is assigned to each task. The determination of the time window needs to consider the duration of the task, the dependencies between the tasks (such as some tasks must be completed after other tasks are completed), and the availability of resources. Based on the time window allocation, heuristic algorithms (such as genetic algorithms, simulated annealing algorithms, etc.) are used to optimize the time window to further reduce the maintenance time. The optimization goals include reducing the waiting time between tasks and improving the continuous utilization of resources. In response to possible emergencies (such as the emergence of new faults, resource failures, etc.), a certain buffer time is reserved in the schedule to enhance the flexibility and robustness of the plan.

[0123] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0124] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid, characterized by: include: The equipment information collection module includes an equipment status sensor unit, an operating parameter recording unit, and an image recognition unit, which is used to collect the working status data, operating parameters, and equipment appearance images of various equipment in the thermal power plant in real time; GIS dynamic grid construction module is used to divide the thermal power plant area into multiple dynamic grids and integrate with the geographic information system to spatially locate the equipment. Each grid contains a corresponding set of equipment. The maintenance standard matching module has a built-in maintenance standard database, which is used to automatically match the corresponding maintenance standards and cycles according to the equipment type, service life and operating status, and generate a maintenance task list; The data analysis module includes an equipment operation status analysis unit, a fault prediction unit, and a maintenance priority sorting unit. This module uses the data from the equipment information acquisition module and combines it with GIS dynamic grid information to analyze abnormal changes in the equipment operation status, predict potential fault points, and prioritize maintenance tasks based on the output of the maintenance standard matching module. The maintenance plan generation module automatically generates a maintenance plan based on the output results of the data analysis module, including maintenance time, required resources, personnel allocation and safety measures; The supervision and feedback module is used to monitor the implementation of the maintenance plan, collect feedback data from on-site maintenance personnel, evaluate the maintenance quality, and adjust subsequent maintenance plans and standards accordingly to form a closed-loop continuous improvement mechanism.

2. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 1 is characterized in that: The GIS dynamic grid construction module dynamically adjusts the grid division and equipment positioning by calculating the adjustment parameters of each grid unit, including: Dynamically update the content of grid cells and device collections based on the addition, removal or location change of devices; Dynamically adjust the density and size of the grid based on the urgency and priority of equipment maintenance.

3. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 2 is characterized in that: The step of generating a maintenance task list by the maintenance standard matching module includes: Step 1: Obtain equipment characteristics from the equipment information acquisition module, wherein the equipment characteristics include equipment type, service life, operating parameter history records and current status data; Step 2: Based on the acquired equipment features, a fuzzy matching algorithm is used to compare and analyze the preset standards in the maintenance standard database. The specific algorithm is: Assume that the device feature vector D = (T, Y, P1, P2, ..., P n ), where T is the equipment type, Y is the service life, P1, P2, …, P n Each standard record in the maintenance standard database is S i =(T i ,Y i ,R i1 ,R i2 ,…,R in ,C i ,M i ) where T i is the specific device type, Y i For the applicable working years, R i1 ,R i2 ,…,R in is the standard range or threshold of each operating parameter, C i is the maintenance cycle, M i is the corresponding maintenance method or standard; for each device D, calculate its comparison with each standard S in the database i The similarity Sim(D,S i ): Among them, α, β, ω j is the weight coefficient, satisfying δ(T,T i ) is the matching degree of the device type. If T = T i Then δ=1, otherwise δ=0; γ(Y,Y i ) is the matching function of working years. If Y is in Y i If it is within the range, then γ=1, otherwise γ=0; η(P j ,R ij ) is the matching function of the running parameters. If P j In R ij If the deviation is within the range, then η = 1, otherwise a value between 0 and 1 is assigned according to the degree of deviation; Step 3: Select the similarity Sim(D,S i ) The highest standard record is taken as the matching result, and the maintenance cycle C is extracted from it. i and maintenance method M i ; Step 4: Based on the matching results, a maintenance task list including equipment identification, recommended maintenance cycle, specific maintenance method and time is automatically generated.

4. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 1 is characterized by: The equipment operation status analysis unit obtains equipment operation parameters and real-time working status data collected by the equipment information acquisition module, and compares the obtained data with the normal state range to monitor in real time whether the equipment is in an abnormal state.

5. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 1 is characterized by: The fault prediction unit establishes a fault mode library in combination with equipment types and known fault cases, and compares real-time data with the patterns in the fault mode library through a pattern matching algorithm to identify potential fault types and fault points.

6. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 5 is characterized by: The pattern matching algorithm adopts a decision tree algorithm.

7. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 6 is characterized in that: The fault prediction unit compares the real-time data with the patterns in the fault pattern library through a decision tree algorithm to identify potential fault types and fault points, and the implementation steps include: S1: Obtain the real-time operation data of the equipment from the equipment information collection module, and select the characteristic parameters closely related to the failure mode in combination with the equipment type and known failure cases to construct the failure feature vector; S2: Use historical fault data and corresponding fault modes as training sample sets; use decision tree algorithm to learn the training sample sets and build a decision tree model; the decision tree model consists of nodes and edges, each node represents a test of a feature attribute, and each edge represents a branch corresponding to the test result; S3: Input the fault feature vector constructed in real time into the decision tree model, starting from the root node, traverse the nodes and edges of the decision tree in sequence according to the test results of the feature attributes; at each node, select the corresponding edge according to the value of the feature attribute until reaching the leaf node; the category represented by the leaf node is the identified fault mode.

8. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 7 is characterized by: When building a decision tree model, the optimal split feature is selected through the Gini index algorithm. The calculation formula of the Gini index algorithm is: Among them, k represents the number of categories, p i Represents the probability of category i appearing in the data set Data.

9. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 1 is characterized in that: Also includes: The mobile maintenance module can quickly obtain equipment information, maintenance standards and historical maintenance records by scanning the equipment QR code or GPS positioning, assist in completing maintenance tasks, and upload maintenance progress and problems found in real time.

10. The thermal power overhaul platform integrating equipment information, maintenance standards and GIS dynamic grid according to claim 1 is characterized in that: When formulating a maintenance plan, the maintenance plan generation module optimizes the sequencing and time arrangement of maintenance tasks with the goal of minimizing maintenance time and maximizing resource utilization.

Images