Power distribution network equipment operation data event association analysis method and system

Abstract

The invention provides a power distribution network equipment operation data event association analysis method and system, and relates to the technical field of power distribution network operation data analysis, and the method comprises the steps: obtaining the operation data of each equipment, and formulating a construction plan containing the construction project, time, required resources, project position and other contents according to the operation data; setting the consulting authority of constructors to the construction project, constructing a consultable, and constructing a single model containing a top layer, a sub-layer and a bottom layer according to the consultable; and associating the construction plan to the single model to obtain a construction path, and sending the path to a construction principal in a preset time before construction, thereby being capable of associating the operation data of the equipment with the construction event, ensuring smooth implementation of the construction project of the power distribution network, and realizing scientific, normalized and intelligent management.

Description

Technical Field

[0001] The present application relates to the technical field of distribution network operation data analysis, and in particular to a method and system for analyzing distribution network equipment operation data event correlation. Background Art

[0002] In the power system, the distribution network is the key link between the transmission network and end users. The stability and reliability of its equipment are directly related to power supply quality and user satisfaction. As the scale and complexity of the distribution network continue to expand, the operating data generated by various devices is growing massively.

[0003] In traditional operation and maintenance models, equipment operating data (such as current, voltage, and temperature) is usually stored in independent monitoring systems (such as SCADA), and construction plans rely on manual experience or fixed cycles rather than real-time equipment status data.

[0004] Therefore, traditional O&M models struggle to correlate equipment operating data with construction events, leading to a disconnect between construction plans and actual equipment status. This is particularly true in multi-equipment collaborative construction scenarios, where the lack of dynamic correlation analysis capabilities across construction projects, resource requirements, and spatial and temporal locations can easily lead to resource conflicts and construction delays, hindering the advancement of intelligent O&M in distribution networks. Summary of the Invention

[0005] In order to associate the operating data and construction events of equipment and improve the intelligent operation and maintenance level of distribution network, the present application provides a distribution network equipment operating data event correlation analysis method and system.

[0006] In the first aspect, the present application provides a method for analyzing distribution network equipment operation data events, which adopts the following technical solutions: A method for analyzing distribution network equipment operation data events, comprising the following steps: Data collection: Obtain the operating data of each device and set a corresponding construction plan based on the operating data of each device. The construction plan includes the construction project, construction time, resources required for construction, and construction project location; Modeling: Set up access permissions for construction projects for each construction worker, build a table of accessible construction projects based on the access permissions, and build a single model based on the table of accessible construction projects. The single model includes: a top layer for representing construction projects, a sub-layer for representing construction links, and a bottom layer for representing construction steps; Allocation: Associate the construction plan in the single model, obtain the construction path based on the single model after the construction plan is associated, and send the construction path to the construction person in charge within the preset time before the construction time.

[0007] This application first obtains the operating data of each device and uses it as the basis for the subsequent formulation of the construction plan. Subsequently, this application will set the access rights of each construction worker to the construction project. The accessible construction project table and the monomer model (including the top layer, sub-layer and bottom layer structure) constructed based on the access rights can clearly present the hierarchical structure and the relationship between each link of the construction project, and provide a data basis for construction path planning. Subsequently, this application associates the construction plan in the monomer model, accurately matches the detailed information in the construction plan with the elements of each level in the monomer model, and finally, obtains the construction path after associating the construction plan in the monomer model. It can plan the optimal construction path according to the location of the construction project, the sequence of construction links and the dependency relationship of construction steps. The optimal construction path can reduce the moving distance and time of construction workers, improve construction efficiency, and send the construction path to the construction manager within the preset time before construction. This application realizes the association between the operating data of the equipment and construction events by setting up a monomer model, thereby improving the level of intelligent operation and maintenance of the distribution network.

[0008] Optionally, the method further includes: An initial risk value is set for each construction step, and the initial risk value of the construction step included in the construction path is recorded as the first data. Equipment parameters and environmental data are collected, and the fuzzy comprehensive evaluation method is used to determine the risk correction value of each construction step based on the equipment parameters and environmental data. The sum of the first data and the risk correction value is used as the final risk value of each construction step. When the final risk value is greater than the preset threshold, an alarm signal is sent to the construction manager.

[0009] This application sets an initial risk value for each construction step and uses a fuzzy comprehensive evaluation method to determine a risk correction value based on equipment parameters and environmental data. By comprehensively considering the inherent risk of the construction step itself (initial value) and the dynamic risk brought about by equipment and environmental factors (corrected value), it can more comprehensively and accurately assess the actual risk status of each construction step, reducing the one-sidedness of single-factor assessments. This application adds the initial risk value and the risk correction value to obtain a final risk value. When the final risk value exceeds the threshold, an alarm signal is issued to the construction manager. By collecting equipment parameters and environmental data and conducting risk assessments and early warnings based on this data, this application makes construction management more scientific and data-based. Construction managers can rationally arrange construction resources based on the risk assessment results, optimize the construction process, and improve construction efficiency.

[0010] Optionally, after issuing the alarm signal, the method further includes: Obtain the construction steps included in the construction path, record them as target steps, and trace back the construction links associated with the target steps in the monomer model after the construction plan is associated, record them as target links; Determine whether the association between the target step and the target link is correct. If so, revise the monomer model after the associated construction plan, record the revised monomer model as the new monomer model, and execute the assigned steps; if not, issue a signal to stop construction.

[0011] After the alarm signal is issued, this application obtains the construction steps in the construction path and records them as target steps. By backtracking the construction links associated with the target steps in the monomer model after the construction plan is linked, and recording them as target links, it further explores the construction links related to the problematic steps. The monomer model integrates various aspects of construction information. With the help of the monomer model after the construction plan is linked, it can comprehensively and systematically sort out the links that are logically connected to the target steps, which in turn helps to identify the cause of the risk alarm.

[0012] Due to various reasons (such as design changes, information transmission errors, etc.), the association relationship in the monomer model may deviate. Through the above scheme, this application can promptly discover the erroneous associations in the monomer model after the associated construction plan, so that the construction logic reflected by the monomer model is consistent with the actual situation. If the association relationship is found to be incorrect, this application will correct the monomer model after the associated construction plan and record the corrected model as the new monomer model. When it is judged that the association relationship between the target step and the target link is correct, a signal to stop construction is issued, indicating that under the construction logic reflected by the current model, the target step has a risk that cannot be ignored, and continuing construction may lead to more serious consequences. Stopping construction in time can reduce casualties and property losses.

[0013] Optionally, the method further includes: The location, status, and available time data of resources required for construction are collected in real time. Based on the new monomer model, a greedy algorithm is used to detect resource conflicts between target steps. If resource conflicts exist between target steps, the priority scoring step is executed. Calculate priority scores: Use the analytic hierarchy process to assign corresponding weights to the final risk value, number of affected users, and construction duration of each target step. Calculate the priority score of each target step based on the final risk value, number of affected users, construction duration, and corresponding weights. Allocate resources required for construction based on the priority score.

[0014] Real-time data collection on the location, status, and availability of resources required for construction provides the construction management team with a comprehensive and timely understanding of the actual resource situation. Subsequently, based on a new single-unit model, this application utilizes a greedy algorithm to detect resource conflicts between target steps, enabling rapid and accurate identification of resource allocation issues. Once a conflict is detected, resources are reallocated by calculating priority scores, reducing the blindness and irrationality of resource allocation. By promptly identifying and resolving resource conflicts, construction pauses and delays caused by insufficient or improper resource allocation are reduced.

[0015] The analytic hierarchy process is used to assign corresponding weights to the final risk value, the number of affected users, and the construction duration of each target step, and the priority score is calculated based on these factors. The above scheme takes into account multiple important factors, making resource allocation decisions more scientific and reasonable. The analytic hierarchy process can decompose complex decision-making problems into multiple levels and factors, and determine the relative importance of each factor through pairwise comparison, thereby assigning a reasonable weight to each factor, reducing the one-sidedness and subjectivity of single-factor decisions, and improving the accuracy and reliability of decisions. This application quantifies the importance and urgency of each construction step by calculating the priority score, providing the construction management team with a clear basis for decision-making. When resources are limited, the management team can reasonably allocate resources according to the priority score to ensure that resources are invested where they are most needed.

[0016] Optionally, before executing the step of calculating the priority score, the method further includes: Determine whether there is an execution order requirement between the conflicting target steps. If so, allocate the required construction resources according to the execution order; if not, execute the step of calculating the priority score.

[0017] When conflicting target steps have a clear order of priority, allocating resources according to this order can minimize ineffective resource occupation and idleness. For steps with order requirements, there is no need to perform complex priority scoring calculations, and resources can be allocated directly in order, reducing decision-making links and time costs.

[0018] Optionally, before sending the construction route to the construction manager within a preset time before the construction start, the method further includes: Obtain a list of construction personnel and add skill tags to each construction personnel, including professional field, qualification level, practical experience and historical construction score; Construction requirements are extracted based on the description in the construction plan. The degree of match between construction personnel and construction requirements is obtained through label matching algorithms and skill labels. The construction personnel with the highest degree of match are designated as construction supervisors.

[0019] A list of construction personnel is obtained and skill tags are added to them. These skill tags cover multiple dimensions, including professional field, qualification level, practical experience, and historical construction ratings. This allows the construction party to gain a comprehensive and detailed understanding of each construction worker's abilities and characteristics, forming a complete personnel information database. Construction requirements are extracted based on the construction plan, and a tag matching algorithm is used to match construction personnel with these requirements, accurately identifying the personnel who best meet the requirements. The tag matching algorithm comprehensively considers multiple skill tag factors. By setting different weights and matching rules, it calculates the degree of match between each construction worker and the construction requirements. The construction worker with the highest degree of match is designated as the construction supervisor, ensuring that the construction supervisor has sufficient skills and experience to organize and direct the construction work. With a deep understanding of the construction task requirements and processes, the construction supervisor can rationally arrange the work of construction personnel and coordinate and resolve problems that arise during the construction process, thereby improving the execution efficiency of the entire construction team.

[0020] The construction supervisors selected through skill tag matching can better understand and grasp construction requirements and formulate scientific and reasonable construction plans and quality control measures. During the construction process, the construction supervisors can use their professional knowledge and skills to effectively guide and supervise the construction personnel, promptly identify and correct quality issues during construction, and ensure that the construction quality meets the standard requirements.

[0021] Because the construction manager is highly aligned with construction requirements and is intimately familiar with construction tasks and processes, he or she is able to effectively communicate and coordinate with other project stakeholders. Before construction begins, the manager can fully communicate with the design and supervision teams regarding construction plans and technical requirements to ensure smooth construction progress. During construction, the manager can promptly communicate with construction personnel on progress and any issues, coordinating and resolving resource allocation and process integration issues, minimizing construction delays caused by poor communication and coordination.

[0022] Optionally, after setting the construction worker with the highest matching degree as the construction manager, the method further includes: Verify the construction plan: Determine whether the same construction items exist in the single model after the construction plan is associated. If so, execute the steps of the verification person in charge; if not, do nothing. Verify the person in charge: Determine whether the construction person in charge is the same. If so, perform the deletion step; if not, perform the desensitization step; Delete: Delete any construction project in the same construction project; Desensitization: Obtain historical construction plans, build a construction terminology semantic library based on the historical construction plans using NLP technology, desensitize the construction plans based on the construction terminology semantic library and the construction path of each construction manager, and send the desensitized construction plans to the corresponding construction managers.

[0023] By verifying the construction plan and determining whether the same construction projects exist in the monomer model after the associated construction plan, this application can promptly discover possible duplicate arrangements in the construction plan. In large and complex construction projects involving multiple sub-projects and numerous construction links, the construction plans formulated by different departments or teams may overlap or repeat. For example, in a construction project, the electrical installation team and the intelligent system team may both be involved in line laying work. If this step is not verified, it may lead to repeated construction in the same area, resulting in waste of resources and delays in construction. Through this step, this application can discover and solve these problems in advance, and improve the rationality and uniqueness of the construction plan.

[0024] When duplicate construction projects are discovered, further verification of the responsible individuals is performed. If the same construction project managers exist, this indicates that the same team may have duplicated planning for the same project. In this case, the deletion step is performed to remove any of the duplicate construction projects, reducing confusion and duplication of resources during the construction process. If different construction project managers exist, the desensitization step is performed.

[0025] In the desensitization step, historical construction plans are obtained and a semantic library of construction terms is constructed based on NLP technology. By analyzing large amounts of text data in historical construction plans, NLP technology can accurately identify and understand the meaning and contextual relationships of these terms, and construct a semantic library that conforms to the characteristics of the construction industry. The construction plan is desensitized based on the semantic library of construction terms and the construction path of each construction manager. This application can remove sensitive information that may be involved in the construction plan, such as the specific construction location, key technical parameters, etc., while retaining the core content and key information of the construction plan. It tries to protect the information security of the construction project and reduce the risk of sensitive information being leaked to competitors or unrelated personnel, while ensuring that the construction manager can understand the construction tasks and requirements.

[0026] The desensitized construction plan is sent to the corresponding construction manager, ensuring they receive accurate and clear construction task information. Because desensitization removes unnecessary sensitive information, the construction plan is more concise and clear, allowing the manager to quickly understand their job responsibilities and task requirements, reducing information comprehension and communication costs.

[0027] Optionally, a semantic library of construction terms is constructed using NLP technology based on historical construction plans, including: Build an empty semantic library of construction terms, extract structured and unstructured data from historical construction plans, and perform syntactic analysis on the unstructured data to split it into multiple semantic units. The Word2Vec model is used to convert semantic units into semantic vectors, and the similarity between semantic vectors is calculated. Any semantic unit with a similarity greater than a preset similarity threshold is deleted to obtain the remaining semantic units. Fill the structured data and remaining semantic units into the construction terminology semantic library.

[0028] This application extracts both structured and unstructured data from historical construction plans, making full use of all information in historical construction plans and minimizing the problem of incomplete semantic libraries due to missing data, thus providing a rich data foundation for building a comprehensive and accurate semantic library of construction terminology. From building an empty semantic library to extracting data, splitting semantic units, converting semantic vectors, calculating similarity, and finally filling the semantic library, the entire process is automated. Compared with traditional manual organization and entry methods, automated processing greatly improves efficiency and reduces the time and cost of manual operations.

[0029] This application also performs syntactic analysis on unstructured data. Syntactic analysis can identify the grammatical structure and semantic relationships in sentences, decompose complex text descriptions into units with independent semantics, and more accurately understand the meaning of each part, providing a more accurate basis for subsequent semantic vector conversion and similarity calculation.

[0030] The Word2Vec model is used to convert semantic units into semantic vectors and calculate the similarity between semantic vectors. The Word2Vec model can map words or phrases into a low-dimensional vector space, so that semantically similar words are closer in the vector space. By calculating the similarity between semantic vectors, it is possible to accurately determine whether two semantic units have similar meanings. After calculating the similarity of the semantic vectors, any semantic unit with a similarity greater than a preset similarity threshold is deleted, thereby removing redundant information from the semantic library and minimizing duplicate semantic units from occupying storage space and affecting query efficiency. Filling the construction terminology semantic library with structured data and the remaining processed semantic units can ensure the standardization and consistency of the data in the semantic library.

[0031] Optionally, the method further includes: adding structured data and synonyms or near synonyms of semantic units to the construction term semantic library to obtain a new construction term semantic library.

[0032] In a second aspect, the present application provides a distribution network equipment operation data event correlation analysis system, which adopts the following technical solutions: A distribution network equipment operation data event correlation analysis system, comprising: a memory and a processor, The memory stores a computer-readable storage medium; When the processor processes the computer program stored on the computer-readable storage medium, the method according to the first aspect is implemented.

[0033] In summary, this application includes at least one of the following beneficial technical effects: 1. This application first obtains the operating data of each device and uses it as the basis for the subsequent formulation of the construction plan. Subsequently, this application will set up the access rights of each construction worker to the construction project. The accessible construction project table and the monomer model (including the top layer, sub-layer and bottom layer structure) constructed based on the access rights can clearly present the hierarchical structure and relationship between each link of the construction project, and provide a data basis for construction path planning. Subsequently, this application associates the construction plan in the monomer model, accurately matches the detailed information in the construction plan with the elements of each level in the monomer model, and finally, obtains the construction path after associating the construction plan in the monomer model. It can plan the optimal construction path based on the location of the construction project, the sequence of construction links and the dependency relationship of construction steps. The optimal construction path can reduce the moving distance and time of construction workers, improve construction efficiency, and send the construction path to the construction manager within the preset time before construction. This application realizes the association between the operating data of the equipment and construction events by setting up a monomer model, thereby improving the level of intelligent operation and maintenance of the distribution network.

[0034] 2. This application sets an initial risk value for each construction step and uses a fuzzy comprehensive evaluation method to determine a risk correction value based on equipment parameters and environmental data. By comprehensively considering the inherent risk of the construction step itself (initial value) and the dynamic risk brought about by equipment and environmental factors (corrected value), it can more comprehensively and accurately assess the actual risk status of each construction step, reducing the one-sidedness of single-factor assessments. This application adds the initial risk value and the risk correction value to obtain a final risk value. When the final risk value exceeds the threshold, an alarm signal is issued to the construction manager. By collecting equipment parameters and environmental data and conducting risk assessments and early warnings based on this data, this application makes construction management more scientific and data-based. Construction managers can rationally arrange construction resources based on the risk assessment results, optimize the construction process, and improve construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 This is a flow chart of Example 2 of the present application; Figure 3 This is a flowchart of Example 3 of the present application. DETAILED DESCRIPTION

[0036] The following combination Figures 1 to 3 This application is described in further detail.

[0037] Example 1: This example discloses a method for analyzing the correlation of distribution network equipment operation data events. Figure 1 The method includes: S11 data collection, S12 modeling, and S13 allocation. First, the operating data of each device is obtained and a construction plan including construction items, time, required resources, and project location is formulated based on the data. Then, the access rights of construction personnel to the construction items are set and a reference table is constructed. Then, a single model including a top layer (construction items), a sub-layer (construction links), and a bottom layer (construction steps) is constructed based on the data. Then, the construction plan is associated with the single model to obtain a construction path, and the path is sent to the construction person in charge at a preset time before construction. This embodiment includes the following steps: S11 data acquisition, through sensors or on-site surveys by construction personnel, obtains the operating data generated by each device in the distribution network during operation. The operating data includes: equipment operating time, load conditions, energy consumption, fault records, maintenance cycles and other information reflecting the status and performance of the equipment.

[0038] Based on operational data, a construction plan tailored to each piece of equipment is developed. For example, if a piece of equipment has recently experienced high loads and an increased risk of failure, its operating hours may be reduced or maintenance may be scheduled in advance. For stable equipment, more construction tasks may be assigned based on its performance. The construction plan includes the construction project, construction time, resources required, and location.

[0039] Construction project: Specific construction projects are determined based on the equipment's operating data and fault diagnosis results. For example, if a temperature sensor indicates a device's motor temperature is consistently too high, and construction personnel analyze the cause as likely due to motor bearing wear, the construction project may be to replace the motor bearing.

[0040] Construction time: Consider the production plan, operating status and impact of the equipment on production, and arrange the construction time reasonably.

[0041] Construction resources required: Estimate the human, material, and financial resources required based on the specific requirements of the construction project. For example, the hoisting of large equipment requires specialized lifting equipment, lifting tools, and experienced riggers; electrical equipment repair requires the necessary electrical components, testing instruments, and insulation tools.

[0042] Construction project location: Clarify the location of the construction project so that construction workers can find the construction project quickly and accurately.

[0043] S12 modeling sets the access rights of each construction worker to the construction project. Construction workers in different positions have different understanding requirements and authority scopes for construction projects.

[0044] For example, ordinary construction workers only need to understand the construction projects they are responsible for, construction time, resources required for construction, and the location of the construction projects; while the construction manager needs to grasp the overall situation of the entire construction project, including construction projects, construction time, construction progress, allocation of resources required for construction, quality control, and the location of construction projects.

[0045] A searchable construction project table is constructed based on the access permissions. The searchable construction project table includes the construction project information that each construction worker has the right to access, including project batches, project names, on-site survey records, locations and other information. The searchable construction project table is used to provide construction workers with an information query channel so that they can quickly find the information they need to know.

[0046] A monomer model is constructed based on the accessible construction project table. The monomer model adopts a hierarchical structure, including a top layer for representing construction projects, a sub-layer for representing construction links, and a bottom layer for representing construction steps.

[0047] Top-level view: This displays an overview of the entire construction project, organized by project, including basic information such as project name, project leader, and project start and end dates. This top-level view allows construction managers to conduct macro-management and scheduling of construction projects, understanding the overall project progress and resource allocation.

[0048] Sub-layer: Each construction project is broken down into several construction links, such as equipment disassembly, parts cleaning, equipment installation, commissioning and operation, etc. The sub-layer view describes in detail the specific content and requirements of each construction link, providing clear operational guidance for construction personnel.

[0049] Bottom layer: Each construction link is further refined into specific construction steps, such as what tools to use, what operating methods to adopt, and what safety precautions to pay attention to. The bottom layer view is the basis for the actual operation of the construction personnel, maintaining the normalization and standardization of the construction process.

[0050] S13 allocation is to associate the construction plan in the monomer model, that is, to correspond the specific information in the construction plan, such as construction projects, construction time, resources required for construction, etc., with the hierarchical structure in the monomer model, to achieve the integration and sharing of construction information. By associating the construction plan with the monomer model, the schedule of each construction project, resource allocation and the logical relationship between each construction link can be intuitively displayed in the monomer model.

[0051] For example, in the top-level view of a single model, the time progress bar of the entire construction project can be displayed, marking the start and end time of each construction stage; in the sub-level view, the quantity and type of resources required for each construction link, as well as the resource usage time and allocation status, can be displayed; in the bottom-level view, the actual execution status of each construction step can be recorded in detail, including the construction personnel's operation records, construction time records, quality inspection records, etc.

[0052] Based on the monomer model associated with the construction plan, the construction path is obtained. The construction path refers to the route that construction personnel take to complete each construction link and step in sequence according to the sequence and logical relationship of the construction project. Finally, the construction path is sent to the construction manager within the preset time before the construction time.

[0053] For example, construction location: construction sites in a certain area are distributed on three parallel streets, A, B, and C are on Street 1 (500 meters apart), and D and E are on Street 2 (800 meters apart). Street 1 and Street 2 are separated by a river (only connected by one bridge).

[0054] The construction sequence analysis process is as follows: This area has three tasks: ①10kV main line power outage (takes 1 hour); ② Replacement of branch line cables (requires 2 hours and depends on completion of ①); ③ Transformer temperature measurement (can be operated under power supply without power outage).

[0055] According to construction experience or construction constraints, the construction order should be: first ③, then ①, and finally ②.

[0056] Construction constraints can include: main line outages must be completed before branch line maintenance can begin; transformer replacements must precede commissioning of the associated switchgear; live work must take precedence over outage work; high-altitude work (such as tower maintenance) must avoid intersecting work below; and tasks within the same construction team must be performed continuously (to reduce personnel / tool ​​transfer costs).

[0057] Based on the location relationship and construction sequence, an algorithm is used to generate a path. The core goal is "shortest total journey + no intersection conflicts". The specific method is as follows: Use the construction sequence as the time axis to connect each construction point in sequence, give priority to "continuous paths within the area" (such as working from east to west on the same street), and try to avoid "cross-area return" (such as going from A to B, and then returning to C near A).

[0058] In other embodiments, a path optimization algorithm (such as an improved Dijkstra algorithm or a genetic algorithm) is introduced for multiple construction points, and the input parameters include "distance matrix of each point", "sequence constraint weight", and "cross-operation penalty coefficient", and the output is "the construction path with the shortest total mileage and no conflicts".

[0059] In other embodiments, if a temporary change occurs during construction (such as a delay at a certain point or a new task), the position and sequence are updated in real time through the monomer model associated with the construction plan, and the path is recalculated (for example, the original path A→B→C is adjusted to A→C→B due to a delay in B, reducing waiting time).

[0060] By adopting the above solution, this embodiment can associate the operating data of the equipment with construction events. In terms of construction planning, the construction time is accurately determined based on the operating data, avoiding the high-incidence period of equipment failure and peak electricity consumption, ensuring progress and reducing costs; in terms of resource allocation, resources are arranged in advance based on operating data such as maintenance cycles to ensure timely and efficient maintenance; in terms of quality control, the construction time is reasonably arranged in combination with fault records, and sufficient inspection and repair time is reserved to improve construction quality and reduce rework; at the information management level, an association model is established to achieve real-time data feedback and dynamic matching, which makes it easier for management personnel to grasp the situation at any time and make timely adjustments, thereby improving the efficiency and level of construction management from multiple dimensions, ensuring the smooth implementation of distribution network construction projects and realizing scientific, standardized and intelligent management.

[0061] Example 2: Reference Figure 2 The difference between this embodiment and embodiment 1 is that before the construction personnel receive the construction path matched in S13 and perform construction according to the construction path, the method further includes: S21 calculates the risk value and sets an initial risk value for each construction step. The initial risk value is determined based on past experience, industry standards or expert experience.

[0062] The process of setting the initial risk value based on past experience is as follows: Data collection: Collect detailed information from past similar construction projects, including construction steps, risk events that occurred, the consequences of risk events (such as casualties, property losses, construction delays, etc.), and the frequency of risk events.

[0063] Statistical Analysis: Statistical analysis is performed on collected historical data to calculate the probability of risk occurring at each construction step and the extent of the loss that risk may cause. Risk probability can be categorized into different levels, such as low, medium, and high, based on the frequency of risk events. Loss levels can also be categorized based on the severity of the loss, such as minor, moderate, severe, and major.

[0064] Determining the Initial Risk Value: An initial risk value is determined for each construction step, combining the risk probability and loss severity. This embodiment uses a risk matrix approach, using the risk probability and loss severity as the two dimensions of the matrix. Initial risk values ​​are determined based on different combinations. For example, when the risk probability is high and the loss severity is severe, the corresponding initial risk value can be set to a higher value.

[0065] During the pre-repair preparation step, repair delays occurred twice due to incomplete tool preparation, accounting for 10% of the total number of similar projects. The resulting losses primarily resulted in extended construction time and increased labor costs, and the degree of loss was assessed as minor. Based on the risk matrix, the initial risk value for this step was set to 2.

[0066] In the transformer power outage operation steps, equipment damage caused by misoperation occurred once, accounting for 5%. The degree of loss was serious, and the initial risk value was set to 8.

[0067] No obvious risk events occurred during the transformer oil draining and internal inspection steps, but considering the complexity of the internal inspection, the initial risk value was set to 3 with reference to industry standards.

[0068] In the winding insulation treatment step, there were three cases of recurrence of failure after repair due to insulation material quality problems, accounting for 15%. The degree of loss was moderate, and the initial risk value was set to 4.

[0069] During the transformer oil return and sealing inspection steps, oil leakage caused environmental pollution and equipment damage once, accounting for 5%. The degree of loss was medium, and the initial risk value was set to 5.

[0070] In the transformer power transmission test steps, equipment failures caused by incorrect test parameter settings occurred twice, accounting for 10%. The degree of loss was serious, and the initial risk value was set to 7.

[0071] The process of setting the initial risk value based on industry standards is as follows: Find industry standards: Consult relevant national and industry construction safety standards, specifications, and guidelines to understand the risk classification and risk value setting methods for similar construction steps. For example, in construction, you can refer to the "Construction Safety Inspection Standards" for risk assessment requirements for different construction processes.

[0072] Comparative analysis: Compare and analyze the construction steps in this embodiment with similar steps in industry standards. Consider the impact of the project's special circumstances (such as the construction environment, construction technology, equipment selection, etc.) on risk, make appropriate adjustments to the risk values ​​in the industry standards, and determine the initial risk value of this project.

[0073] The initial risk value of the construction steps included in the construction path in S13 matching is recorded as the first data, and equipment parameters and environmental data are collected. The equipment parameters include the operating status, performance indicators, aging degree, etc. of the equipment; the environmental data includes temperature, humidity, weather conditions, geographical conditions of the construction site, etc.

[0074] The fuzzy comprehensive evaluation method is used to determine the risk correction value of each construction step based on equipment parameters and environmental data. The fuzzy comprehensive evaluation method can handle the uncertainty and ambiguity of data. By establishing an evaluation factor set, a comment set and a weight set, and applying the fuzzy transformation principle, the risk correction value of each construction step is obtained.

[0075] The evaluation factor set is a collection of factors that influence the risk correction value of a construction step. Taking into account equipment parameters and environmental data, this set of factors can include equipment operating status (such as equipment failure rate, equipment aging, and equipment load), equipment performance indicators (such as equipment accuracy, efficiency, and stability), ambient temperature, ambient humidity, weather conditions (such as sunny, rainy, snowy, and windy), and construction site geological conditions.

[0076] For example, for the transformer outage operation step, the evaluation factor set is expressed as ,in The operating life of the equipment, For operator skill level, For weather conditions, The completeness of safety measures.

[0077] The evaluation set is a collection of various possible evaluation results of the evaluation factors. The evaluation results of the risk adjustment value can usually be divided into different levels, such as low risk adjustment, relatively low risk adjustment, medium risk adjustment, relatively high risk adjustment, and high risk adjustment.

[0078] For the hoisting construction steps of large mechanical equipment, the comment set is expressed as ,in Indicates a low-risk correction, Indicates a lower risk correction, Indicates a medium risk correction, Indicates a higher risk correction, Indicates a high risk correction.

[0079] The single factor fuzzy evaluation matrix is ​​obtained by evaluating each evaluation factor separately. , according to its actual data and evaluation criteria, determine its membership degree for each comment in the comment set ,in (n is the number of evaluation factors), (m is the number of comments).

[0080] Membership Indicates evaluation factors Belong to the comments The degree of membership of all evaluation factors is formed into an n×m matrix R, which is the single-factor fuzzy evaluation matrix: ; For example, for the equipment operating life , after evaluation, its membership to the comment set V is determined to be =0.1, =0.2, =0.3, =0.3, =0.1, then the single factor evaluation vector of equipment failure rate is (0.1, 0.2, 0.3, 0.3, 0.1).

[0081] Similarly, according to the above scheme, the single factor evaluation vectors of other evaluation factors can be obtained, and then the single factor fuzzy evaluation matrix R can be obtained. The specific form of the single factor fuzzy evaluation matrix in this embodiment is as follows: .

[0082] The evaluation factors are divided into target layer, criterion layer and indicator layer. The target layer is the risk correction value evaluation of the construction steps, the criterion layer is the equipment parameters and environmental parameters, and the indicator layer is the specific evaluation factors. For each factor at the same level, the relative importance is determined by pairwise comparison and a judgment matrix is ​​constructed. The elements of the judgment matrix are: Indicates the importance of factor i relative to factor j, and is quantified using a 1-9 scale.

[0083] By solving the eigenvalues ​​and eigenvectors of the judgment matrix, the weight vector W = (0.2, 0.3, 0.2, 0.3) of each evaluation factor is obtained.

[0084] The consistency index and random consistency ratio of the judgment matrix are calculated. When the random consistency ratio is less than 0.1, the judgment matrix is ​​considered to have satisfactory consistency and the weight vector is acceptable; otherwise, the judgment matrix needs to be adjusted and the weight vector needs to be recalculated.

[0085] According to the single-factor fuzzy evaluation matrix R and the weight vector, the fuzzy synthesis operator is used to perform fuzzy comprehensive evaluation, and the fuzzy comprehensive evaluation result vector is (0.32, 0.29, 0.2, 0.14, 0.05) According to the fuzzy comprehensive evaluation result vector, the maximum membership principle is adopted to determine the risk correction value of the construction step. The maximum membership principle means selecting the comment corresponding to the maximum value in the fuzzy comprehensive evaluation result vector as the evaluation result, and determining the final risk correction value based on the correspondence between the pre-set comments and the risk correction value.

[0086] According to the maximum membership principle, 0.32 is the largest value in the fuzzy comprehensive evaluation result vector, corresponding to the low risk correction. Referring to the pre-set corresponding relationship, the risk correction value for this step is determined to be -1 (a negative sign indicates a reduced risk).

[0087] The construction steps in the construction path and their risk correction values ​​are shown in Table 1.

[0088] Table 1 Schematic diagram of construction steps and their risk correction values ​​in the construction path Repair steps Risk Modified Value Preparation before maintenance -0.5 Transformer power outage operation -1 Transformer oil draining and internal inspection 0 Winding insulation treatment 1 Transformer oil return and seal inspection 0.5 Transformer power transmission test 1.5 The sum of the first data and the risk correction value is taken as the final risk value of each construction step, as shown in Table 2.

[0089] Table 2 Schematic diagram of construction steps and their final risk values ​​in the construction path Repair steps Risk Modified Value Initial risk value Terminal Value at Risk Preparation before maintenance -0.5 2 1.5 Transformer power outage operation -1 8 7 Transformer oil draining and internal inspection 0 3 3 Winding insulation treatment 1 4 5 Transformer oil return and seal inspection 0.5 5 5.5 Transformer power transmission test 1.5 7 8.5 When the final risk value is greater than the preset threshold, an alarm signal is sent to the construction manager and the S22 model check is performed.

[0090] S22 model verification, obtain the construction steps included in the construction path, record them as target steps, and trace back the construction links associated with the target steps in the monomer model after the associated construction plan, record them as target links.

[0091] Traverse the monomer model after the associated construction plan to determine whether the association relationship between the target step and the target link is incorrect. If so, it means that the monomer model cannot accurately reflect the actual construction situation. Correct the monomer model after the associated construction plan, record the corrected monomer model as the new monomer model, and execute S13 allocation; if not, send a signal to stop construction.

[0092] S23 resource conflict judgment collects the location, status and available time data of resources required for construction in real time to obtain the dynamic situation of resources required for construction.

[0093] Based on the new single-unit model, a greedy algorithm is used to detect resource conflicts between target steps. This algorithm gradually approaches the global optimal solution by selecting the optimal solution in the current state at each step. During resource conflict detection, the algorithm quickly identifies any temporal overlap in the demand for the same resource between target steps, thereby determining whether a resource conflict exists. If a resource conflict exists between target steps, S24 is executed to determine the execution time.

[0094] S24 performs time judgment to determine whether there is an execution order requirement between the conflicting target steps. If so, the resources required for construction are allocated according to the execution order; if not, S25 is executed to calculate the priority score.

[0095] S25 calculates the priority score and uses the hierarchical analysis method to assign corresponding weights to the final risk value, number of affected users and construction duration of each target step.

[0096] The hierarchical structure of the AHP is as follows: Goal layer: Determine the priority score of each goal step.

[0097] Criteria layer: includes three evaluation indicators - final risk value, number of affected users, and construction time.

[0098] Solution level: all target steps that need to be evaluated.

[0099] The 1-9 scale method was used to compare the relative importance of the three evaluation indicators, and the judgment matrix was constructed as shown in Table 3.

[0100] Table 3 Judgment Matrix index Terminal Value at Risk Number of affected users Construction time Terminal Value at Risk 1 3 5 Number of affected users 1 / 3 1 2 Construction time 1 / 5 1 / 2 1 When calculating the weight, you can first calculate the geometric mean of the elements in each row of the judgment matrix, and then normalize the geometric mean to obtain the weight of each indicator. The calculation process of the geometric mean of the final risk value is as follows: ; Similarly, the geometric mean of the number of affected users is approximately 0.87, and the geometric mean of construction duration is approximately 0.46.

[0101] After normalization, the weight of the final risk value is approximately 0.65, the weight of the number of affected users is approximately 0.23, and the weight of the construction duration is approximately 0.23.

[0102] The priority score of each target step is calculated based on the final risk value, number of affected users, construction duration and corresponding weight of each target step, and the resources required for construction are allocated according to the priority score.

[0103] Obtain the indicator values ​​of each target step, then normalize the various indicator values ​​according to categories to obtain normalized results. The priority score of each target step is equal to the sum of the products of each weight and the normalized results of each indicator.

[0104] By adopting the above solution, this embodiment realizes the setting of resource priority allocation before construction. This embodiment also realizes alarm by combining equipment parameters and environmental data.

[0105] Example 3: Reference Figure 3 The difference between this embodiment and embodiment 1 is that it further includes: S31 adds tags and obtains basic information of construction personnel (name, work number, department, etc.) from the human resources system (HRMS).

[0106] The skills of each construction worker are obtained through questionnaires, skill certification records, historical project logs and other channels, and skill tags are added to each construction worker. The skill tags include professional fields, qualification levels, practical experience and historical construction scores.

[0107] S32 sets the construction manager, extracts structured requirements (such as "must have first-level firefighting qualifications") from construction plan documents (such as PDF, Word), and uses regular expressions and template matching technology to parse unstructured text (such as "must be familiar with super high-rise building construction technology").

[0108] The degree of match between construction personnel and construction requirements is obtained through label matching algorithms (such as the VSM model) and skill labels. The VSM model converts construction requirements (such as qualifications and years of experience) into feature vectors, calculates cosine similarity with the construction personnel label vector, and then designates the construction personnel with the highest degree of match as the construction manager.

[0109] S33 verifies the construction plan and determines whether there are identical construction projects in the monomer model after the associated construction plan. If the project name, construction scope, and time interval completely overlap, they are determined to be identical. If so, S34 verifies the person in charge; if not, no processing is performed.

[0110] S34 verifies the person in charge, compares the construction person in charge of the duplicate projects and their work numbers, and determines whether the construction persons in charge are the same. If so, execute S35 to delete; if not, execute S36 to desensitize.

[0111] S35 deletes any one of the same construction projects.

[0112] S36 desensitization was performed to obtain documents related to historical construction plans. Tesseract OCR was used to extract text from the documents. Apache POI was used to parse the structured tables in the documents. A semantic library of construction terminology was constructed using NLP technology, including: We built an empty semantic library of construction terminology, extracted structured and unstructured data from historical construction plans, and used Stanford CoreNLP to perform dependency parsing on the unstructured text, breaking it down into "subject-predicate-object" semantic units (e.g., "prefabricated components" → "construction method: prefabrication"), thus splitting the unstructured data into multiple semantic units.

[0113] The Word2Vec model is used to convert semantic units into semantic vectors, and the similarity between semantic vectors is calculated. Any semantic unit with a similarity greater than a preset similarity threshold is deleted to obtain the remaining semantic units.

[0114] Fill the structured data and remaining semantic units into the construction terminology semantic library.

[0115] In the construction term semantic database, structured data and synonyms or near synonyms of semantic units are added to obtain a new construction term semantic database.

[0116] The construction plan is desensitized based on the new construction terminology semantic library and the construction path of each construction manager. The process is as follows: A three-level desensitization rule base is established, and the structure of the desensitization rule base is as follows: Basic layer: regular expression matching (ID number, contact information, etc.) Domain layer: Construction term replacement (e.g. "deep foundation pit" → "[project type]") Business layer: project-specific sensitive information filtering A sliding window algorithm is used to implement context-aware desensitization to avoid semantic discontinuity caused by partial replacement. For example, "Zhang San (project manager)" is replaced as a whole with "[Construction Manager]".

[0117] Based on the construction workers' access rights to construction projects in the S12 model, the visible range of each construction worker is obtained. The following solutions can be used for desensitization: Word-level desensitization: directly replace sensitive words; Sentence-level desensitization: retain key construction actions and hide specific parameters; Segment-level desensitization: shielding technical solutions that need to be kept confidential; Finally, the desensitized construction plan is sent to the corresponding construction manager.

[0118] By adopting the above solution, this embodiment achieves precision, automation and security in construction plan management.

[0119] Embodiment 4: This embodiment discloses a distribution network equipment operation data event correlation analysis system, the system comprising: a memory and a processor, The memory stores a computer-readable storage medium; When the processor processes the computer program stored on the computer-readable storage medium, it implements the distribution network equipment operation data event correlation analysis method.

[0120] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for analyzing distribution network equipment operation data events, characterized in that: include: Data collection: Obtain the operating data of each device and set a corresponding construction plan based on the operating data of each device. The construction plan includes the construction project, construction time, resources required for construction, and construction project location; Modeling: Set up access permissions for construction projects for each construction worker, build a table of accessible construction projects based on the access permissions, and build a single model based on the table of accessible construction projects. The single model includes: a top layer for representing construction projects, a sub-layer for representing construction links, and a bottom layer for representing construction steps; Allocation: Associate the construction plan in the single model, obtain the construction path based on the single model after the construction plan is associated, and send the construction path to the construction person in charge within the preset time before the construction time.

2. The method for analyzing distribution network equipment operation data events according to claim 1, characterized in that: The method further comprises: An initial risk value is set for each construction step, and the initial risk value of the construction step included in the construction path is recorded as the first data. Equipment parameters and environmental data are collected, and the fuzzy comprehensive evaluation method is used to determine the risk correction value of each construction step based on the equipment parameters and environmental data. The sum of the first data and the risk correction value is used as the final risk value of each construction step. When the final risk value is greater than the preset threshold, an alarm signal is sent to the construction manager.

3. The method for analyzing distribution network equipment operation data events according to claim 2, characterized in that: After issuing the alarm signal, the method further includes: Obtain the construction steps included in the construction path, record them as target steps, and trace back the construction links associated with the target steps in the monomer model after the construction plan is associated, record them as target links; Determine whether the association between the target step and the target link is correct. If so, revise the monomer model after the associated construction plan, record the revised monomer model as the new monomer model, and execute the assigned steps; if not, issue a signal to stop construction.

4. The method for analyzing distribution network equipment operation data events according to claim 3, characterized in that: The method further comprises: The location, status, and available time data of resources required for construction are collected in real time. Based on the new monomer model, a greedy algorithm is used to detect resource conflicts between target steps. If resource conflicts exist between target steps, the priority scoring step is executed. Calculate priority scores: Use the analytic hierarchy process to assign corresponding weights to the final risk value, number of affected users, and construction duration of each target step. Calculate the priority score of each target step based on the final risk value, number of affected users, construction duration, and corresponding weights. Allocate resources required for construction based on the priority score.

5. The method for analyzing distribution network equipment operation data events according to claim 4, characterized in that: Before executing the step of calculating the priority score, the method further includes: Determine whether there is an execution order requirement between the conflicting target steps. If so, allocate the required construction resources according to the execution order; if not, execute the step of calculating the priority score.

6. The method for analyzing distribution network equipment operation data events according to any one of claims 1 to 5, characterized in that: Before sending the construction route to the construction person in charge within a preset time before the construction start, the method further includes: Obtain a list of construction personnel and add skill tags to each construction personnel, including professional field, qualification level, practical experience and historical construction score; Construction requirements are extracted based on the description in the construction plan. The degree of match between construction personnel and construction requirements is obtained through label matching algorithms and skill labels. The construction personnel with the highest degree of match are designated as construction supervisors.

7. The method for analyzing distribution network equipment operation data events according to claim 6, characterized in that: After setting the construction worker with the highest matching degree as the construction manager, the method further includes: Verify the construction plan: Determine whether the same construction items exist in the single model after the construction plan is associated. If so, execute the steps of the verification person in charge; if not, do nothing. Verify the person in charge: Determine whether the construction person in charge is the same. If so, perform the deletion step; if not, perform the desensitization step; Delete: Delete any construction project in the same construction project; Desensitization: Obtain historical construction plans, build a construction terminology semantic library based on the historical construction plans using NLP technology, desensitize the construction plans based on the construction terminology semantic library and the construction path of each construction manager, and send the desensitized construction plans to the corresponding construction managers.

8. The method for analyzing distribution network equipment operation data events according to claim 7, characterized in that: Based on historical construction plans, a semantic library of construction terminology is constructed using NLP technology, including: Build an empty semantic library of construction terms, extract structured and unstructured data from historical construction plans, and perform syntactic analysis on the unstructured data to split it into multiple semantic units. The Word2Vec model is used to convert semantic units into semantic vectors, and the similarity between semantic vectors is calculated. Any semantic unit with a similarity greater than a preset similarity threshold is deleted to obtain the remaining semantic units. Fill the structured data and remaining semantic units into the construction terminology semantic library.

9. The method for analyzing distribution network equipment operation data events according to claim 8, characterized in that: The method further includes: adding structured data and synonyms or near synonyms of semantic units to the construction term semantic library to obtain a new construction term semantic library.

10. A distribution network equipment operation data event correlation analysis system, characterized in that: include: memory and processor, The memory stores a computer-readable storage medium; When the processor processes the computer program stored on the computer-readable storage medium, the method according to any one of claims 1 to 9 is implemented.

Images