Safety monitoring method and system based on intelligent power transmission and transformation project

Abstract

According to the safety monitoring method and system based on the intelligent power transmission and transformation project provided by the embodiment of the invention, the basic power transmission and transformation project safety monitoring network is generated by combining large-scale public template project monitoring data with the monitoring data of the specific target project environment and utilizing neural network training, so that rich public data resources are effectively utilized, and the safety of the power transmission and transformation project is improved. And customized training is carried out for the target project, so that the generalization ability and accuracy of the monitoring model are improved. And further, performing security risk label prediction on the to-be-learned template engineering monitoring data, optimizing the target machine learning model according to the security risk label prediction, and finally generating a target power transmission and transformation engineering security monitoring network capable of accurately predicting the security risk characteristics in the target engineering monitoring data. Therefore, the intelligent level of power transmission and transformation project safety monitoring is remarkably improved, and the timeliness and accuracy of safety risk early warning are enhanced.

Description

Technical Field

[0001] The present application relates to the technical field of power transmission and transformation engineering, and more specifically, to a safety monitoring method and system based on smart power transmission and transformation engineering. Background Art

[0002] With the continuous advancement of smart grid construction, safety monitoring of power transmission and transformation projects has become an important part of ensuring the stable operation of the power grid. However, traditional safety monitoring methods mainly rely on manual inspections and empirical judgments, and have problems such as low monitoring efficiency and insufficient accuracy. Especially in the complex and changeable power transmission and transformation project environment, how to accurately and efficiently identify potential safety risks has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] In view of this, the purpose of this application is to provide a safety monitoring method and system based on smart power transmission and transformation projects.

[0004] In conjunction with the first aspect of the present application, a safety monitoring method based on a smart power transmission and transformation project is provided, which is applied to a safety monitoring system based on a smart power transmission and transformation project, and the method comprises:

[0005] Obtaining public template engineering monitoring data that is adapted to each power transmission and transformation engineering environment and carries public safety risk label data, and target template engineering monitoring data that is adapted to the target engineering environment and carries target safety risk label data, as well as to-be-labeled to-be-learned template engineering monitoring data under the target engineering environment, wherein the sample size of the public template engineering monitoring data is larger than the sample size of the target template engineering monitoring data;

[0006] Performing neural network training based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network;

[0007] Based on the basic power transmission and transformation engineering safety monitoring network, predict the safety risk label of the template engineering monitoring data to be learned, and generate template safety risk label data of the template engineering monitoring data to be learned;

[0008] Based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data;

[0009] Acquire target project monitoring data, load the target project monitoring data into the target power transmission and transformation project safety monitoring network, and generate safety risk label data of the predicted target safety risk characteristics in the target project monitoring data.

[0010] In a possible implementation of the first aspect, the neural network training is performed based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data, and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network, including:

[0011] Parsing first template engineering monitoring data and second template engineering monitoring data from the target template engineering monitoring data, and parsing first security risk label data of the first template engineering monitoring data and second security risk label data of the second template engineering monitoring data from the target security risk label data, wherein the sample size of the second template engineering monitoring data is larger than the sample size of the first template engineering monitoring data;

[0012] Based on the public template engineering monitoring data and the public safety risk label data, as well as the first template engineering monitoring data and the first safety risk label data, the initialized neural network is subjected to parameter learning, and the neural network that has completed parameter learning is used as a candidate power transmission and transformation engineering safety monitoring network for predicting engineering monitoring data;

[0013] Based on the second template project monitoring data and the second safety risk label data, the candidate power transmission and transformation project safety monitoring network is subjected to enhanced parameter learning to generate a basic power transmission and transformation project safety monitoring network after parameter learning.

[0014] In a possible implementation of the first aspect, the predicting of security risk labels of the template engineering monitoring data to be learned based on the basic power transmission and transformation engineering security monitoring network to generate template security risk label data of the template engineering monitoring data to be learned includes:

[0015] Loading the to-be-learned template engineering monitoring data into the basic power transmission and transformation engineering safety monitoring network, generating basic safety risk label data of the to-be-learned template engineering monitoring data predicted by the basic power transmission and transformation engineering safety monitoring network;

[0016] Adjusting the basic security risk label data according to the monitoring feature vector of the template engineering monitoring data to be learned;

[0017] The adjusted basic safety risk label data is used as template safety risk label data of the template engineering monitoring data to be learned.

[0018] In a possible implementation manner of the first aspect, before performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network, the method further includes:

[0019] Performing data enhancement on the template engineering monitoring data to be learned to generate enhanced template engineering monitoring data, wherein the data enhancement includes at least one of data segment clipping, time series reversal, and noise addition;

[0020] According to the template security risk label data and an enhancement strategy for data enhancement of the template engineering monitoring data to be learned, iterative template security risk label data for the enhanced template engineering monitoring data is determined.

[0021] In a possible implementation of the first aspect, the method of performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network includes:

[0022] Loading the enhanced template engineering monitoring data into the template engineering monitoring data to be learned before the enhancement process, and loading the iterative template safety risk label data into the predicted template safety risk label data to generate an enhanced training data sequence;

[0023] Using the monitoring data of each template project to be learned in the enhanced training data sequence and the corresponding template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation project safety monitoring network.

[0024] In a possible implementation of the first aspect, the target machine learning model is parameter-learned based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk features in the engineering monitoring data, including:

[0025] Loading the to-be-learned template engineering monitoring data into a target machine learning model, generating security risk prediction data of the to-be-learned template engineering monitoring data, wherein the security risk prediction data includes a security risk category and a probability value of the security risk category;

[0026] Optimizing the model parameter information of the target machine learning model based on the security risk prediction data and the template security risk label data to complete a round of parameter learning;

[0027] Adding the security risk category whose probability value is greater than a set value to the template security risk label data to generate extended predicted target security risk label data;

[0028] The target safety risk label data after expanded prediction is used as the template safety risk label data, and the step of loading the template engineering monitoring data to be learned into the target machine learning model to generate safety risk prediction data of the template engineering monitoring data to be learned is returned until the target machine learning model meets the model convergence requirements, and the target machine learning model after parameter learning is used as the target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data.

[0029] In a possible implementation of the first aspect, the target machine learning model is parameter-learned based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk features in the engineering monitoring data, including:

[0030] Based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the basic power transmission and transformation engineering safety monitoring network to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data.

[0031] In a possible implementation manner of the first aspect, after performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk features in the engineering monitoring data, the method includes:

[0032] According to the node information of the target security risk feature represented in the template security risk label data, the candidate risk data segment corresponding to the target security risk feature is parsed from the template engineering monitoring data to be learned;

[0033] Performing waveform transformation processing on the candidate risk data segment to generate the transformed candidate risk data segment;

[0034] A neural network training is performed based on the characteristic information of the target safety risk characteristics represented in the template safety risk label data and the transformed candidate risk data segment to generate a corresponding target power transmission and transformation project risk identification model for identifying the target safety risk characteristic information in the project monitoring data.

[0035] In combination with the second aspect of the present application, a safety monitoring system based on a smart power transmission and transformation project is provided, wherein the safety monitoring system based on a smart power transmission and transformation project comprises a machine-readable storage medium and a processor, wherein the machine-readable storage medium stores machine-executable instructions, and when the processor executes the machine-executable instructions, the safety monitoring system based on the smart power transmission and transformation project implements the aforementioned safety monitoring method based on the smart power transmission and transformation project.

[0036] In combination with the third aspect of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When the computer-executable instructions are executed, the aforementioned safety monitoring method based on smart power transmission and transformation projects is implemented.

[0037] In combination with any of the above aspects, the embodiment of the present application combines large-scale public template engineering monitoring data with monitoring data of a specific target engineering environment, and uses neural network training to generate a basic power transmission and transformation engineering safety monitoring network, which not only effectively utilizes abundant public data resources, but also performs customized training for the target project, thereby improving the generalization ability and accuracy of the monitoring model. Furthermore, by predicting the safety risk labels of the learning template engineering monitoring data and optimizing the target machine learning model accordingly, a target power transmission and transformation engineering safety monitoring network is finally generated that can accurately predict the safety risk characteristics in the target engineering monitoring data, thereby significantly improving the intelligent level of power transmission and transformation engineering safety monitoring and enhancing the timeliness and accuracy of safety risk warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained by combining these drawings without paying creative work.

[0039] Figure 1 A flow chart of a safety monitoring method based on a smart power transmission and transformation project provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or ends.

[0042] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0043] Figure 1 The flowchart of the safety monitoring method based on the smart power transmission and transformation project provided by the embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps in the safety monitoring method based on the smart power transmission and transformation project of this embodiment can be shared with each other based on actual needs, or some steps can be omitted or maintained. The details of the safety monitoring method based on the smart power transmission and transformation project include:

[0044] Step S110, obtain public template engineering monitoring data that are adapted to each power transmission and transformation engineering environment and carry public safety risk label data, and target template engineering monitoring data that are adapted to the target engineering environment and carry target safety risk label data, as well as template engineering monitoring data to be labeled and learned under the target engineering environment, wherein the sample size of the public template engineering monitoring data is larger than the sample size of the target template engineering monitoring data.

[0045] In this embodiment, in a large-scale power enterprise data center, the server power enterprise has numerous power transmission and transformation engineering projects in different regions, and each project has its own engineering environment.

[0046] For the public template project monitoring data, the server obtains data from the company's historical database and the public power project data sharing platform. These data come from power transmission and transformation projects in different regions and at different times, covering a variety of different environmental conditions, such as different climates (cold northern regions, hot and humid southern regions, dry and cold high-altitude areas, etc.), and different terrains (plains, mountainous areas, coastal areas, etc.). The public safety risk label data attached to these public template project monitoring data are labeled by professional engineers and safety experts based on relevant standards and actual accident cases. For example, in some mountainous power transmission and transformation projects, landslides may cause the foundation of the transmission line tower to be unstable, and this risk is marked on the corresponding monitoring data; in coastal areas, the risk label of equipment corrosion may be marked due to factors such as sea breeze corrosion. The sample size of these public template project monitoring data is very large, and may contain tens of thousands of relevant monitoring data points for power transmission and transformation projects.

[0047] The target template engineering monitoring data is targeted at a specific target engineering environment. For example, an enterprise is now building a power transmission and transformation project in a newly developed industrial park. There are some small factories near this industrial park that may generate electromagnetic interference. At the same time, the soil conditions in the park also have a certain impact on the stability of the tower base. The server obtains the target template engineering monitoring data from the preliminary survey and test data of this industrial park, and these data have been labeled with target security risk label data by the security team responsible for the project based on preliminary analysis and experience. For example, the risk of electromagnetic interference is labeled with a risk label that may affect the normal operation of equipment in the substation, and the soil conditions are labeled with a label for the risk of tower base settlement. The sample size of these target template engineering monitoring data is smaller than that of the public template engineering monitoring data, and may only have a few hundred data points.

[0048] In addition, in this target engineering environment, there is still monitoring data of the template engineering to be learned that needs to be labeled. These data are collected by some temporary monitoring equipment in the early stage of engineering construction. For example, sensors installed on the planned paths of several transmission lines selected in the industrial park collected monitoring data such as soil moisture and electromagnetic field strength. However, these data have not been labeled with safety risk labels, and the server needs to obtain these data for subsequent processing and model training.

[0049] Step S120, performing neural network training based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network.

[0050] In this embodiment, after acquiring the above data, the server starts to perform neural network training based on the public template project monitoring data, public safety risk label data, target template project monitoring data and target safety risk label data to generate a basic power transmission and transformation project safety monitoring network.

[0051] First, the server parses the first template engineering monitoring data and the second template engineering monitoring data from the target template engineering monitoring data, and parses the first security risk label data of the first template engineering monitoring data and the second security risk label data of the second template engineering monitoring data from the target security risk label data. For example, in a power transmission and transformation project in a new industrial park where the target engineering environment is a new industrial park, the server uses some of the data collected in the early stage with fewer data and more typical features as the first template engineering monitoring data. These data may be collected at several specific locations in the park, and the first security risk label data marks the risks such as small electromagnetic interference initially found at these locations. The data with a relatively large amount of data and more comprehensive information is used as the second template engineering monitoring data. The second security risk label data marks risks such as the comprehensive impact of soil conditions on the tower foundation and the comprehensive evaluation of electromagnetic interference at multiple locations, and the sample size of the second template engineering monitoring data is larger than the sample size of the first template engineering monitoring data.

[0052] Then, the server performs parameter learning on the initialized neural network based on the public template engineering monitoring data and the public safety risk label data, as well as the first template engineering monitoring data and the first safety risk label data. This initialized neural network may be a multi-layer perceptron network structure. The server inputs various environmental factor data (such as climate and terrain related data) and corresponding public safety risk label data (such as risk labels of natural disasters to power transmission and transformation equipment) in the public template engineering monitoring data, as well as the first template engineering monitoring data (such as the electromagnetic field strength of a specific location, etc.) and the first safety risk label data (such as the electromagnetic interference risk of the location) in the target project into the neural network. The neurons in the neural network perform calculations based on these input data and corresponding weights, and continuously adjust the weights through the back propagation algorithm to reduce the error between the prediction results and the actual label data. After a large number of iterative calculations, the neural network that has completed parameter learning is used as a candidate power transmission and transformation engineering safety monitoring network for predicting engineering monitoring data.

[0053] Next, the server conducts enhanced parameter learning on the candidate power transmission and transformation project safety monitoring network based on the second template project monitoring data and the second safety risk label data. For example, the second template project monitoring data containing more comprehensive information (such as comprehensive data such as electromagnetic field strength and soil moisture at multiple locations) and the second safety risk label data (such as a comprehensive risk assessment of electromagnetic interference and tower foundation stability for the entire park) are input into the candidate power transmission and transformation project safety monitoring network. The network calculates again based on these data and further adjusts parameters such as the weights of the neural network to better adapt to the complex situation in the target engineering environment. After this process, a basic power transmission and transformation project safety monitoring network is generated after parameter learning. This basic network can have a preliminary prediction capability for safety risks in power transmission and transformation projects based on multiple data sources.

[0054] Step S130 , predicting safety risk labels for the template engineering monitoring data to be learned based on the basic power transmission and transformation engineering safety monitoring network, and generating template safety risk label data for the template engineering monitoring data to be learned.

[0055] In this embodiment, after the basic power transmission and transformation project safety monitoring network is generated, the safety risk label prediction of the template project monitoring data to be learned is started based on the network to generate template safety risk label data of the template project monitoring data to be learned.

[0056] The server first loads the monitoring data of the template project to be learned into the basic power transmission and transformation project safety monitoring network. For example, the monitoring data of the template project to be learned is the electromagnetic field strength, soil moisture and other data collected at other locations in the new industrial park that have not been fully analyzed. After inputting these data into the basic power transmission and transformation project safety monitoring network, the network generates basic safety risk label data based on its internal neuron structure and the parameters that have been learned. Assume that the basic power transmission and transformation project safety monitoring network predicts that there may be a risk of equipment failure at a certain location due to excessive electromagnetic field intensity, and there may be a risk of tower foundation settlement due to changes in soil moisture. This is the basic safety risk label data.

[0057] Then, the server adjusts the basic safety risk label data based on the monitoring feature vector of the monitoring data of the template project to be learned. For example, although the electromagnetic field intensity in the monitoring data of the template project to be learned is high, the monitoring feature vector shows that some electromagnetic shielding facilities are under construction near this location, which may reduce the risk of equipment failure. The server adjusts the previously predicted equipment failure risk based on this monitoring feature vector to reduce its risk level. Similarly, for the tower foundation settlement risk corresponding to soil moisture changes, if the monitoring feature vector shows that the soil type at this location has good stability, the server will also adjust the risk level accordingly.

[0058] Finally, the adjusted basic safety risk label data is used as the template safety risk label data of the template engineering monitoring data to be learned. These template safety risk label data will serve as an important basis for subsequent model training and optimization.

[0059] Step S140, based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data.

[0060] In this step, the parameters of the target machine learning model are learned based on the template engineering monitoring data to be learned and the template safety risk label data, and a target power transmission and transformation engineering safety monitoring network is generated to predict the target safety risk characteristics in the engineering monitoring data.

[0061] Before starting parameter learning, the server performs data enhancement on the monitoring data of the template engineering to be learned. For example, for the electromagnetic field strength data in the monitoring data of the template engineering to be learned, the server performs data segment clipping and selects some data segments with large fluctuations and may be more representative; the time series data of soil moisture is reversed to increase the diversity of the data; some small noise can also be added to simulate the small errors in actual monitoring. Through these data enhancement operations, enhanced template engineering monitoring data is generated. At the same time, based on the template safety risk label data and the enhancement strategy for data enhancement of the template engineering monitoring data to be learned, the iterative template safety risk label data for the enhanced template engineering monitoring data is determined. For example, for the electromagnetic field strength data that has been clipped, the risk level in the corresponding iterative template safety risk label data is adjusted according to its risk label situation in the original data and the clipped features.

[0062] Next, the server loads the enhanced template engineering monitoring data into the template engineering monitoring data to be learned before the enhancement process, and loads the iterative template safety risk label data into the predicted template safety risk label data to generate an enhanced training data sequence. For example, the data such as electromagnetic field strength and soil moisture after data enhancement are combined with the original other unenhanced data, and the adjusted iterative template safety risk label data is matched with the previously predicted template safety risk label data to form a complete enhanced training data sequence.

[0063] Then, the server uses the template engineering monitoring data to be learned and the corresponding template safety risk label data in the enhanced training data sequence to perform parameter learning on the target machine learning model. This target machine learning model can be a support vector machine model. The server inputs each data point and the corresponding label in the data sequence into the support vector machine model. The model finds an optimal hyperplane to divide different risk categories based on the distribution of these data points in the feature space. In this process, by adjusting the parameters of the model (such as the parameters of the kernel function, etc.), the model's classification accuracy of the data is continuously improved. After a large number of iterative calculations, when the model reaches a certain convergence condition (such as the accuracy is no longer significantly improved, etc.), the target machine learning model after parameter learning is used as the target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data.

[0064] Another way is that the server loads the monitoring data of the template engineering to be learned into the target machine learning model, and generates the safety risk prediction data of the monitoring data of the template engineering to be learned, and the safety risk prediction data includes the safety risk category and the probability value of the safety risk category. For example, after the electromagnetic field strength and soil moisture in the monitoring data of the template engineering to be learned are input into the target machine learning model, the model predicts that there is a risk of equipment failure at this location with a probability value of 30%, and there is a risk of tower foundation settlement with a probability value of 20%, etc. Then, based on the safety risk prediction data and the template safety risk label data, the model parameter information of the target machine learning model is optimized to complete a round of parameter learning. For example, if the equipment failure risk level marked here in the template safety risk label data is high, and the probability value predicted by the model is low, the parameters of the model are adjusted to improve the prediction probability of the equipment failure risk. Next, the safety risk category with a probability value greater than the set value (such as 50%) is added to the template safety risk label data to generate the target safety risk label data after the extended prediction. For example, after a round of adjustment, the probability value of the equipment failure risk is increased to 55%, and this risk category is added to the target safety risk label data after the extended prediction. Then, the target safety risk label data after the expanded prediction is used as the template safety risk label data, and the step of loading the template engineering monitoring data to be learned into the target machine learning model to generate the safety risk prediction data of the template engineering monitoring data to be learned is returned until the target machine learning model meets the model convergence requirements, and the target machine learning model after parameter learning is used as the target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data.

[0065] Alternatively, the server performs parameter learning on the basic power transmission and transformation engineering safety monitoring network based on the template engineering monitoring data to be learned and the template safety risk label data, and generates a target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data. For example, the electromagnetic field intensity, soil moisture and other data in the template engineering monitoring data to be learned and the corresponding template safety risk label data are input into the basic power transmission and transformation engineering safety monitoring network. The network further adjusts its internal neuron weights and other parameters based on these new data and labels. After multiple rounds of iterative learning, when the network's prediction error for the data reaches a minimum or no longer decreases significantly, the target power transmission and transformation engineering safety monitoring network is obtained.

[0066] Step S150, acquiring target project monitoring data, loading the target project monitoring data into the target power transmission and transformation project safety monitoring network, and generating safety risk label data of the predicted target safety risk features in the target project monitoring data.

[0067] In this embodiment, after obtaining the target power transmission and transformation project safety monitoring network, the target project monitoring data is acquired, the target project monitoring data is loaded into the target power transmission and transformation project safety monitoring network, and safety risk label data of the target safety risk characteristics in the predicted target project monitoring data is generated.

[0068] For example, the target project monitoring data is real-time monitoring data during the construction of the power transmission and transformation project in the new industrial park, including the operating parameters of the equipment in the substation, the current and voltage data of the transmission line, the electromagnetic field strength of the surrounding environment, etc. The server loads this data into the target power transmission and transformation project safety monitoring network. The network analyzes this data based on the parameters and structures it has learned. Assuming that the operating parameters of a certain device in the target project monitoring data fluctuate abnormally, the target power transmission and transformation project safety monitoring network predicts that this device has a failure risk based on the previously learned rules, and gives the corresponding safety risk label data, marking the level of failure risk (such as high risk) and the possible types of failures involved (such as circuit short circuit risk). For the current and voltage data of the transmission line, if there is a large voltage fluctuation, the network may predict that there is a risk of increased transmission loss and give the corresponding safety risk label data. In this way, the server uses the target power transmission and transformation project safety monitoring network to analyze the target project monitoring data, providing accurate risk prediction and evaluation basis for the safety management of the power transmission and transformation project.

[0069] After generating the target power transmission and transformation project safety monitoring network, the server can also parse the candidate risk data segments corresponding to the target safety risk features from the template engineering monitoring data to be learned based on the node information of the target safety risk features represented in the template safety risk label data. For example, if the template safety risk label data marks the equipment failure risk corresponding to the abnormal electromagnetic field strength, the server finds the data segment with abnormal electromagnetic field strength from the template engineering monitoring data to be learned as the candidate risk data segment. Then, the candidate risk data segment is subjected to waveform transformation processing, such as Fourier transformation and other operations, to generate the transformed candidate risk data segment. Finally, neural network training is performed based on the feature information of the target safety risk features represented in the template safety risk label data and the transformed candidate risk data segment to generate the corresponding target power transmission and transformation project risk identification model for identifying the target safety risk feature information in the engineering monitoring data. This target power transmission and transformation project risk identification model can more accurately identify specific safety risk features.

[0070] Based on the above steps, the embodiment of the present application combines large-scale public template engineering monitoring data with monitoring data of specific target engineering environments, and uses neural network training to generate a basic power transmission and transformation engineering safety monitoring network. It not only effectively utilizes rich public data resources, but also performs customized training for target projects, thereby improving the generalization ability and accuracy of the monitoring model. Furthermore, by predicting the safety risk labels of the learning template engineering monitoring data and optimizing the target machine learning model accordingly, a target power transmission and transformation engineering safety monitoring network is finally generated that can accurately predict the safety risk characteristics in the target engineering monitoring data, thereby significantly improving the intelligent level of power transmission and transformation engineering safety monitoring and enhancing the timeliness and accuracy of safety risk warnings.

[0071] In a possible implementation, the neural network training is performed based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network, including:

[0072] The first template engineering monitoring data and the second template engineering monitoring data are parsed from the target template engineering monitoring data, and the first security risk label data of the first template engineering monitoring data and the second security risk label data of the second template engineering monitoring data are parsed from the target security risk label data, and the sample size of the second template engineering monitoring data is larger than the sample size of the first template engineering monitoring data.

[0073] Based on the public template engineering monitoring data and the public safety risk label data, as well as the first template engineering monitoring data and the first safety risk label data, parameter learning is performed on the initialized neural network, and the neural network that has completed parameter learning is used as a candidate power transmission and transformation engineering safety monitoring network for predicting engineering monitoring data.

[0074] Based on the second template project monitoring data and the second safety risk label data, the candidate power transmission and transformation project safety monitoring network is subjected to enhanced parameter learning to generate a basic power transmission and transformation project safety monitoring network after parameter learning.

[0075] In a possible implementation, the predicting of security risk labels of the template engineering monitoring data to be learned based on the basic power transmission and transformation engineering security monitoring network to generate template security risk label data of the template engineering monitoring data to be learned includes:

[0076] The template engineering monitoring data to be learned is loaded into the basic power transmission and transformation engineering safety monitoring network to generate basic safety risk label data of the template engineering monitoring data to be learned predicted by the basic power transmission and transformation engineering safety monitoring network.

[0077] The basic security risk label data is adjusted according to the monitoring feature vector of the template engineering monitoring data to be learned.

[0078] The adjusted basic safety risk label data is used as template safety risk label data of the template engineering monitoring data to be learned.

[0079] In a possible implementation manner, before performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network, the method further includes:

[0080] Data enhancement is performed on the template engineering monitoring data to be learned to generate enhanced template engineering monitoring data, wherein the data enhancement includes at least one of data segment clipping, time series reversal, and noise addition.

[0081] According to the template security risk label data and an enhancement strategy for data enhancement of the template engineering monitoring data to be learned, iterative template security risk label data for the enhanced template engineering monitoring data is determined.

[0082] In a possible implementation, the target machine learning model is parameter-learned based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network, including:

[0083] The enhanced template engineering monitoring data is loaded into the template engineering monitoring data to be learned before the enhancement process, and the iterative template safety risk label data is loaded into the predicted template safety risk label data to generate an enhanced training data sequence.

[0084] Using the monitoring data of each template project to be learned in the enhanced training data sequence and the corresponding template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation project safety monitoring network.

[0085] In a possible implementation, the target machine learning model is parameter-learned based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data, including:

[0086] The template engineering monitoring data to be learned is loaded into the target machine learning model to generate security risk prediction data of the template engineering monitoring data to be learned, wherein the security risk prediction data includes security risk categories and probability values ​​of the security risk categories.

[0087] Based on the security risk prediction data and the template security risk label data, the model parameter information of the target machine learning model is optimized to complete a round of parameter learning.

[0088] The security risk category whose probability value is greater than a set value is added to the template security risk label data to generate target security risk label data after extended prediction.

[0089] The target safety risk label data after expanded prediction is used as the template safety risk label data, and the step of loading the template engineering monitoring data to be learned into the target machine learning model to generate safety risk prediction data of the template engineering monitoring data to be learned is returned until the target machine learning model meets the model convergence requirements, and the target machine learning model after parameter learning is used as the target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data.

[0090] In a possible implementation, the target machine learning model is parameter-learned based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data, including:

[0091] Based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the basic power transmission and transformation engineering safety monitoring network to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data.

[0092] In a possible implementation, after performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk features in the engineering monitoring data, the method includes:

[0093] According to the node information of the target security risk feature represented in the template security risk label data, the candidate risk data segment corresponding to the target security risk feature is parsed from the template engineering monitoring data to be learned.

[0094] The candidate risk data segment is subjected to waveform transformation processing to generate the transformed candidate risk data segment.

[0095] A neural network training is performed based on the characteristic information of the target safety risk characteristics represented in the template safety risk label data and the transformed candidate risk data segment to generate a corresponding target power transmission and transformation project risk identification model for identifying the target safety risk characteristic information in the project monitoring data.

[0096] In the above embodiments, the safety monitoring system based on the smart power transmission and transformation project for executing the above method embodiments has at least one processor, a control module (chip set) coupled to at least one of the (at least one) processors, a memory coupled to the control module, a non-volatile memory (NVM) / storage device coupled to the control module, at least one load / output device coupled to the control module, and a network interface coupled to the control module.

[0097] The processor may include at least one single-core or multi-core processor, and the processor may include any combination of general-purpose processors or special-purpose processors (such as graphics processors, application processors, baseband processors, etc.). For some alternative implementations, the safety monitoring system based on the smart power transmission and transformation project can be used as the gateway described in the embodiments of the present application. Based on electronic equipment.

[0098] For some alternative embodiments, the safety monitoring system based on the smart power transmission and transformation project may include at least one computer-readable medium (e.g., a memory or NVM / storage device) having instructions and at least one processor integrated with the at least one computer-readable medium and configured to execute instructions to implement a module to perform the actions described in the present disclosure.

[0099] For one embodiment, the control module may include any suitable interface controller to provide any suitable interface to at least one of the processor(s) and / or any suitable device or component in communication with the control module.

[0100] The control module may include a memory controller module to provide an interface to the memory. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0101] The memory may be used, for example, to load and store data and / or instructions for a safety monitoring system based on a smart power transmission and transformation project. For one embodiment, the memory may include any appropriate volatile memory, for example, an appropriate DRAM.

[0102] For one embodiment, the control module may include at least one load-in / load-out controller to provide an interface to the NVM / storage device and the (at least one) load-in / load-out device.

[0103] For example, the NVM / storage device may be used to store data and / or instructions. The NVM / storage device may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable (at least one) non-volatile storage device (e.g., at least one hard disk drive (HDD), at least one compact disk (CD) drive, and / or at least one digital versatile disk (DVD) drive).

[0104] The NVM / storage device may include a storage resource that is physically part of a device on which the safety monitoring system based on the smart power transmission and transformation project is installed, or it may be accessible to the device without being part of the device. For example, the NVM / storage device may be accessed via (at least one) load / output device based on a network.

[0105] (At least one) load-in / output device may provide an interface for the safety monitoring system based on the smart power transmission and transformation project to communicate with any other appropriate device, and the load-in / output device may include a communication component, a pinyin component, a sensor component, etc. The network interface may provide an interface for the safety monitoring system based on the smart power transmission and transformation project to communicate based on at least one network, and the safety monitoring system based on the smart power transmission and transformation project may wirelessly communicate with at least one component of the wireless network based on any of the priors and / or protocols of at least one wireless network, for example, access a wireless network based on a communication prior.

[0106] For one embodiment, at least one of the (at least one) processor may be loaded together with the logic of at least one controller of the control module (e.g., a memory controller module). For one embodiment, at least one of the (at least one) processor may be loaded together with the logic of at least one controller of the control module to form a system level load. For one embodiment, at least one of the (at least one) processor may be fused on the same die with the logic of at least one controller of the control module. For one embodiment, at least one of the (at least one) processor may be fused on the same die with the logic of at least one controller of the control module to form a system on chip (SoC).

[0107] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for those skilled in the art, based on the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0108] An embodiment of the present invention discloses a computer-readable storage medium, which stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the safety monitoring method based on smart power transmission and transformation projects described in the aforementioned embodiment.

[0109] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the safety monitoring method based on smart power transmission and transformation projects described in the aforementioned embodiment.

[0110] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed over multiple modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Those of ordinary skill in the art may understand and implement the present invention without creative effort.

[0111] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, a magnetic disk storage, a magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0112] Finally, it should be noted that what is disclosed above is only a preferred embodiment of the present invention, which is only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A safety monitoring method based on smart power transmission and transformation engineering, characterized in that: The method comprises: Obtaining public template engineering monitoring data that is adapted to each power transmission and transformation engineering environment and carries public safety risk label data, and target template engineering monitoring data that is adapted to the target engineering environment and carries target safety risk label data, as well as to-be-labeled to-be-learned template engineering monitoring data under the target engineering environment, wherein the sample size of the public template engineering monitoring data is larger than the sample size of the target template engineering monitoring data; Performing neural network training based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network; Based on the basic power transmission and transformation engineering safety monitoring network, predict the safety risk label of the template engineering monitoring data to be learned, and generate template safety risk label data of the template engineering monitoring data to be learned; Based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data; Acquire target project monitoring data, load the target project monitoring data into the target power transmission and transformation project safety monitoring network, and generate safety risk label data of the predicted target safety risk characteristics in the target project monitoring data.

2. The safety monitoring method based on smart power transmission and transformation engineering according to claim 1 is characterized in that: The neural network training is performed based on the public template engineering monitoring data, the public safety risk label data, the target template engineering monitoring data and the target safety risk label data to generate a corresponding basic power transmission and transformation engineering safety monitoring network, including: Parsing first template engineering monitoring data and second template engineering monitoring data from the target template engineering monitoring data, and parsing first security risk label data of the first template engineering monitoring data and second security risk label data of the second template engineering monitoring data from the target security risk label data, wherein the sample size of the second template engineering monitoring data is larger than the sample size of the first template engineering monitoring data; Based on the public template engineering monitoring data and the public safety risk label data, as well as the first template engineering monitoring data and the first safety risk label data, the initialized neural network is subjected to parameter learning, and the neural network that has completed parameter learning is used as a candidate power transmission and transformation engineering safety monitoring network for predicting engineering monitoring data; Based on the second template project monitoring data and the second safety risk label data, the candidate power transmission and transformation project safety monitoring network is subjected to enhanced parameter learning to generate a basic power transmission and transformation project safety monitoring network after parameter learning.

3. The safety monitoring method based on intelligent power transmission and transformation engineering according to claim 1 is characterized in that: The method of predicting the safety risk label of the template engineering monitoring data to be learned based on the basic power transmission and transformation engineering safety monitoring network to generate template safety risk label data of the template engineering monitoring data to be learned includes: Loading the to-be-learned template engineering monitoring data into the basic power transmission and transformation engineering safety monitoring network, generating basic safety risk label data of the to-be-learned template engineering monitoring data predicted by the basic power transmission and transformation engineering safety monitoring network; Adjusting the basic security risk label data according to the monitoring feature vector of the template engineering monitoring data to be learned; The adjusted basic safety risk label data is used as template safety risk label data of the template engineering monitoring data to be learned.

4. The safety monitoring method based on intelligent power transmission and transformation engineering according to claim 1 is characterized in that: Before performing parameter learning on the target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate the target power transmission and transformation engineering safety monitoring network, the method further includes: Performing data enhancement on the template engineering monitoring data to be learned to generate enhanced template engineering monitoring data, wherein the data enhancement includes at least one of data segment clipping, time series reversal, and noise addition; According to the template security risk label data and an enhancement strategy for data enhancement of the template engineering monitoring data to be learned, iterative template security risk label data for the enhanced template engineering monitoring data is determined.

5. The safety monitoring method based on smart power transmission and transformation engineering according to claim 4 is characterized in that: The method of performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network includes: Loading the enhanced template engineering monitoring data into the template engineering monitoring data to be learned before the enhancement process, and loading the iterative template safety risk label data into the predicted template safety risk label data to generate an enhanced training data sequence; Using the monitoring data of each template project to be learned in the enhanced training data sequence and the corresponding template safety risk label data, parameter learning is performed on the target machine learning model to generate a target power transmission and transformation project safety monitoring network.

6. The safety monitoring method based on intelligent power transmission and transformation engineering according to claim 1 is characterized in that: The method of performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data includes: Loading the to-be-learned template engineering monitoring data into a target machine learning model, generating security risk prediction data of the to-be-learned template engineering monitoring data, wherein the security risk prediction data includes a security risk category and a probability value of the security risk category; Optimizing the model parameter information of the target machine learning model based on the security risk prediction data and the template security risk label data to complete a round of parameter learning; Adding the security risk category whose probability value is greater than a set value to the template security risk label data to generate target security risk label data after extended prediction; The target safety risk label data after expanded prediction is used as the template safety risk label data, and the step of loading the template engineering monitoring data to be learned into the target machine learning model to generate safety risk prediction data of the template engineering monitoring data to be learned is returned until the target machine learning model meets the model convergence requirements, and the target machine learning model after parameter learning is used as the target power transmission and transformation engineering safety monitoring network for predicting the target safety risk characteristics in the engineering monitoring data.

7. The safety monitoring method based on intelligent power transmission and transformation engineering according to claim 1 is characterized in that: The method of performing parameter learning on a target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data includes: Based on the template engineering monitoring data to be learned and the template safety risk label data, parameter learning is performed on the basic power transmission and transformation engineering safety monitoring network to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk characteristics in the engineering monitoring data.

8. The safety monitoring method based on intelligent power transmission and transformation engineering according to claim 1 is characterized in that: After performing parameter learning on the target machine learning model based on the template engineering monitoring data to be learned and the template safety risk label data to generate a target power transmission and transformation engineering safety monitoring network for predicting target safety risk features in the engineering monitoring data, the method includes: According to the node information of the target security risk feature represented in the template security risk label data, the candidate risk data segment corresponding to the target security risk feature is parsed from the template engineering monitoring data to be learned; Performing waveform transformation processing on the candidate risk data segment to generate the transformed candidate risk data segment; A neural network training is performed based on the characteristic information of the target safety risk characteristics represented in the template safety risk label data and the transformed candidate risk data segment to generate a corresponding target power transmission and transformation project risk identification model for identifying the target safety risk characteristic information in the project monitoring data.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by a computer, the safety monitoring method based on the smart power transmission and transformation project described in any one of claims 1-8 is implemented.

10. A safety monitoring system based on smart power transmission and transformation engineering, characterized in that: It includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by a computer, the safety monitoring method based on the smart power transmission and transformation project described in any one of claims 1 to 8 is implemented.

Images