A power transmission and transformation project construction monitoring method and system

Abstract

This invention discloses a method and system for monitoring the construction of power transmission and transformation projects, applied in the field of power transmission and transformation project construction management technology. The method includes: acquiring engineering instruction information for the power transmission and transformation project to be constructed; determining various initial construction sub-schemes based on the engineering instruction information to form an equipment-result association mapping diagram; performing graph embedding encoding on the equipment-result association mapping diagram to obtain various embedding vectors; determining the target construction time series based on construction event information; processing the target construction time series and embedding vectors based on a pre-built SAHP model to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed; correcting the target construction scheme based on the green assessment index of the distribution network; and performing real-time monitoring based on the corrected target construction scheme during the actual construction process of the power transmission and transformation project to be constructed. The method of this invention can achieve precise monitoring of the construction progress of large and complex power transmission and transformation projects.

Description

Technical Field

[0001] This invention relates to the field of power transmission and transformation project construction management technology, and in particular to a method and system for monitoring the construction of power transmission and transformation projects. Background Technology

[0002] The construction progress of power transmission and transformation projects is directly related to regional economic development, energy security, and the timely consumption of clean energy. Therefore, accurate and effective monitoring of construction progress is the foundation for optimizing social resources.

[0003] In existing technologies, project managers break down the entire project into a series of specific construction procedures based on their experience, and estimate the construction period for each procedure. They periodically compare the actual progress of each procedure with the planned progress and adjust the corresponding construction period accordingly. However, the complex and ever-changing construction site environment can easily lead to changes in the logic and progress between procedures. Furthermore, this monitoring method is a passive, post-hoc record; deviations can only be detected by comparing the actual progress with the planned progress. Therefore, it cannot provide forward-looking predictions of construction progress and cannot meet the high standards of precision and green development required for progress control in modern large-scale and complex power transmission and transformation projects. Summary of the Invention

[0004] This invention provides a method and system for monitoring the construction progress of power transmission and transformation projects, in order to solve the technical problem that manual periodic inspections cannot make forward-looking predictions of the construction progress, and to achieve the effect of precise monitoring of the construction progress of large and complex power transmission and transformation projects.

[0005] To address the aforementioned technical problems, this invention provides a method for monitoring the construction progress of power transmission and transformation projects, comprising: Obtain engineering instructions for the power transmission and transformation project to be constructed; Based on the engineering instruction information, each initial construction sub-plan is determined, and the construction equipment data in each initial construction sub-plan is assigned a corresponding prediction result label to form an equipment-result association mapping diagram. The device-result association mapping graph is subjected to graph embedding encoding to obtain each embedding vector; Extract construction event information from selected historical construction projects, and determine the target construction time sequence based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. Based on the pre-built SAHP model, the target construction time series and the embedding vector are processed to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. Based on the green assessment indicators of the power distribution network, the target construction plan was revised. During the actual construction of the power transmission and transformation project to be constructed, real-time monitoring is carried out based on the revised target construction plan.

[0006] As one preferred embodiment, the step of determining each initial construction sub-scheme based on the engineering instruction information, assigning corresponding prediction result labels to the construction equipment data in each initial construction sub-scheme, and forming an equipment-result association mapping diagram includes: Extract the spatial topology of the power transmission and transformation project to be constructed from the engineering instruction information; Based on the spatial topology, the power transmission and transformation project to be constructed is divided into multiple construction sub-regions, and each construction sub-region corresponds to an initial construction sub-scheme. Assign a corresponding prediction result label to the construction equipment data in each of the initial construction sub-schemes; Using the construction equipment data as nodes and the prediction result labels as node attributes, a device-result association mapping graph is established.

[0007] As one preferred embodiment, the step of performing graph embedding encoding on the device-result association mapping graph to obtain various embedding vectors includes: Based on the graph neural network model, local neighborhood feature analysis is performed on the device-result association mapping graph to obtain the initial embedding vector. The initial embedding vector is subjected to global feature integration processing to obtain each embedding vector corresponding to the device-result association mapping graph.

[0008] As one preferred embodiment, the step of extracting construction event information from selected historical construction projects and determining the target construction time series based on the construction event information includes: Based at least on the project type of the power transmission and transformation project to be constructed, the historical project database is filtered to obtain the selected historical construction projects; Extract the construction event information corresponding to the power transmission and transformation project to be constructed from the project management logs of the historical construction projects; Based on timestamp information, the construction event information is time-coded to obtain the target construction time sequence.

[0009] As one preferred embodiment, the process of processing the target construction time series and the embedding vector based on the pre-built SAHP model to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed includes: The target construction time series is fused with the embedded vector to obtain the construction event representation result; A multi-head attention mechanism is introduced into the SAHP model to perform temporal dependency modeling on the construction event representation results, thereby obtaining a comprehensive attention representation result. Based on the event intensity function, a nonlinear transformation is performed on the comprehensive attention representation result to obtain the event intensity value of each event to be constructed in the time dimension. Based on the intensity value of each event, the corresponding events to be constructed are associated and mapped to obtain the target construction plan corresponding to the power transmission and transformation project to be constructed.

[0010] Another aspect of the present invention provides a construction monitoring system for power transmission and transformation projects, comprising: The instruction information acquisition module is used to acquire engineering instruction information for the power transmission and transformation project to be constructed. The association mapping graph construction module is used to determine each initial construction sub-scheme based on the engineering instruction information, assign corresponding prediction result labels to the construction equipment data in each initial construction sub-scheme, and form an equipment-result association mapping graph. The graph embedding encoding module is used to perform graph embedding encoding on the device-result association mapping graph to obtain various embedding vectors; The construction time series determination module is used to extract construction event information from selected historical construction projects and determine the target construction time series based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. The construction scheme acquisition module is used to process the target construction time series and the embedding vector based on the pre-built SAHP model to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. The construction plan correction module is used to correct the target construction plan based on the green assessment indicators of the power distribution network. The power transmission and transformation project monitoring module is used to perform real-time monitoring based on the modified target construction plan during the actual construction process of the power transmission and transformation project to be constructed.

[0011] As one preferred embodiment, the association mapping graph construction module includes: A spatial topology extraction unit is used to extract the spatial topology of the power transmission and transformation project to be constructed from the engineering indication information. The sub-scheme division unit is used to divide the power transmission and transformation project to be constructed into multiple construction sub-regions based on the spatial topology relationship, and each construction sub-region corresponds to an initial construction sub-scheme. The prediction result labeling unit is used to assign a corresponding prediction result label to the construction equipment data in each of the initial construction sub-schemes; The association mapping graph construction unit is used to establish the equipment-result association mapping graph with the construction equipment data as nodes and the prediction result labels as node attributes.

[0012] As one preferred embodiment, the graph embedding encoding module includes: The initial embedding vector unit is used to perform local neighborhood feature analysis on the device-result association mapping graph based on the graph neural network model to obtain the initial embedding vector. The feature integration processing unit is used to perform global feature integration processing on the initial embedding vector to obtain each of the embedding vectors corresponding to the device-result association mapping.

[0013] As one preferred embodiment, the construction time sequence determination module includes: The construction project screening unit is used to screen the historical project database based at least on the project type of the power transmission and transformation project to be constructed, and to obtain the selected historical construction projects. The construction event information extraction unit is used to extract the construction event information corresponding to the power transmission and transformation project to be constructed from the project management log of the historical construction projects; The time encoding processing unit is used to perform time encoding processing on the construction event information based on timestamp information to obtain the target construction time sequence.

[0014] As one preferred embodiment, the construction plan acquisition module includes: The feature fusion unit is used to fuse the target construction time series with the embedded vector to obtain a construction event representation result. The attention mechanism unit is used to introduce a multi-head attention mechanism into the SAHP model, perform time-series dependency modeling on the construction event representation results, and obtain a comprehensive attention representation result. The nonlinear transformation unit is used to perform a nonlinear transformation on the comprehensive attention representation result based on the event intensity function to obtain the event intensity value of each event to be constructed in the time dimension. The association mapping processing unit is used to perform association mapping processing on the corresponding events to be constructed based on each event intensity value, so as to obtain the target construction plan corresponding to the power transmission and transformation project to be constructed.

[0015] Compared to existing technologies, the beneficial effects of the embodiments of the present invention are at least one of the following: The present invention constructs a device-result association mapping graph and performs graph embedding encoding, enabling subsequent construction progress prediction to be based on solid engineering physical information, greatly improving the rationality and credibility of the prediction results; The present invention utilizes the SAHP model to perform feature encoding and fusion of the target construction time series and embedding vector, which can accurately capture the complex nonlinear dependencies in the construction process, thereby achieving advanced and accurate prediction of progress dynamics; The present invention corrects the target construction plan through green evaluation indicators, ensuring the accuracy of the construction progress prediction results, providing project managers with accurate and reliable decision-making basis, and ultimately realizing intelligent, refined, and green closed-loop monitoring of the construction progress of power transmission and transformation projects. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a construction monitoring method for power transmission and transformation projects in one embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a power transmission and transformation engineering construction monitoring system in one embodiment of the present invention; Figure label: Among them, 11. Indication information acquisition module; 12. Association mapping graph construction module; 13. Graph embedding encoding module; 14. Construction time series determination module; 15. Construction plan acquisition module; 16. Construction plan correction module; 17. Power transmission and transformation project monitoring module. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] In existing technologies, project managers typically rely on personal experience to divide power transmission and transformation projects into several specific construction procedures, such as foundation excavation, concrete pouring, tower erection, and conductor laying. They manually estimate the required construction period for each procedure, forming a preliminary schedule. During construction, managers periodically collect data on the actual completion status of each procedure and compare it with the original planned schedule. If deviations are found, corrections are made by adjusting resource allocation for subsequent procedures or extending working hours. However, the construction site environment for power transmission and transformation projects is complex and variable, often affected by multiple factors such as strong winds, rain, smog, traffic restrictions, environmental regulations, and delays in equipment and material supply. These disturbances not only directly affect the execution efficiency of individual procedures but also... This monitoring method can trigger a chain reaction of delays due to the dependencies between work processes, causing the original plan to quickly become invalid. More importantly, this monitoring method is essentially a post-event recording management system. Deviations can only be identified after actual progress data is collected and compared, making it impossible to provide early warnings or interventions before deviations occur, and lacking forward-looking predictive capabilities. At the same time, this method does not incorporate green construction indicators such as energy consumption control and carbon emission constraints into the generation and adjustment process of the schedule plan. It often ignores environmental protection requirements when pursuing the schedule target, making it difficult to achieve synergistic optimization of schedule and green performance. Therefore, facing the high standards of accuracy, dynamic adaptability and greenness required for schedule control in modern large-scale and complex power transmission and transformation projects, traditional monitoring methods that rely on manual experience and passive feedback are clearly insufficient.

[0022] One embodiment of the present invention provides a method for monitoring the construction of power transmission and transformation projects. For details, please refer to [link / reference]. Figure 1 , Figure 1 The diagram shown is a flowchart illustrating a construction monitoring method for power transmission and transformation projects according to one embodiment of the present invention. The method includes steps S1 to S7: S1. Obtain engineering instruction information for the power transmission and transformation project to be constructed; S2. Based on the engineering instruction information, determine each initial construction sub-plan, assign corresponding prediction result labels to the construction equipment data in each initial construction sub-plan, and form an equipment-result association mapping diagram. S3. Perform graph embedding encoding on the device-result association mapping graph to obtain each embedding vector; S4. Extract construction event information from the selected historical construction projects, and determine the target construction time sequence based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. S5. Based on the pre-built SAHP model, the target construction time series and the embedding vector are processed to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. S6. Based on the green assessment indicators of the power distribution network, the target construction plan is revised; S7. During the actual construction of the power transmission and transformation project to be constructed, real-time monitoring is carried out based on the modified target construction plan.

[0023] Furthermore, in step S1, obtaining the engineering instruction information of the power transmission and transformation project to be constructed is to fully understand the structural characteristics, construction logic, and resource requirements of the project. Specifically, multi-source engineering information is extracted from the BIM (Building Information Modeling) model, including component geometric information, process logic relationships, and resource consumption parameters. The component geometric information includes the location of the transmission tower, foundation dimensions, and spatial layout of the substation structure. The process logic relationships describe the sequential dependencies, such as the tower erection only after the foundation is poured. The resource consumption parameters cover data such as steel consumption, equipment power, and manpower allocation.

[0024] The advantage of this step is that it provides a structured and multi-dimensional input basis for the subsequent construction of equipment result association mapping diagrams, generation of initial construction sub-schemes, and integration of green constraints, thereby ensuring that the construction progress prediction not only conforms to the actual logic of the project, but also supports the dynamic monitoring needs of greening and intelligence.

[0025] Furthermore, in step S2, determining each initial construction sub-scheme based on the engineering instruction information and assigning prediction result labels to the construction equipment data within it is to decompose the overall project into executable and evaluable local construction units and establish a quantitative correlation between equipment operation behavior and its expected results. Specifically, by dividing the construction into segments based on the spatial distribution of components and process dependencies, matching the corresponding construction equipment type and its operation tasks for each sub-scheme, and then combining the standard ergonomics library, equipment rated power, and carbon emission factors to calculate the expected completion time, energy consumption, and carbon emissions of each piece of equipment under that sub-scheme, these are used as prediction result labels and bound to the corresponding equipment, thereby constructing an equipment result association mapping graph with equipment as nodes, prediction labels as attributes, and collaborative or adjacency relationships as edges.

[0026] The advantage of this step is that it structurally links discrete equipment resources with construction goals, providing subsequent graph embedding encoding and deep learning models with fusion features rich in process logic, resource consumption and green indicators, which significantly improves the rationality, interpretability and green constraint embedding ability of construction progress prediction.

[0027] Further, in step S3, in order to transform the equipment-result association mapping graph into a low-dimensional vector representation suitable for subsequent deep learning model processing, graph embedding encoding of the graph is required. First, the importance of each construction device and its interrelationships in the project is understood, and this goal is achieved by constructing a graph neural network model. Specifically, each construction device is extracted as a node from the equipment-result association mapping graph, and its corresponding predicted result label (such as expected completion time, energy consumption value, and carbon emissions) is input as node attributes into the graph neural network model; the information of each node and its neighboring nodes is aggregated using graph convolutional layers to generate initial embedding vectors; then, global feature information is further integrated through multi-layer graph convolution or by introducing an attention mechanism, so that the embedding vector of each node not only contains its own attributes but also incorporates its contextual information in the network; finally, the node embedding vectors are adjusted through nonlinear transformations such as fully connected layers to obtain the final embedding vectors.

[0028] This approach effectively captures the complex dependencies and synergistic effects among construction equipment. Secondly, the embedding vectors provide a compact and rich feature representation, facilitating efficient computation during subsequent model training and inference. Thirdly, graph embedding encoding seamlessly integrates structured information with time-series data, providing high-quality input for construction progress prediction based on the SAHP (Self-Attentive Hawkes Process) model, ensuring that the prediction results conform to both actual construction logic and green constraints. Furthermore, this representation method enhances the model's adaptability to dynamically changing environments, allowing for real-time updates to equipment status and adjustments to construction plans, thereby improving the overall intelligence level and decision-making accuracy of project management.

[0029] Furthermore, in step S4, in order to extract construction event information from the selected historical construction projects and determine the target construction time series based on this information, it is first necessary to clarify that the selected historical construction projects are of the same type as the power transmission and transformation project to be constructed. This is mainly because similar types of projects share similarities in construction processes, logical relationships between procedures, resource consumption patterns, and potential problems. In this way, actual data from past projects can be used to guide the progress forecasting and management of new projects, improving the accuracy and rationality of forecasts.

[0030] Specifically, first, detailed records of selected historical construction projects are collected and organized, including but not limited to key information such as the start and end times of each process, required resources, and involved equipment and personnel. Then, this information is analyzed and processed to extract content related to construction events, such as the execution status of specific processes, unexpected events that occurred during the process, and their resolution. Next, time-series data is constructed based on the extracted construction event information. This typically involves arranging each construction event in chronological order and labeling the specific time nodes and duration of each event. Ultimately, the resulting target construction time series not only includes details at the process level but also reflects various dynamic changes that may occur throughout the entire construction process.

[0031] Furthermore, in step S5, in order to achieve forward-looking prediction and green management of the construction progress of power transmission and transformation projects, it is necessary to jointly process the target construction time series and embedding vector based on the pre-built SAHP model to generate a scientific and reasonable target construction plan. Since relying solely on historical time series cannot reflect the structural characteristics and resource constraints of the current project, and relying solely on static graph embedding lacks the ability to model the dynamic evolution of construction events, only by integrating the two into the SAHP model with time series modeling capabilities can we simultaneously capture the logical dependencies between processes, historical construction patterns, and the impact of green constraints, thereby outputting a schedule that conforms to the actual project.

[0032] Specifically, firstly, the target construction time series extracted and sorted from historical projects is input into the time encoding layer of the SAHP model, transforming it into vectorized time features containing timestamps and time interval information. Simultaneously, the embedding vector obtained by graph embedding encoding of the equipment result association mapping graph is used as the structural prior information of the current project. Subsequently, in the multi-head attention layer of the SAHP model, the time features and embedding vectors are fused, and a comprehensive attention representation result is obtained by calculating the dynamic attention weights between events. The event intensity value in the continuous time dimension is generated by using the dual-channel structural modeling process trigger probability of the event intensity function. Finally, the event intensity value is unfolded over time, and the expected start and completion times of each process are determined by identifying the intensity peak or integral threshold, forming a target construction plan that includes process sequence, time nodes, and resource arrangements.

[0033] Preferably, the dual-channel structure includes a progress channel and a green channel. The result of the comprehensive attention representation is input into the progress channel of the event intensity function. In the progress channel, an event triggering probability distribution is established through nonlinear transformation, and the process triggering probability vector is output. The specific scoring formula for the progress channel is as follows: in, The integrated attention representation sequence represents the sequence at time step. Vector features; The weight matrix represents the progress channel and consists of learnable parameters; The bias vector representing the progress channel is a learnable parameter. This represents the process trigger probability vector, indicating the probability at time step [missing information]. Trigger distribution for each process type; This indicates the number of process event categories.

[0034] The scoring formula for the green channel is as follows: in, This indicates the set of green constraint parameters at time step. The constraint vector is obtained by mapping carbon emission budget, energy consumption limit, and safety regulations; The mapping matrix from the set of constraint parameters to the green representation space represents the learnable parameters; Represents the bias term of the mapping, a learnable parameter; Represents the element-wise nonlinear function ReLU, used to enhance the nonlinear characterization capability of green constraints; The green constraint representation vector represents the vector at time step [0, 1]. Energy consumption and carbon emission constraints; The weight matrix representing the green channel consists of learnable parameters. The bias vector representing the green channel is a learnable parameter. The green constraint correction vector represents the result of the constraint condition correcting the event intensity.

[0035] Furthermore, in step S6, to ensure that the target construction plan meets the requirements of green construction while satisfying the schedule and procedural logic, it needs to be revised based on the green assessment indicators of the power distribution network. This is because although the initially generated construction plan considers historical construction patterns and engineering structural characteristics, it may exceed environmental constraints in terms of energy consumption or total carbon emissions. Especially under the current dual-carbon target context, power transmission and transformation engineering construction must balance schedule efficiency and green performance to avoid situations where excessive use of high-energy-consuming equipment or concentrated operations lead to localized carbon emission exceedances in order to meet deadlines.

[0036] Specifically, the process begins by acquiring green assessment indicators for the power distribution network, including the upper limit of energy consumption per unit project, the total budget for carbon emissions, and relevant thresholds in construction safety and environmental protection regulations. These indicators are then combined with time-series resource consumption data to establish sliding window constraints over time, such as ensuring that cumulative carbon emissions within each seven-day period do not exceed a preset limit. Next, the projected energy consumption and carbon emissions of each process in the target construction plan are compared with these constraints to identify time periods or processes that violate green indicators. Based on this, an optimization problem is constructed with the goal of minimizing schedule deviations and satisfying green constraints. The start time, duration, or equipment configuration of processes is iteratively adjusted through a differentiable optimization layer. For example, high-energy-consuming processes are scheduled to run during clean energy power supply periods, or intensive operations are broken down into segments to reduce instantaneous load. Correction signals generated during the optimization process are also fed back to the event intensity function and attention layer of the SAHP model, dynamically updating the prediction results until a corrected target construction plan that conforms to both construction logic and green constraints is output.

[0037] Preferably, the energy consumption constraints and carbon emission constraints in the green assessment indicators are converted into modulation factors, wherein the expression for the modulation factor is: Where M represents the modulation factor and E represents the energy consumption constraint value. This represents the upper limit of the energy consumption budget, and C represents the carbon emission constraint value. Indicates the carbon emissions budget ceiling. This represents the energy-consuming modulation weighting coefficient. This represents the carbon emission modulation weighting coefficient.

[0038] Preferably, the purpose of writing back the correction signal during the optimization process to the event intensity function and the multi-head attention layer via a chain-like derivative path is to update the model parameters. The backpropagation process updates the parameters by calculating the gradient of the loss function with respect to the model parameters. Assuming the event intensity function is E and the mapping of the multi-head attention layer is A, the update process is as follows: in, These are the parameters of the model. It is a loss function. and These are the outputs of the event intensity function and the multi-head attention layer, respectively.

[0039] In this way, the signal is corrected. It is incorporated into the backpropagation calculation to help adjust the event intensity sequence to better align with the target schedule and meet green assessment metrics.

[0040] In each iteration, the mapping parameters of the dual-channel weights and multi-head attention are updated via a chained derivative path. Assume the weights of the progress channel are... The weight of the green channel is And the mapping of the multi-head attention layer as The update rule is as follows: in, It's the learning rate. It is the gradient of the corresponding parameters. The parameters of each layer are updated through backpropagation, which ultimately makes the event intensity sequence more consistent with the corrected progress variable and meets the green evaluation index.

[0041] Furthermore, in step S7, real-time monitoring is conducted based on the revised target construction plan during the actual construction process of the power transmission and transformation project to be constructed. This is to ensure that the intelligent prediction results are effectively implemented in on-site management and to achieve closed-loop control from planning to execution. Specifically, the revised target construction plan imports the planned time nodes, equipment configuration requirements, and green indicator thresholds for each process into the digital monitoring platform at the construction site. Then, real-time construction data is collected via IoT sensors, mobile terminal data entry, video recognition, or the project management system. This includes the actual start and end times of each process, equipment operating status, energy consumption, and estimated carbon emissions. The system compares this real-time data with the corresponding planned items in the revised plan, calculating the progress deviation rate and the compliance with green indicators. When the actual progress of a process lags behind the preset tolerance range, or when energy consumption and carbon emissions within a certain time period approach or exceed the upper limit of the green assessment indicators, the system automatically generates dynamic correction prompts. These prompts include suggested adjustments to the sequence of subsequent processes, alternative low-energy-consuming equipment types, suggestions for optimizing work periods, or resource reallocation strategies. Simultaneously, this deviation information can also be fed back to the SAHP model for online fine-tuning of event intensity predictions, supporting the rolling updates of the next phase of the construction plan.

[0042] This step enables a shift in construction management from passive response to proactive early warning, allowing for timely intervention in the early stages of deviations and preventing problems from accumulating and escalating. By incorporating green indicators into real-time monitoring, it ensures that low-carbon goals are integrated throughout the entire construction process rather than merely remaining at the planning level. Furthermore, it enables construction progress control to have continuous learning and adaptive capabilities, significantly improving the stability, resource utilization efficiency, and green construction level of large-scale power transmission and transformation projects in complex environments.

[0043] Another embodiment of the present invention provides a construction monitoring system for power transmission and transformation projects. For details, please refer to [link / reference needed]. Figure 2 , Figure 2The diagram illustrates a power transmission and transformation engineering construction monitoring system according to one embodiment of the present invention. The system includes: The instruction information acquisition module 11 is used to acquire the engineering instruction information of the power transmission and transformation project to be constructed; The association mapping diagram construction module 12 is used to determine each initial construction sub-scheme according to the engineering instruction information, assign corresponding prediction result labels to the construction equipment data in each initial construction sub-scheme, and form an equipment-result association mapping diagram. Graph embedding encoding module 13 is used to perform graph embedding encoding on the device-result association mapping graph to obtain various embedding vectors; The construction time series determination module 14 is used to extract construction event information from selected historical construction projects and determine the target construction time series based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. The construction scheme acquisition module 15 is used to process the target construction time series and the embedding vector based on the pre-built SAHP model to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. The construction plan correction module 16 is used to correct the target construction plan based on the green assessment indicators of the power distribution network. The power transmission and transformation project monitoring module 17 is used to perform real-time monitoring based on the modified target construction plan during the actual construction process of the power transmission and transformation project to be constructed.

[0044] Furthermore, in the above embodiments, the association mapping graph construction module includes: A spatial topology extraction unit is used to extract the spatial topology of the power transmission and transformation project to be constructed from the engineering indication information. The sub-scheme division unit is used to divide the power transmission and transformation project to be constructed into multiple construction sub-regions based on the spatial topology relationship, and each construction sub-region corresponds to an initial construction sub-scheme. The prediction result labeling unit is used to assign a corresponding prediction result label to the construction equipment data in each of the initial construction sub-schemes; The association mapping graph construction unit is used to establish the equipment-result association mapping graph with the construction equipment data as nodes and the prediction result labels as node attributes.

[0045] Furthermore, in the above embodiments, the graph embedding encoding module includes: The initial embedding vector unit is used to perform local neighborhood feature analysis on the device-result association mapping graph based on the graph neural network model to obtain the initial embedding vector. The feature integration processing unit is used to perform global feature integration processing on the initial embedding vector to obtain each of the embedding vectors corresponding to the device-result association mapping.

[0046] Furthermore, in the above embodiments, the construction time sequence determination module includes: The construction project screening unit is used to screen the historical project database based at least on the project type of the power transmission and transformation project to be constructed, and to obtain the selected historical construction projects. The construction event information extraction unit is used to extract the construction event information corresponding to the power transmission and transformation project to be constructed from the project management log of the historical construction projects; The time encoding processing unit is used to perform time encoding processing on the construction event information based on timestamp information to obtain the target construction time sequence.

[0047] Furthermore, in the above embodiments, the construction plan acquisition module includes: The feature fusion unit is used to fuse the target construction time series with the embedded vector to obtain a construction event representation result. The attention mechanism unit is used to introduce a multi-head attention mechanism into the SAHP model, perform time-series dependency modeling on the construction event representation results, and obtain a comprehensive attention representation result. The nonlinear transformation unit is used to perform a nonlinear transformation on the comprehensive attention representation result based on the event intensity function to obtain the event intensity value of each event to be constructed in the time dimension. The association mapping processing unit is used to perform association mapping processing on the corresponding events to be constructed based on each event intensity value, so as to obtain the target construction plan corresponding to the power transmission and transformation project to be constructed.

[0048] Compared to existing technologies, the beneficial effects of the embodiments of the present invention are at least one of the following: The present invention constructs a device-result association mapping graph and performs graph embedding encoding, enabling subsequent construction progress prediction to be based on solid engineering physical information, greatly improving the rationality and credibility of the prediction results; The present invention utilizes the SAHP model to perform feature encoding and fusion of the target construction time series and embedding vector, which can accurately capture the complex nonlinear dependencies in the construction process, thereby achieving advanced and accurate prediction of progress dynamics; The present invention corrects the target construction plan through green evaluation indicators, ensuring the accuracy of the construction progress prediction results, providing project managers with accurate and reliable decision-making basis, and ultimately realizing intelligent, refined, and green closed-loop monitoring of the construction progress of power transmission and transformation projects.

[0049] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A power transmission and distribution construction monitoring method, characterized by, include: Obtain engineering instructions for the power transmission and transformation project to be constructed; Based on the engineering instruction information, each initial construction sub-plan is determined, and the construction equipment data in each initial construction sub-plan is assigned a corresponding prediction result label to form an equipment-result association mapping diagram. The step of determining each initial construction sub-plan based on the engineering instruction information, assigning corresponding prediction result labels to the construction equipment data in each initial construction sub-plan, and forming an equipment-result association mapping diagram includes: Extract the spatial topology of the power transmission and transformation project to be constructed from the engineering instruction information; Based on the spatial topology, the power transmission and transformation project to be constructed is divided into multiple construction sub-regions, and each construction sub-region corresponds to an initial construction sub-scheme. Assign a corresponding prediction result label to the construction equipment data in each of the initial construction sub-schemes; Using the construction equipment data as nodes and the prediction result labels as node attributes, establish the equipment-result association mapping graph; The device-result association mapping graph is subjected to graph embedding encoding to obtain each embedding vector; Extract construction event information from selected historical construction projects, and determine the target construction time sequence based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. Based on the pre-built SAHP model, the target construction time series and the embedding vector are processed to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. The pre-built SAHP model is used to process the target construction time series and the embedding vector to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed, including: The target construction time series is fused with the embedded vector to obtain the construction event representation result; A multi-head attention mechanism is introduced into the SAHP model to perform temporal dependency modeling on the construction event representation results, thereby obtaining a comprehensive attention representation result. Based on the event intensity function, the comprehensive attention representation result is nonlinearly transformed to obtain the event intensity value of each event to be constructed in the time dimension. Based on the intensity value of each event, the corresponding events to be constructed are associated and mapped to obtain the target construction plan corresponding to the power transmission and transformation project to be constructed. Based on the green assessment indicators of the power distribution network, the target construction plan was revised. During the actual construction of the power transmission and transformation project to be constructed, real-time monitoring is carried out based on the revised target construction plan.

2. The power transmission line construction monitoring method of claim 1, wherein The graph embedding encoding of the device-result association mapping graph to obtain various embedding vectors includes: Based on the graph neural network model, local neighborhood feature analysis is performed on the device-result association mapping graph to obtain the initial embedding vector. The initial embedding vector is subjected to global feature integration processing to obtain each embedding vector corresponding to the device-result association mapping graph.

3. The power transmission line construction monitoring method of claim 1, wherein The step of extracting construction event information from selected historical construction projects and determining the target construction time series based on the construction event information includes: Based at least on the project type of the power transmission and transformation project to be constructed, the historical project database is filtered to obtain the selected historical construction projects; Extract the construction event information corresponding to the power transmission and transformation project to be constructed from the project management logs of the historical construction projects; Based on timestamp information, the construction event information is time-coded to obtain the target construction time sequence.

4. A power transmission line construction monitoring system characterized by comprising: include: The instruction information acquisition module is used to acquire engineering instruction information for the power transmission and transformation project to be constructed. The association mapping graph construction module is used to determine each initial construction sub-scheme based on the engineering instruction information, assign corresponding prediction result labels to the construction equipment data in each initial construction sub-scheme, and form an equipment-result association mapping graph. The step of determining each initial construction sub-plan based on the engineering instruction information, assigning corresponding prediction result labels to the construction equipment data in each initial construction sub-plan, and forming an equipment-result association mapping diagram includes: Extract the spatial topology of the power transmission and transformation project to be constructed from the engineering instruction information; Based on the spatial topology, the power transmission and transformation project to be constructed is divided into multiple construction sub-regions, and each construction sub-region corresponds to an initial construction sub-scheme. Assign a corresponding prediction result label to the construction equipment data in each of the initial construction sub-schemes; Using the construction equipment data as nodes and the prediction result labels as node attributes, establish the equipment-result association mapping graph; The graph embedding encoding module is used to perform graph embedding encoding on the device-result association mapping graph to obtain various embedding vectors; The construction time series determination module is used to extract construction event information from selected historical construction projects and determine the target construction time series based on the construction event information. The historical construction projects are of the same type as the power transmission and transformation project to be constructed. The construction scheme acquisition module is used to process the target construction time series and the embedding vector based on the pre-built SAHP model to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed. The pre-built SAHP model is used to process the target construction time series and the embedding vector to obtain the target construction scheme corresponding to the power transmission and transformation project to be constructed, including: The target construction time series is fused with the embedded vector to obtain the construction event representation result; A multi-head attention mechanism is introduced into the SAHP model to perform temporal dependency modeling on the construction event representation results, thereby obtaining a comprehensive attention representation result. Based on the event intensity function, the comprehensive attention representation result is nonlinearly transformed to obtain the event intensity value of each event to be constructed in the time dimension. Based on the intensity value of each event, the corresponding events to be constructed are associated and mapped to obtain the target construction plan corresponding to the power transmission and transformation project to be constructed. The construction plan correction module is used to correct the target construction plan based on the green assessment indicators of the power distribution network. The power transmission and transformation project monitoring module is used to perform real-time monitoring based on the modified target construction plan during the actual construction process of the power transmission and transformation project to be constructed.

5. The power transmission line construction monitoring system of claim 4, wherein, The graph embedding encoding module includes: The initial embedding vector unit is used to perform local neighborhood feature analysis on the device-result association mapping graph based on the graph neural network model to obtain the initial embedding vector. The feature integration processing unit is used to perform global feature integration processing on the initial embedding vector to obtain each of the embedding vectors corresponding to the device-result association mapping.

6. The power transmission line construction monitoring system of claim 4, wherein, The construction time sequence determination module includes: The construction project screening unit is used to screen the historical project database based at least on the project type of the power transmission and transformation project to be constructed, and to obtain the selected historical construction projects. The construction event information extraction unit is used to extract the construction event information corresponding to the power transmission and transformation project to be constructed from the project management log of the historical construction projects. The time encoding processing unit is used to perform time encoding processing on the construction event information based on timestamp information to obtain the target construction time sequence.

Images