Substation construction monitoring and early warning method and device based on three-dimensional model

Abstract

The present application provides a substation construction monitoring and early warning method and device based on a three-dimensional model, and relates to the field of substation early warning technology. The method includes: constructing a target three-dimensional model; obtaining target sensor information through a sensor monitoring device group; calling a digital preprocessing middle station to obtain target preprocessing sensor information; obtaining a target construction visualization model; reading a first predetermined sensor indicator set to obtain a first real-time indicator parameter set for a first monitoring category; obtaining first output data as input data for an integrated prediction model; when the first early warning index reaches the first early warning index limit, activating the target early warning device for abnormal early warning. The present application can solve the technical problems in the prior art of construction monitoring and early warning, such as poor early warning timeliness and accuracy, and inability to effectively prevent construction safety accidents, and achieves the technical effect of improving the timeliness and accuracy of abnormal early warning for substation construction, thereby ensuring the safety and reliability of substation construction.

Description

Technical Field

[0001] The present application relates to the field of substation early warning technology, and in particular to a substation construction monitoring and early warning method and device based on a three-dimensional model. Background Art

[0002] As the complexity and scale of power systems continue to expand, substations, as an important part of the power system, are crucial to the safe and stable operation of the entire power grid. In the field of substation construction monitoring and early warning, it is crucial to ensure safety and efficiency during the construction process. With the development of technology, traditional monitoring and early warning methods have gradually revealed their limitations, especially in processing large amounts of monitoring data. For example, although modern power systems have gradually been digitized, a large amount of sensor data requires efficient processing and analysis methods to quickly and accurately identify potential problems. At present, data processing and analysis methods are often not efficient enough to meet the needs of real-time monitoring and early warning.

[0003] In summary, in order to improve the efficiency and effectiveness of substation construction monitoring and early warning, it is necessary to develop a system that can perform efficient data processing and analysis on real-time construction monitoring information, so as to provide timely warning of construction anomalies. Summary of the invention

[0004] The purpose of this application is to provide a substation construction monitoring and early warning method and device based on a three-dimensional model, so as to solve the technical problems that the construction monitoring and early warning analysis in the prior art lacks integration and intelligence, cannot effectively extract key information from a large amount of monitoring data, and thus cannot identify potential risks and provide effective early warnings in a timely manner, resulting in threats to the safety of substation construction.

[0005] In view of the above problems, the present application provides a substation construction monitoring and early warning method and device based on a three-dimensional model.

[0006] In a first aspect, the present application provides a substation construction monitoring and early warning method based on a three-dimensional model, the method is implemented by a substation construction monitoring and early warning device based on a three-dimensional model, wherein the substation construction monitoring and early warning method based on a three-dimensional model includes: constructing a target three-dimensional model, the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing target point cloud data of a target substation based on the principle of random sampling consistency; performing dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; calling a digital preprocessing middle station to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information; rendering the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; reading a first predetermined sensor indicator set, and collecting data on the target construction visualization model based on the first predetermined sensor indicator set to obtain a first real-time indicator parameter set for a first monitoring category; using the first real-time indicator parameter set as input data of an integrated prediction model to obtain first output data, the first output data refers to a first warning index for the first monitoring category; when the first warning index reaches the first warning index limit, activating the target warning device to perform an abnormal warning on the first monitoring category of the target substation.

[0007] In the second aspect, the present application also provides a substation construction monitoring and early warning device based on a three-dimensional model, which is used to execute the substation construction monitoring and early warning method based on a three-dimensional model as described in the first aspect, wherein the substation construction monitoring and early warning device based on a three-dimensional model includes: a model construction module, the model construction module is used to construct a target three-dimensional model, and the target three-dimensional model refers to a three-dimensional model obtained by aligning and fusing the target point cloud data of the target substation based on the principle of random sampling consistency; a dynamic monitoring module, the dynamic monitoring module is used to perform dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; a preprocessing module, the preprocessing module is used to call a digital preprocessing middle station to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information; a visualization module The visualization module is used to render the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; a data collection module, the data collection module is used to read a first predetermined sensor indicator set, and based on the first predetermined sensor indicator set, collect data for the target construction visualization model to obtain a first real-time indicator parameter set for a first monitoring category; an early warning prediction module, the early warning prediction module is used to use the first real-time indicator parameter set as input data of an integrated prediction model to obtain first output data, and the first output data refers to a first early warning index for the first monitoring category; an early warning execution module, the early warning execution module is used to activate a target early warning device to issue an abnormal early warning for the first monitoring category of the target substation when the first early warning index reaches a first early warning index limit.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By constructing a target three-dimensional model, the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing the target point cloud data of the target substation based on the principle of random sampling consistency; performing dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; calling a digital preprocessing middle station to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information; rendering the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; reading a first predetermined sensor indicator set, and performing data collection on the target construction visualization model based on the first predetermined sensor indicator set. A first real-time indicator parameter set for a first monitoring category is obtained; the first real-time indicator parameter set is used as input data of an integrated prediction model to obtain first output data, wherein the first output data refers to a first warning index for the first monitoring category; when the first warning index reaches a first warning index limit, the target warning device is activated to issue an abnormal warning for the first monitoring category of the target substation. That is to say, by constructing a three-dimensional model and real-time sensor monitoring, combined with preprocessing analysis and an integrated prediction model, dynamic monitoring and early warning of the construction process of the target substation are achieved, so as to improve the construction safety and construction management efficiency of the substation.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A flow chart of a substation construction monitoring and early warning method based on a three-dimensional model in this application;

[0013] Figure 2 This is a structural schematic diagram of a substation construction monitoring and early warning device based on a three-dimensional model in this application.

[0014] Description of reference numerals:

[0015] Model building module 11, dynamic monitoring module 12, preprocessing module 13, visualization module 14, data collection module 15, early warning prediction module 16, early warning execution module 17. DETAILED DESCRIPTION

[0016] This application solves the technical problem that the construction monitoring and early warning analysis in the prior art lacks integration and intelligence, cannot effectively extract key information from a large amount of monitoring data, and thus cannot identify potential risks and provide effective early warnings in a timely manner, resulting in threats to the safety of substation construction by providing a substation construction monitoring and early warning method and device based on a three-dimensional model. By constructing a high-precision three-dimensional model, dynamic multi-dimensional sensor monitoring is achieved, and efficient data processing and analysis are performed through a digital preprocessing platform, and finally, a visual construction status display and real-time early warning are provided for the construction of the substation through a visualization model and an integrated prediction model, thereby improving the construction management level and safety stability of the substation.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] For example, please refer to the attached Figure 1 The present application provides a substation construction monitoring and early warning method based on a three-dimensional model, wherein the substation construction monitoring and early warning method based on a three-dimensional model is applied to a substation construction monitoring and early warning device based on a three-dimensional model, and the substation construction monitoring and early warning method based on a three-dimensional model specifically includes the following steps:

[0019] A target three-dimensional model is constructed, wherein the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing target point cloud data of a target substation based on a principle of random sampling consistency.

[0020] Further, it includes: taking the first point cloud sample set as input data of the intelligent registration and fusion model to obtain a first parameter estimation result of the first three-dimensional model, wherein the first point cloud sample set is any group of point cloud samples in the target point cloud data, wherein the intelligent registration and fusion model refers to a model parameter estimation model constructed based on the principle of random sampling consistency; obtaining a first point cloud verification set according to the first point cloud sample set and the target point cloud data; obtaining a first verification result of the first point cloud verification set on the first three-dimensional model, and determining the target three-dimensional model according to the first verification result.

[0021] Further, it includes: judging whether a first spatial distance from a first point cloud verification sample to the first three-dimensional model meets a spatial distance limit, the first point cloud verification sample refers to any point cloud sample in the first point cloud verification set; if so, adding the first point cloud verification sample to the first consistency point cloud set; judging whether the ratio of the number of point cloud samples in the first consistency point cloud set to the first point cloud verification set meets a preset ratio constraint, and obtaining the first verification result.

[0022] Further, it includes: according to the first verification result, if it is satisfied, taking the first three-dimensional model as the target three-dimensional model; if it is not satisfied, reading a predetermined re-evaluation verification mechanism, and based on the predetermined re-evaluation verification mechanism, performing verification analysis on the second three-dimensional model obtained by analyzing the second point cloud sample set by the intelligent registration fusion model to obtain the target three-dimensional model.

[0023] Specifically, the core goal is to build a target 3D model, especially point cloud data for the construction status of the target substation. The so-called point cloud data here refers to the spatial displacement information of each point in the substation obtained by 3D scanning and other technologies. Using the principle of random sampling consistency, these point cloud data are processed through an intelligent registration fusion model to generate an accurate 3D model. This 3D model is crucial for substation construction monitoring because the target 3D model can provide intuitive and accurate substation construction status information.

[0024] Specifically, first, a set of point cloud samples is selected from the target point cloud data as input, and this set of samples is selected arbitrarily. These point cloud samples are input into the intelligent registration fusion model constructed based on the principle of random sampling consistency. The model processes these data and outputs the parameter estimation results of the first three-dimensional model. It should be noted here that the intelligent registration fusion model is a core component that can process point cloud data and estimate the parameters of the three-dimensional model. Next, a first point cloud verification set is generated based on the selected point cloud samples and the entire target point cloud data. The function of this verification set is to verify the first three-dimensional model. The verification process includes determining whether the spatial distance from the point cloud verification sample to the three-dimensional model meets the preset spatial distance limit. If it meets, the point cloud verification sample is added to the first consistent point cloud set. Further, it is determined whether the ratio of the number of point cloud samples in the consistent point cloud set to the verification set meets the preset ratio constraint. This step is to ensure the accuracy of the three-dimensional model. Finally, according to the verification result, if the condition is met, that is, the ratio of the number of point cloud samples in the consistent point cloud set to the verification set meets the preset ratio constraint, then the first three-dimensional model is the target three-dimensional model. If it is not satisfied, it is necessary to read the predetermined re-evaluation verification mechanism, and perform verification analysis on the second three-dimensional model obtained by analyzing the second point cloud sample set by the intelligent registration and fusion model based on the mechanism until a target three-dimensional model that meets the requirements is obtained. The predetermined re-evaluation verification mechanism refers to extracting another set of samples different from the first point cloud sample set from the target point cloud data for registration and fusion, and obtaining the parameter estimation result of another three-dimensional model, and then verifying the model until its verification result meets the preset ratio constraint, and then using the corresponding model as the construction model of the target substation, that is, as the target three-dimensional model.

[0025] Through the precise processing of intelligent registration fusion model and point cloud data, the accurate construction of the target substation 3D model is achieved, the efficiency and accuracy of model construction are improved, and a reliable foundation is provided for the subsequent construction monitoring, analysis and management of the substation. This method can more effectively process complex point cloud data, ensuring that the final 3D model can accurately reflect the actual construction situation of the substation, thereby supporting more in-depth technical analysis and decision-making.

[0026] Dynamic multi-dimensional sensor monitoring is performed on the target substation through a sensor monitoring equipment group to obtain target sensor information.

[0027] Further, it includes: the sensor monitoring equipment group includes multiple displacement sensors and multiple vibration sensors, the target displacement sensor information of the multiple displacement sensors is matched in the target sensor information, and the target displacement sensor information is preprocessed by the displacement processing middle station in the digital preprocessing middle station to obtain target preprocessed displacement sensor information; the target vibration sensor information of the multiple vibration sensors is matched in the target sensor information, and the target vibration sensor information is preprocessed by the vibration processing middle station in the digital preprocessing middle station to obtain target preprocessed vibration sensor information; the target preprocessed displacement sensor information and the target preprocessed vibration sensor information constitute the target sensor information.

[0028] Specifically, dynamic multi-dimensional sensor monitoring is carried out by deploying a sensor monitoring equipment group at the target substation to obtain comprehensive target sensor information. This sensor information reflects the real-time construction status of the substation, which is of great significance for ensuring the safe operation of the substation in the later stage and preventing potential failures.

[0029] Specifically, first, the sensor monitoring equipment group includes multiple displacement sensors and vibration sensors, which are used to monitor the dynamic changes of the substation. The displacement sensor is used to monitor the construction displacement changes of the substation structure or components, while the vibration sensor is used to monitor the vibration of each object in the substation during the construction process. The data collected by these sensors constitute the target sensing information. In other words, the construction process of the substation includes the construction of multiple objects. Therefore, when the substation is monitored for construction, it is actually the monitoring of the construction process of multiple objects, and each object is equipped with multiple displacement sensors and vibration sensors for simultaneous monitoring. Among them, the displacement sensor is used to monitor the construction progress of each object in the substation, and the vibration sensor is used to monitor the construction quality of each object in the substation. Then, in the target sensing information, the data collected by multiple displacement sensors and vibration sensors are matched respectively. In other words, the data of the displacement sensor is matched with the target displacement sensing information, and the target displacement sensing information is preprocessed by the displacement processing middle station in the digital preprocessing middle station to eliminate noise, correct data, etc. In addition, the most important thing is to verify and support the monitoring data of multiple similar sensors of each substation construction object, so as to obtain more accurate target preprocessing displacement sensing information. Similarly, the data of the vibration sensor is matched with the target vibration sensing information, and similar preprocessing is performed through the vibration processing middle station in the digital preprocessing middle station to obtain the target preprocessed vibration sensing information. Finally, the preprocessed displacement sensing information and vibration sensing information are combined to form the complete target sensing information. This information can be used for further analysis and processing to evaluate the health status of the substation, predict potential failures, or for other operation and maintenance decision support.

[0030] Through multi-dimensional sensor monitoring and effective data preprocessing, comprehensive status monitoring of the construction of the target substation is achieved. By accurately processing displacement and vibration sensor information, a deeper understanding of the substation construction status can be achieved, providing important support for the safe operation and intelligent management of the substation in the later stage. This construction monitoring and early warning method helps to detect abnormalities in advance, so as to take preventive measures and ensure the stability and reliability of power supply.

[0031] The digital preprocessing middle station is called to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information.

[0032] Further, it includes: extracting a first displacement sensor from the multiple displacement sensors, the first displacement sensor is used to monitor the displacement of a first construction object, and the first displacement sensor corresponds to first displacement sensing information; extracting a second displacement sensor from the multiple displacement sensors for the first construction object, and obtaining second displacement sensing information of the second displacement sensor; introducing a monitoring information support evaluation function to evaluate the support of the second displacement sensing information for the first displacement sensing information, and obtaining a support degree; judging whether the support degree reaches a support limit; if so, taking an average of the first displacement sensing information and the second displacement sensing information as the first target preprocessing displacement sensing information of the first construction object; and constructing the target preprocessing displacement sensing information based on the first target preprocessing displacement sensing information.

[0033] Furthermore, it also includes: if it is not reached, retrieving any historical pre-processed displacement sensor information of any historical moment of the first construction object; obtaining the time series of the historical pre-processed displacement sensor information based on the correspondence between the any historical moment and the any historical pre-processed displacement sensor information; performing regression fitting prediction analysis on the time series of the historical pre-processed displacement sensor information to obtain the first target predicted real-time displacement sensor information; and using the first target predicted real-time displacement sensor information as the first target pre-processed displacement sensor information of the first construction object.

[0034] Specifically, the target sensor information is preprocessed and analyzed through the digital preprocessing center to obtain more accurate target preprocessing sensor information. By analyzing this data, the displacement of the construction object can be monitored more effectively, which is crucial to ensure construction safety and improve monitoring accuracy.

[0035] Specifically, first, extract the first displacement sensor and the second displacement sensor for displacement monitoring of the first construction object from multiple displacement sensors, and obtain the respective first displacement sensing information and the second displacement sensing information. Then, introduce the monitoring information support evaluation function, evaluate the support degree of the second displacement sensing information for the first displacement sensing information, and obtain the support degree. Then, judge whether this support degree reaches the preset support limit. If the support degree reaches the limit, take the mean of the first displacement sensing information and the second displacement sensing information as the first target preprocessing displacement sensing information of the first construction object. Build the target preprocessing displacement sensing information based on this information. If the support degree does not reach the limit, it is necessary to retrieve any historical preprocessing displacement sensing information of any historical moment of the first construction object. Then, based on the correspondence between the historical moment and the historical preprocessing displacement sensing information, obtain the time series of the historical preprocessing displacement sensing information. Next, perform regression fitting prediction analysis on this time series to obtain the first target predicted real-time displacement sensing information. Finally, use this predicted real-time displacement sensing information as the first target preprocessing displacement sensing information of the first construction object. By introducing support evaluation and historical data prediction, accurate displacement monitoring results can be provided even when sensor data is not completely consistent. This is of great significance for ensuring construction safety, preventing accidents and improving construction efficiency.

[0036] The target preprocessed sensor information is rendered to the target three-dimensional model to obtain a target construction visualization model.

[0037] Specifically, combining the preprocessed sensor information with the target 3D model to generate a target construction visualization model helps to more intuitively understand the status of the construction object, which is of great value for construction monitoring, analysis and decision support. First, the preprocessed sensor information is obtained, which may include various types of data such as displacement, vibration, temperature, etc. Then, the sensor information is matched with the target 3D model. The target 3D model is a 3D representation of the construction object or scene, which may be created by 3D scanning or other modeling techniques. Next, the sensor information is rendered on the 3D model in a visual form. For example, displacement data may be represented by different colored marks on the model, and vibration data may be displayed by vibration animation of elements on the model. By combining sensor information with the 3D model, the construction process is visualized. This helps to more intuitively monitor and analyze the status of the construction object, detect anomalies in time, predict potential problems, and thus improve the safety and efficiency of construction. Through this visualization method, complex data can be presented in a more understandable form, providing a powerful decision support tool for engineers, project managers and other relevant personnel.

[0038] A first predetermined sensor indicator set is read, and data is collected on the target construction visualization model based on the first predetermined sensor indicator set to obtain a first real-time indicator parameter set of a first monitoring category.

[0039] Specifically, by reading a predetermined set of sensor indicators, data is collected on the target construction visualization model to obtain a real-time indicator parameter set for a specific monitoring category. First, the first predetermined sensor indicator set is read, which contains various indicators for evaluating and monitoring the performance of the construction object. These indicators may include displacement, vibration amplitude, temperature range, etc., depending on the needs and objectives of the monitoring. Then, based on these predetermined sensor indicators, data is collected on the target construction visualization model. This means extracting data related to these indicators from the visualization model. For example, if the indicator set includes temperature monitoring, the system will extract temperature-related data points from the model. Finally, through this process, the first real-time indicator parameter set of the first monitoring category is obtained. This parameter set contains real-time data collected according to the predetermined indicators, which provides important information for monitoring and analyzing the current state of the construction object. Among them, the first monitoring category may be construction progress or construction quality. Through the predetermined sensor indicator set, efficient data collection of the target construction visualization model is achieved. This method ensures the consistency of the collected data with the monitoring target and improves the relevance and availability of the data.

[0040] The first real-time indicator parameter set is used as input data of the integrated prediction model to obtain first output data, where the first output data refers to a first warning index of the first monitoring category.

[0041] Further, it includes: forming a first training data set for the first monitoring category based on big data, and performing supervised learning on the first training data set to obtain a first prediction layer; wherein the first training data set includes a first historical indicator parameter set and a first historical warning index corresponding to the first predetermined sensor indicator set; and building the integrated prediction model according to the first prediction layer.

[0042] Specifically, a training data set for the first monitoring category is constructed based on big data, and a prediction layer is obtained through supervised learning, and then an integrated prediction model is built. The model is trained using historical data and early warning indexes to achieve predictions of future states, which is of great significance for preventing and responding to potential problems.

[0043] First, a first training data set of the first monitoring category is constructed based on big data. This data set includes a first historical indicator parameter set and a first historical warning index corresponding to the first predetermined sensor indicator set. The historical indicator parameter set contains data collected based on the predetermined sensor indicators over a period of time in the past, while the historical warning index is a corresponding warning record. Then, supervised learning is performed on the first training data set. Supervised learning is a machine learning method that uses labeled training data to train a model so that it can classify or predict new data. In this scenario, the historical indicator parameter set and warning index of the training data set are used to train the model to identify indicator patterns that may lead to warnings. Then, the first prediction layer is obtained through this process. This prediction layer is a part of the model that can predict future warning indices based on the input real-time indicator parameter set. Finally, an integrated prediction model is built based on the first prediction layer. The integrated prediction model may combine the outputs of multiple prediction layers to improve the accuracy and reliability of the prediction. This model can be used to monitor and predict the future state of the construction object in real time, so as to detect potential problems in advance and take preventive measures.

[0044] Through big data analysis and machine learning technology, the future status of construction objects can be predicted. By building and training prediction models, historical data can be used more effectively to predict future warning situations, which is of great significance for improving construction safety and efficiency. This method can assist engineers and project managers to make more informed decisions and ensure the smooth progress of the construction process.

[0045] When the first warning index reaches a first warning index limit, the target warning device is activated to issue an abnormal warning for the first monitoring category of the target substation.

[0046] Specifically, when the first warning index reaches the warning index limit, the warning device is activated to warn the first monitoring category of the target substation for abnormalities. The key is to set a warning threshold and trigger the warning mechanism when this threshold is reached, which is of great significance for timely responding to potential safety risks and preventing accidents. First, monitor the first warning index, which is obtained based on the analysis of real-time data and historical data and is used to evaluate the status of the target monitoring category. Then, set a warning index limit, which is predetermined and used to determine when the warning should be triggered. When the first warning index reaches or exceeds this warning index limit, the system will activate the target warning device. This warning device may be an audible and visual alarm, a text message notification system, or other types of alarm devices, which are used to alert relevant personnel. By setting the warning threshold and activating the warning device, timely warning of potential abnormalities in the target substation is achieved, which can ensure that prompt action is taken when key indicators exceed the safety range, thereby reducing safety risks and ensuring the safe and stable operation of the substation. Through this warning mechanism, the monitoring efficiency of the substation status can be improved, providing important safety support for operation and maintenance personnel.

[0047] Furthermore, it also includes: training a second prediction layer for a second monitoring category; using the first prediction layer and the second prediction layer as initial learners; obtaining a meta-learner, wherein a predetermined coefficient of variation mechanism is embedded in the meta-learner; normalizing and weighting the first warning index of the first prediction layer and the second warning index of the second prediction layer according to the predetermined coefficient of variation mechanism to obtain a target warning index; when the target warning index reaches the second warning index limit, activating the target warning device to issue an abnormal warning to the target substation.

[0048] Specifically, by training the prediction layer of the second monitoring category, combined with the first prediction layer, using the meta-learner and the predetermined coefficient of variation mechanism, the target warning index is calculated, and the warning device is activated for abnormal warning when the index reaches the warning limit. The accuracy and reliability of the warning are improved by integrating and weighting multiple prediction layers. First, the second prediction layer of the second monitoring category is trained. This prediction layer is similar to the first prediction layer and is obtained through supervised learning based on the historical data and warning index of the second monitoring category. Then, the first prediction layer and the second prediction layer are used as initial learners. This means that the outputs of the two prediction layers are used as inputs for subsequent analysis. Next, a meta-learner is obtained, which is embedded with a predetermined coefficient of variation mechanism. The coefficient of variation is a statistic that measures the degree of discreteness of data. The predetermined coefficient of variation mechanism is used to normalize and weight the first warning index of the first prediction layer and the second warning index of the second prediction layer. The purpose of this is to unify the outputs of different prediction layers to a common scale for effective comparison and synthesis. Finally, the first warning index and the second warning index are normalized and weighted according to the predetermined coefficient of variation mechanism to obtain the target warning index. When this target warning index reaches the second warning index limit, the target warning device is activated to issue an abnormal warning to the target substation.

[0049] By integrating multiple prediction layers and using meta-learners, the complexity and accuracy of the early warning system are improved. By normalizing and weighting the outputs of different prediction layers, potential safety risks can be more comprehensively assessed and early warnings can be issued in a timely manner when necessary. This approach can enhance the monitoring capabilities of substation status, provide stronger safety support for operation and maintenance personnel, and help ensure the safe and stable operation of substations.

[0050] In summary, the substation construction monitoring and early warning method based on the three-dimensional model provided in this application has the following technical effects:

[0051] By constructing a target three-dimensional model, the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing the target point cloud data of the target substation based on the principle of random sampling consistency; performing dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; calling a digital preprocessing middle station to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information; rendering the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; reading a first predetermined sensor indicator set, and performing data collection on the target construction visualization model based on the first predetermined sensor indicator set. A first real-time indicator parameter set for a first monitoring category is obtained; the first real-time indicator parameter set is used as input data of an integrated prediction model to obtain first output data, wherein the first output data refers to a first warning index for the first monitoring category; when the first warning index reaches a first warning index limit, the target warning device is activated to issue an abnormal warning for the first monitoring category of the target substation. That is to say, by constructing a three-dimensional model and real-time sensor monitoring, combined with preprocessing analysis and an integrated prediction model, dynamic monitoring and early warning of the construction process of the target substation are achieved, so as to improve the construction safety and construction management efficiency of the substation.

[0052] Embodiment 2, based on the same inventive concept as the substation construction monitoring and early warning method based on the three-dimensional model in the above embodiment, the present application also provides a substation construction monitoring and early warning device based on the three-dimensional model, please refer to the attached Figure 2 The substation construction monitoring and early warning device based on the three-dimensional model includes:

[0053] The model building module 11 is used to build a target three-dimensional model. The target three-dimensional model refers to a three-dimensional model obtained by registering and fusing target point cloud data of a target substation based on the principle of random sampling consistency.

[0054] The dynamic monitoring module 12 is used to perform dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information.

[0055] The preprocessing module 13 is used to call the digital preprocessing center to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information.

[0056] The visualization module 14 is used to render the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model.

[0057] The data collection module 15 is used to read a first predetermined sensor indicator set, and collect data on the target construction visualization model based on the first predetermined sensor indicator set to obtain a first real-time indicator parameter set of a first monitoring category.

[0058] The early warning prediction module 16 is used to use the first real-time indicator parameter set as input data of the integrated prediction model to obtain first output data, and the first output data refers to the first early warning index of the first monitoring category.

[0059] The early warning execution module 17 is used to activate the target early warning device to issue an abnormal early warning for the first monitoring category of the target substation when the first early warning index reaches a first early warning index limit.

[0060] Furthermore, the model construction module 11 in the substation construction monitoring and early warning device based on the three-dimensional model is also used to: use the first point cloud sample set as input data of the intelligent registration and fusion model to obtain a first parameter estimation result of the first three-dimensional model, wherein the first point cloud sample set is any group of point cloud samples in the target point cloud data, wherein the intelligent registration and fusion model refers to a model parameter estimation model constructed based on the principle of random sampling consistency; obtain a first point cloud verification set based on the first point cloud sample set and the target point cloud data; obtain a first verification result of the first point cloud verification set for the first three-dimensional model, and determine the target three-dimensional model based on the first verification result.

[0061] Furthermore, the model building module 11 in the substation construction monitoring and early warning device based on a three-dimensional model is also used to: determine whether the first spatial distance from the first point cloud verification sample to the first three-dimensional model meets the spatial distance limit, the first point cloud verification sample refers to any point cloud sample in the first point cloud verification set; if so, add the first point cloud verification sample to the first consistency point cloud set; determine whether the ratio of the number of point cloud samples in the first consistency point cloud set to the first point cloud verification set meets the preset ratio constraint, and obtain the first verification result.

[0062] Furthermore, the model building module 11 in the three-dimensional model-based substation construction monitoring and early warning device is also used to: according to the first verification result, if it is satisfied, use the first three-dimensional model as the target three-dimensional model; if it is not satisfied, read the predetermined re-evaluation verification mechanism, and based on the predetermined re-evaluation verification mechanism, perform verification analysis on the second three-dimensional model obtained by analyzing the second point cloud sample set by the intelligent registration fusion model to obtain the target three-dimensional model.

[0063] Furthermore, the preprocessing module 13 in the substation construction monitoring and early warning device based on the three-dimensional model is also used for: the sensor monitoring equipment group includes multiple displacement sensors and multiple vibration sensors, the target displacement sensor information of the multiple displacement sensors is matched in the target sensor information, and the target displacement sensor information is preprocessed through the displacement processing middle station in the digital preprocessing middle station to obtain target preprocessed displacement sensor information; the target vibration sensor information of the multiple vibration sensors is matched in the target sensor information, and the target vibration sensor information is preprocessed through the vibration processing middle station in the digital preprocessing middle station to obtain target preprocessed vibration sensor information; the target preprocessed displacement sensor information and the target preprocessed vibration sensor information constitute the target sensor information.

[0064] Furthermore, the preprocessing module 13 in the substation construction monitoring and early warning device based on the three-dimensional model is also used to: extract a first displacement sensor from the multiple displacement sensors, the first displacement sensor is used to perform displacement monitoring on a first construction object, and the first displacement sensor corresponds to first displacement sensing information; extract a second displacement sensor of the first construction object from the multiple displacement sensors, and obtain second displacement sensing information of the second displacement sensor; introduce a monitoring information support evaluation function to perform support evaluation of the second displacement sensing information for the first displacement sensing information to obtain a support degree; determine whether the support degree reaches a support limit; if so, take the average of the first displacement sensing information and the second displacement sensing information as the first target preprocessing displacement sensing information of the first construction object; and construct the target preprocessing displacement sensing information based on the first target preprocessing displacement sensing information.

[0065] Furthermore, the preprocessing module 13 in the substation construction monitoring and early warning device based on the three-dimensional model is also used for: if it is not reached, retrieving any historical preprocessing displacement sensor information of any historical moment of the first construction object; obtaining the time series of the historical preprocessing displacement sensor information based on the correspondence between the any historical moment and the any historical preprocessing displacement sensor information; performing regression fitting prediction analysis on the time series of the historical preprocessing displacement sensor information to obtain the first target predicted real-time displacement sensor information; and using the first target predicted real-time displacement sensor information as the first target preprocessing displacement sensor information of the first construction object.

[0066] Furthermore, the early warning prediction module 16 in the substation construction monitoring and early warning device based on the three-dimensional model is also used to: form a first training data set of the first monitoring category based on big data, and perform supervised learning on the first training data set to obtain a first prediction layer; wherein the first training data set includes a first historical indicator parameter set and a first historical early warning index corresponding to the first predetermined sensor indicator set; and build the integrated prediction model according to the first prediction layer.

[0067] Furthermore, the substation construction monitoring and early warning device based on the three-dimensional model also includes a second early warning execution module, wherein the second early warning execution module is used to: train the second prediction layer of the second monitoring category; use the first prediction layer and the second prediction layer as initial learners; obtain a meta-learner, in which a predetermined coefficient of variation mechanism is embedded; normalize and weight the first early warning index of the first prediction layer and the second early warning index of the second prediction layer according to the predetermined coefficient of variation mechanism to obtain a target early warning index; when the target early warning index reaches the second early warning index limit, activate the target early warning device to issue an abnormal early warning to the target substation.

[0068] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The substation construction monitoring and early warning method based on the three-dimensional model and the specific examples in the first embodiment are also applicable to the substation construction monitoring and early warning device based on the three-dimensional model in this embodiment. Through the above detailed description of the substation construction monitoring and early warning method based on the three-dimensional model, those skilled in the art can clearly know the substation construction monitoring and early warning device based on the three-dimensional model in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0069] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A substation construction monitoring and early warning method based on a three-dimensional model, characterized in that: include: Constructing a target three-dimensional model, wherein the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing target point cloud data of a target substation based on a principle of random sampling consistency; Performing dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; Calling a digital preprocessing middle station to preprocess and analyze the target sensor information to obtain target preprocessed sensor information; Rendering the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; Reading a first predetermined sensor indicator set, and collecting data on the target construction visualization model based on the first predetermined sensor indicator set to obtain a first real-time indicator parameter set of a first monitoring category; Using the first real-time indicator parameter set as input data of the integrated prediction model to obtain first output data, where the first output data refers to a first warning index of the first monitoring category; When the first warning index reaches a first warning index limit, activating a target warning device to issue an abnormal warning for the first monitoring category of the target substation; Building a first training data set for the first monitoring category based on big data, and performing supervised learning on the first training data set to obtain a first prediction layer; Wherein, the first training data set includes a first historical indicator parameter set and a first historical warning index corresponding to the first predetermined sensor indicator set; Building the integrated prediction model according to the first prediction layer; Wherein, the method comprises: Training a second prediction layer for a second monitoring category; Using the first prediction layer and the second prediction layer as initial learners; Obtaining a meta-learner, wherein the meta-learner has a predetermined coefficient of variation mechanism embedded therein; Normalizing and weighting the first warning index of the first prediction layer and the second warning index of the second prediction layer according to the predetermined coefficient of variation mechanism to obtain a target warning index; When the target early warning index reaches a second early warning index limit, the target early warning device is activated to issue an abnormality early warning to the target substation.

2. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 1 is characterized in that: include: Using a first point cloud sample set as input data of an intelligent registration and fusion model, obtaining a first parameter estimation result of a first three-dimensional model, wherein the first point cloud sample set is any set of point cloud samples in the target point cloud data, and wherein the intelligent registration and fusion model refers to a model parameter estimation model constructed based on a principle of random sampling consistency; Obtaining a first point cloud verification set according to the first point cloud sample set and the target point cloud data; A first verification result of the first point cloud verification set on the first three-dimensional model is obtained, and the target three-dimensional model is determined according to the first verification result.

3. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 2 is characterized in that: include: Determine whether a first spatial distance from a first point cloud verification sample to the first three-dimensional model meets a spatial distance limit, wherein the first point cloud verification sample refers to any point cloud sample in the first point cloud verification set; If it is consistent, adding the first point cloud verification sample to the first consistent point cloud set; It is determined whether a ratio of the number of point cloud samples in the first consistency point cloud set to the number of point cloud samples in the first point cloud verification set satisfies a preset ratio constraint, and the first verification result is obtained.

4. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 3 is characterized in that: According to the first verification result, if it is satisfied, the first three-dimensional model is used as the target three-dimensional model; if it is not satisfied, the predetermined re-evaluation verification mechanism is read, and the second three-dimensional model obtained by analyzing the second point cloud sample set by the intelligent registration fusion model is verified and analyzed based on the predetermined re-evaluation verification mechanism to obtain the target three-dimensional model.

5. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 1 is characterized in that: The sensor monitoring equipment group includes a plurality of displacement sensors and a plurality of vibration sensors, including: Matching the target displacement sensing information of the plurality of displacement sensors in the target sensing information, and preprocessing the target displacement sensing information through a displacement processing middle stage in the digital preprocessing middle stage to obtain target preprocessed displacement sensing information; Matching the target vibration sensor information of the multiple vibration sensors in the target sensor information, and preprocessing the target vibration sensor information through the vibration processing middle station in the digital preprocessing middle station to obtain target preprocessed vibration sensor information; The target preprocessed displacement sensing information and the target preprocessed vibration sensing information constitute the target sensing information.

6. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 5 is characterized in that: include: Extracting a first displacement sensor from the plurality of displacement sensors, where the first displacement sensor is used to monitor the displacement of a first construction object, and the first displacement sensor corresponds to first displacement sensing information; Extracting a second displacement sensor of the first construction object from the plurality of displacement sensors, and acquiring second displacement sensing information of the second displacement sensor; Introducing a monitoring information support evaluation function to evaluate the support of the second displacement sensing information for the first displacement sensing information to obtain a support degree; Determining whether the support reaches a support limit; If so, taking the average of the first displacement sensing information and the second displacement sensing information as the first target pre-processed displacement sensing information of the first construction object; The target preprocessed displacement sensing information is constructed based on the first target preprocessed displacement sensing information.

7. The substation construction monitoring and early warning method based on a three-dimensional model according to claim 6 is characterized in that: Also includes: If not, retrieve any historical pre-processed displacement sensing information of the first construction object at any historical moment; Obtaining a time sequence of the historical pre-processed displacement sensing information based on a correspondence between the arbitrary historical moment and the arbitrary historical pre-processed displacement sensing information; Performing regression fitting prediction analysis on the historical preprocessed displacement sensor information time series to obtain first target predicted real-time displacement sensor information; The first target predicted real-time displacement sensing information is used as the first target preprocessed displacement sensing information of the first construction object.

8. A substation construction monitoring and early warning device based on a three-dimensional model, characterized in that: The steps for implementing the substation construction monitoring and early warning method based on a three-dimensional model as described in any one of claims 1 to 7 include: A model building module, wherein the model building module is used to build a target three-dimensional model, wherein the target three-dimensional model refers to a three-dimensional model obtained by registering and fusing target point cloud data of a target substation based on a principle of random sampling consistency; A dynamic monitoring module, the dynamic monitoring module is used to perform dynamic multi-dimensional sensor monitoring on the target substation through a sensor monitoring equipment group to obtain target sensor information; A preprocessing module, the preprocessing module is used to call a digital preprocessing center to perform preprocessing analysis on the target sensor information to obtain target preprocessed sensor information; A visualization module, wherein the visualization module is used to render the target preprocessed sensor information to the target three-dimensional model to obtain a target construction visualization model; A data collection module, the data collection module is used to read a first predetermined sensor indicator set, and collect data on the target construction visualization model based on the first predetermined sensor indicator set to obtain a first real-time indicator parameter set of a first monitoring category; An early warning prediction module, the early warning prediction module is used to use the first real-time indicator parameter set as input data of an integrated prediction model to obtain first output data, wherein the first output data refers to a first early warning index of the first monitoring category; An early warning execution module is used to activate a target early warning device to issue an abnormal early warning for the first monitoring category of the target substation when the first early warning index reaches a first early warning index limit.

Images