Integrated line intelligent monitoring system and method

Abstract

The present invention discloses an integrated line intelligent monitoring system and method, which belongs to the field of smart grid technology, and specifically includes: by collecting current, temperature waveforms and multi-dimensional environmental parameters at the connection points of the transmission line, after preprocessing, using feature extraction to build a dynamic fault waveform feature library; using an AI enhanced algorithm to automatically determine the fault type, and building a temperature prediction model based on time series analysis and machine learning to achieve early warning of temperature anomalies; after the warning is triggered, using Beidou satellite navigation to accurately locate the fault point, and through telemetry short messages combined with adaptive retransmission, efficiently returning fault data to the control center; and continuously combining Beidou positioning and map services to build a monitoring line spatial data model, realize visual display of monitoring data, and improve the intelligence and efficiency of transmission line operation and maintenance.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart grid, and in particular to an integrated line intelligent monitoring system and method. Background Art

[0002] With the continuous advancement of smart grid construction, the demand for real-time monitoring of the operating status and health of power lines is increasing. The line monitoring devices in the existing technology generally have problems such as inaccurate fault point positioning, high manual inspection costs, low inspection efficiency, non-intuitive inspection results, and insufficient data analysis and processing capabilities. It is difficult to meet the needs of efficient and accurate monitoring of smart grids.

[0003] For example, the patent with publication number CN110632373A discloses an integrated low-voltage line intelligent monitoring device, including: an open measurement current sensor, an open protection current sensor, a voltage sensor, a temperature sensor, a power supply unit, an intelligent monitoring unit, and a communication unit; the measurement current signal collected by the open measurement current sensor is transmitted to the intelligent detection unit, the protection current signal collected by the open protection current sensor is transmitted to the intelligent detection unit, the voltage signal collected by the voltage sensor is transmitted to the intelligent detection unit, the line temperature signal collected by the temperature sensor is transmitted to the intelligent detection unit, the communication unit is connected to the intelligent monitoring unit, and the power supply unit provides working power for the intelligent monitoring unit and the communication unit. This technical solution solves the problem of miniaturized and flexible monitoring of low-voltage lines.

[0004] For example, the patent with publication number CN117907735A discloses an online intelligent monitoring system and method for power transmission lines, which includes: firstly, obtaining the current signals of each power transmission line branch within a predetermined time period to obtain a sequence of current signals, then uploading the sequence of current signals to the online intelligent monitoring platform for power transmission lines, then extracting the current waveform association features of the sequence of current signals to obtain a sequence of contextual current waveform feature vectors, and finally, identifying the transmission line branch fault features in the sequence of contextual current waveform feature vectors to determine the faulty branch. In this way, advanced data processing and analysis algorithms can be used to process and diagnose current signals without the need for manual time-consuming inspections.

[0005] The above existing technologies all have the following problems: CN110632373A mainly focuses on basic monitoring functions, such as current, voltage, temperature and power supply monitoring, lacks complex and multi-level fault diagnosis, predictive maintenance and data analysis capabilities like the latter, and is limited to simple processing of the collected data or basic alarm judgment, with limited flexibility and scalability; CN117907735A ignores the important factors that temperature and environmental parameters affect the state of the transmission line, and after identifying the fault branch, no specific fault location method is mentioned. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention proposes an integrated line intelligent monitoring system and method, which collects the current, temperature waveforms and multi-dimensional environmental parameters of the transmission line connection points, and constructs a dynamic fault waveform feature library by feature extraction after preprocessing; uses an AI enhanced algorithm to automatically determine the fault type, and constructs a temperature prediction model based on time series analysis and machine learning to achieve early warning of temperature anomalies; after the warning is triggered, the Beidou satellite navigation is used to accurately locate the fault point, and the fault data is efficiently transmitted back to the control center through telemetry short messages combined with adaptive retransmission; and then combines Beidou positioning and map services to construct a monitoring line spatial data model to achieve visual display of monitoring data, thereby improving the intelligence and efficiency of transmission line operation and maintenance.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The integrated line intelligent monitoring method comprises:

[0009] Step S1: collecting current and temperature waveform data and multi-dimensional environmental parameters of the connection point of the transmission line, and performing pre-processing to obtain multi-dimensional data of electric and temperature waveforms;

[0010] Step S2: extracting features from the multi-dimensional data of the electric temperature waveform and constructing a dynamic fault waveform feature library;

[0011] Step S3: Use the AI ​​enhanced fault analysis algorithm to perform preliminary automatic analysis on the input multi-dimensional data of the electrical temperature waveform, and determine the fault type by comparing the dynamic fault waveform feature library;

[0012] Step S4: Based on different types of fault waveform data, a temperature prediction model is constructed using time series analysis and machine learning methods to predict the line temperature change trend. Combined with the preset temperature threshold, an early warning mechanism for temperature anomalies is established. When the predicted temperature exceeds the temperature threshold, a warning signal is automatically triggered.

[0013] Step S5: After the temperature prediction model triggers the warning signal, the positioning module of the Beidou satellite navigation system is used to locate the fault and determine the location of the fault point. The Beidou telemetry short message function is used in combination with the adaptive retransmission mechanism to transmit the current and temperature waveform data, fault type, and fault location back to the control center;

[0014] Step S6: Based on the Beidou positioning function, the monitoring terminal is continuously positioned and monitored. The monitoring data is integrated with the geographic space information in combination with the map service to construct a spatial distribution data model of the monitoring line and visualize the monitoring data.

[0015] Specifically, the specific steps of constructing the dynamic fault waveform feature library in step S2 include:

[0016] S2.1: Acquire the multi-dimensional data of the electric temperature waveform, perform denoising and smoothing, and use an adaptive algorithm to dynamically divide the time series of the processed multi-dimensional data of the electric temperature waveform into N intervals according to the local changes and trends in the multi-dimensional data of the electric temperature waveform;

[0017] S2.2: Use the neural network model to perform convolution operations on the multi-dimensional data of the electric temperature waveform to automatically learn the spatial and temporal features. At the same time, use wavelet transform to extract multi-scale frequency features and form feature vectors through feature fusion methods;

[0018] S2.3: Set three fault types: porcelain bottle grounding, tree barrier grounding, and transformer damage. According to the mapping relationship between the feature vector and the fault type, mark the feature vector of each interval and mark the corresponding fault type;

[0019] S2.4: Store the marked feature vectors into a database to form a dynamic fault waveform feature library.

[0020] Specifically, the specific steps of the AI ​​enhanced fault analysis algorithm in step S3 include:

[0021] S3.1: Select the trained bidirectional deep network as the basic framework, use the dynamic fault waveform feature library to train the bidirectional deep network, and obtain the AI-based fault analysis model;

[0022] S3.2: Use a lightweight deployment strategy to integrate the AI-based fault analysis model into the monitoring system;

[0023] S3.3: When new multi-dimensional data of electrical temperature waveforms are input into the monitoring system, the fault diagnosis algorithm automatically extracts features and, combined with the attention mechanism, automatically performs fault diagnosis based on the AI-based fault diagnosis model;

[0024] S3.4: Compare the analysis results with the dynamic fault waveform feature library, and determine the most matching fault type based on the comparison results. For the fault type that cannot be determined by the model, introduce an expert system for secondary confirmation;

[0025] S3.5: Output the research and judgment results to the user interface and collect feedback from actual use;

[0026] S3.6: Evaluate the actual effect of the fault analysis algorithm, and iteratively optimize the fault analysis algorithm based on feedback and new fault waveform feature data.

[0027] Specifically, the specific steps of step S4 include:

[0028] S4.1: Formulate an integration strategy to integrate the prediction results of the time series analysis model and the machine learning model to form a comprehensive temperature prediction model. The formula of the integration strategy is:

[0029] ;

[0030] in, represents the prediction results of the comprehensive temperature prediction model, represents the prediction result of the time series analysis model, represents the prediction result of the machine learning model, Represents the weight of the time series analysis model;

[0031] S4.2: Input different types of fault waveform data into the comprehensive temperature prediction model to obtain the predicted temperature ,in, represents the predicted temperature value of the Mth fault waveform data, where M represents the number of fault waveform data;

[0032] S4.3: Setting temperature threshold ,like , the early warning signal is automatically triggered.

[0033] Specifically, the specific steps of constructing the monitoring line spatial distribution data model in step S6 include:

[0034] S6.1: Use the Beidou positioning system to continuously collect positioning data from the monitoring terminal and perform preprocessing;

[0035] S6.2: Map, match and integrate the preprocessed positioning data with the geospatial information to form a spatial data set.

[0036] S6.3: Based on the spatial distribution characteristics of the monitoring points, a line model is constructed to connect the continuous monitoring points into lines to form the spatial path of the monitoring line;

[0037] S6.4: Perform buffer analysis and overlay analysis on the constructed spatial distribution data model to obtain the spatial relationship between the monitoring line and the surrounding environment;

[0038] S6.5: Use geographic information system visualization tools to display the monitoring line spatial distribution data model and its analysis results in the form of graphics combined with visual elements.

[0039] Specifically, the multi-dimensional data features of the electric temperature waveform in S2.2 include: the peak value, valley value, average value, standard deviation, energy in a specific frequency band, and waveform duration of the waveform.

[0040] The integrated line intelligent monitoring system includes: data processing module, fault analysis module, temperature prediction module, fault location module, and space model building module;

[0041] The data processing module is used to collect current and temperature waveform data of the connection points of the transmission line and perform feature extraction;

[0042] The fault analysis module is used to identify and classify the multi-dimensional data of the electric temperature waveform using an AI model based on a deep learning method;

[0043] The temperature prediction module is used to construct a temperature prediction model using time series analysis and machine learning methods, predict the line temperature change trend, and establish an early warning mechanism for temperature anomalies;

[0044] The fault location module is used to locate the location of the fault using the Beidou satellite navigation system and transmit the fault information;

[0045] The spatial model building module is used to build a monitoring line spatial distribution data model and perform visual display.

[0046] Specifically, the data processing module includes: a data acquisition unit and a feature library construction unit;

[0047] The data acquisition unit is used to collect current and temperature waveform data from the connection point of the transmission line;

[0048] The feature library construction unit is used to extract features from the multi-dimensional data of the electric temperature waveform, and to organize and store the extracted fault features in a dynamic fault waveform feature library.

[0049] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of an integrated line intelligent monitoring method when executing the computer program.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. The present invention proposes an integrated line intelligent monitoring method, which integrates Beidou high-precision positioning technology, AI fault analysis algorithm and mobile intelligent monitoring service to realize real-time monitoring, intelligent analysis and rapid positioning of characteristic information of temperature and current of transmission lines, thereby improving the safe operation level and operation and maintenance efficiency of the power grid.

[0052] 2. The present invention proposes an integrated line intelligent monitoring method, which realizes preliminary automatic analysis of transmission line faults by collecting current, temperature waveform data and multi-dimensional environmental parameters at the connection points of the transmission lines, and combining AI-enhanced fault analysis algorithms and dynamic fault waveform feature libraries, thereby improving the accuracy and efficiency of fault detection, reducing the need for manual intervention, and accelerating the fault response speed; establishing an early warning mechanism to issue a warning signal in advance before the temperature abnormally rises to a dangerous level, thereby ensuring the safe and stable operation of the power grid and reducing the risk and losses of power outages caused by faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a step framework diagram of the integrated line intelligent monitoring method of the present invention;

[0054] Figure 2 This is a principle flow chart of the integrated line intelligent monitoring method of the present invention;

[0055] Figure 3 A flow chart for constructing a dynamic fault waveform feature library for the integrated line intelligent monitoring method of the present invention;

[0056] Figure 4 This is a flowchart of the AI ​​enhanced fault analysis algorithm implementation of the integrated line intelligent monitoring method of the present invention;

[0057] Figure 5 This is a structural diagram of the integrated line intelligent monitoring system of the present invention. DETAILED DESCRIPTION

[0058] Example 1

[0059] See also Figure 1-Figure 2 , an embodiment provided by the present invention: an integrated line intelligent monitoring method, comprising the following steps:

[0060] Step S1: collecting current and temperature waveform data and multi-dimensional environmental parameters of the connection point of the transmission line, and performing pre-processing to obtain multi-dimensional data of electric and temperature waveforms;

[0061] Furthermore, the specific steps of step S1 include:

[0062] (1) Data collection stage:

[0063] 1) Humidity, wind speed, and air pressure sensors are installed at the connection points of the transmission lines and in the surrounding areas to ensure the accuracy and safety of data collection;

[0064] 2) Set the sampling rate and data format parameters of the data acquisition device to ensure that waveform details and environmental parameter changes can be captured;

[0065] 3) Start the data acquisition system and begin recording current, temperature waveforms and environmental parameter data;

[0066] (2) Data preprocessing stage:

[0067] 1) Perform data cleaning, data calibration, and data conversion on the collected current and temperature waveform data and multi-dimensional environmental parameters;

[0068] 2) Extract characteristic values ​​from waveform data, such as mean, variance, peak value, frequency component, and statistical values ​​of environmental parameters;

[0069] (3) Data integration and storage:

[0070] 1) Integrate current and temperature waveform data with environmental parameter data into a multidimensional data set;

[0071] 2) Storing the preprocessed multidimensional dataset in a database;

[0072] Step S2: extracting features from the multi-dimensional data of the electric temperature waveform and constructing a dynamic fault waveform feature library;

[0073] Step S3: Use the AI ​​enhanced fault analysis algorithm to perform preliminary automatic analysis on the input multi-dimensional data of the electrical temperature waveform, and determine the fault type by comparing the dynamic fault waveform feature library;

[0074] Step S4: Based on different types of fault waveform data, a temperature prediction model is constructed using time series analysis and machine learning methods to predict the line temperature change trend. Combined with the preset temperature threshold, an early warning mechanism for temperature anomalies is established. When the predicted temperature exceeds the temperature threshold, a warning signal is automatically triggered.

[0075] Step S5: After the temperature prediction model triggers the warning signal, the positioning module of the Beidou satellite navigation system is used to locate the fault and determine the location of the fault point. The Beidou telemetry short message function is used in combination with the adaptive retransmission mechanism to transmit the current and temperature waveform data, fault type, and fault location back to the control center;

[0076] Furthermore, the specific steps of fault location include:

[0077] (1) Positioning module initialization: Ensure that the Beidou satellite navigation system positioning module works normally and is initialized;

[0078] (2) Receiving warning signals: When the temperature prediction model triggers a warning, the positioning function is activated;

[0079] (3) Obtaining location information: Obtain accurate latitude, longitude and altitude information through the Beidou satellite navigation system to locate the fault point.

[0080] At the same time, the process of transmitting data back to the control center is:

[0081] (1) Data encapsulation: encapsulate the current waveform, temperature waveform data, fault type, and fault location information into a data packet in a specific format, where the fault location refers to the longitude and latitude;

[0082] (2) Using Beidou telemetry short messages: Using the short message communication function of Beidou satellites to send encapsulated data packets;

[0083] (3) Adaptive retransmission mechanism:

[0084] 1) Detect transmission status: After sending, check whether a confirmation receipt or transmission failure signal is received;

[0085] 2) Retransmission strategy: If the transmission fails, resend the data packet according to the set retransmission strategy, such as retransmission at a fixed time interval or exponential backoff retransmission;

[0086] 3) Retransmission limit: Set the maximum number of retransmissions to prevent infinite loops.

[0087] Step S6: Based on the Beidou positioning function, the monitoring terminal is continuously positioned and monitored. Combined with the map service, the monitoring data is integrated with the geographic space information, a spatial distribution data model of the monitoring line is constructed, and the monitoring data is visualized. The visualization results include the precise location of the monitoring terminal, real-time temperature, and hot spot analysis.

[0088] Example 2

[0089] See also Figure 3-Figure 4 In this embodiment, the specific steps of constructing a dynamic fault waveform feature library in step S2 include:

[0090] S2.1: Acquire multidimensional data of electric temperature waveforms, perform denoising and smoothing processing, and use an adaptive algorithm to dynamically divide the processed multidimensional data of electric temperature waveforms into N intervals according to local changes and trends in the multidimensional data of electric temperature waveforms. The calculation formulas for denoising and smoothing processing are the prior art content in this field, and are not the inventive solution of the present application, and are not described in detail here.

[0091] S2.2: Use a neural network model to perform convolution operations on the multidimensional data of the electric temperature waveform, automatically learn spatial and temporal features, and at the same time, use wavelet transform to extract multi-scale frequency features, and form feature vectors through feature fusion methods, wherein the calculation formula of wavelet transform is the prior art content in this field, and is not an inventive solution of this application, and will not be repeated here;

[0092] Furthermore, the multi-dimensional data features of the electric temperature waveform in S2.2 include: peak value, valley value, average value, standard deviation, energy in a specific frequency band, and waveform duration of the waveform.

[0093] S2.3: Set three fault types: porcelain bottle grounding, tree barrier grounding, and transformer damage. According to the mapping relationship between the feature vector and the fault type, mark the feature vector of each interval and mark the corresponding fault type;

[0094] S2.4: Store the marked feature vectors into a database to form a dynamic fault waveform feature library.

[0095] The specific steps of the AI ​​enhanced fault analysis algorithm in step S3 include:

[0096] S3.1: Select the trained bidirectional deep loop network as the basic framework, use the dynamic fault waveform feature library to train the bidirectional deep loop network, and obtain an AI-based fault analysis model, wherein the training process of the model is the prior art content in this field, and is not the inventive solution of this application, and will not be described in detail here;

[0097] It should be noted that, in the present invention, the bidirectional recurrent deep network combines a long short-term memory network with a variant of the recurrent neural network, GRU.

[0098] Furthermore, the specific steps of S3.1 include:

[0099] (1) Obtain dynamic fault waveform feature library;

[0100] (2) Select the basic network: Bi-RDN is used as the core structure because it can capture the context information of time series data.

[0101] Input layer: accepts preprocessed fault waveform features as input;

[0102] The bidirectional recurrent layer selects LSTM and GRU: used to learn the bidirectional dependencies of the sequence;

[0103] Hidden layer: add fully connected layer and attention mechanism;

[0104] Output layer: Design the output layer according to the specific task, such as classification;

[0105] (3) Model training:

[0106] 1) Prepare training set and validation set: divide data from dynamic fault waveform feature library;

[0107] 2) Set training parameters: such as learning rate, batch size, number of iterations, and optimizer;

[0108] 3) Training process: including forward propagation, loss calculation, back propagation, and evaluation and verification;

[0109] (4) Model evaluation and optimization;

[0110] (5) Deployment and application.

[0111] S3.2: Use a lightweight deployment strategy to integrate the AI-based fault analysis model into the monitoring system;

[0112] S3.3: When new multi-dimensional data of electrical temperature waveforms are input into the monitoring system, the fault diagnosis algorithm automatically extracts features and, combined with the attention mechanism, automatically performs fault diagnosis based on the AI-based fault diagnosis model;

[0113] S3.4: Compare the analysis results with the dynamic fault waveform feature library, and determine the most matching fault type based on the comparison results. For the fault type that cannot be determined by the model, introduce an expert system for secondary confirmation;

[0114] S3.5: Output the research and judgment results to the user interface and collect feedback from actual use;

[0115] S3.6: Evaluate the actual effect of the fault analysis algorithm, and iteratively optimize the fault analysis algorithm based on feedback and new fault waveform feature data.

[0116] The specific steps of step S4 include:

[0117] S4.1: Formulate an integration strategy to integrate the prediction results of the time series analysis model and the machine learning model to form a comprehensive temperature prediction model. The formula of the integration strategy is:

[0118] ;

[0119] in, represents the prediction results of the comprehensive temperature prediction model, represents the prediction result of the time series analysis model, represents the prediction result of the machine learning model, Represents the weight of the time series analysis model;

[0120] S4.2: Input different types of fault waveform data into the comprehensive temperature prediction model to obtain the predicted temperature ,in, represents the predicted temperature value of the Mth fault waveform data, where M represents the number of fault waveform data;

[0121] S4.3: Setting temperature threshold ,like , the early warning signal is automatically triggered.

[0122] The specific steps of constructing the monitoring line spatial distribution data model in step S6 include:

[0123] S6.1: Use the Beidou positioning system to continuously collect positioning data from the monitoring terminal and perform preprocessing;

[0124] S6.2: Map, match and integrate the preprocessed positioning data with the geospatial information to form a spatial data set.

[0125] S6.3: Based on the spatial distribution characteristics of the monitoring points, a line model is constructed to connect the continuous monitoring points into lines to form the spatial path of the monitoring line;

[0126] S6.4: Perform buffer analysis and overlay analysis on the constructed spatial distribution data model to obtain the spatial relationship between the monitoring line and the surrounding environment;

[0127] S6.5: Use geographic information system visualization tools to display the monitoring line spatial distribution data model and its analysis results in the form of graphics combined with visual elements.

[0128] Example 3

[0129] See also Figure 5 Another embodiment provided by the present invention is an integrated line intelligent monitoring system, comprising:

[0130] Data processing module, fault analysis module, temperature prediction module, fault location module, and space model building module;

[0131] A data processing module is used to collect current and temperature waveform data at the connection points of the transmission line and perform feature extraction;

[0132] Fault analysis module, which uses an AI model based on deep learning methods to identify and classify multi-dimensional data of electrical temperature waveforms;

[0133] Temperature prediction module, which uses time series analysis and machine learning methods to build a temperature prediction model, predict line temperature change trends, and establish an early warning mechanism for temperature anomalies;

[0134] Fault location module, used to locate the location of the fault using the Beidou satellite navigation system and transmit fault information;

[0135] The spatial model building module is used to build a spatial distribution data model of the monitoring line and display it visually.

[0136] The data processing module includes: a data acquisition unit, a data preprocessing unit, and a feature library construction unit;

[0137] A data acquisition unit for collecting current and temperature waveform data from a connection point of a transmission line;

[0138] A data preprocessing unit, used for cleaning, standardizing and denoising the collected current and temperature waveform data;

[0139] The feature library construction unit is used to extract features from the multi-dimensional data of the electric temperature waveform, and to organize and store the extracted features in the dynamic fault waveform feature library.

[0140] The temperature prediction module includes: temperature prediction unit and early warning mechanism unit;

[0141] Temperature prediction unit, used to predict future temperature trends using time series analysis;

[0142] The early warning mechanism unit is used to automatically trigger an early warning signal when the predicted temperature exceeds the threshold value in combination with the preset temperature threshold value.

[0143] The fault location module includes: Beidou positioning unit and Beidou communication unit;

[0144] Beidou positioning unit, which is used to provide real-time centimeter-level and post-processing millimeter-level high-precision positioning services based on Beidou high-precision positioning module;

[0145] The Beidou communication unit is used to utilize the Beidou telemetry short message function to safely and quickly transmit current, temperature waveform data, fault type, and fault location information back to the control center.

[0146] The spatial model building module includes: positioning acquisition unit, monitoring data acquisition unit, data integration unit, and visualization unit;

[0147] The positioning acquisition unit is used to continuously acquire the positioning data of the monitoring terminal based on the Beidou high-precision positioning method;

[0148] A monitoring data acquisition unit, used to collect current and temperature monitoring data of the monitoring terminal;

[0149] Data integration unit, used to integrate monitoring data with geospatial information in conjunction with map services;

[0150] The visualization unit is used to realize the visualization display of monitoring data through spatial grid model algorithms and geospatial data visual analysis tools.

[0151] Example 4

[0152] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of an integrated line intelligent monitoring method when executing the computer program. For details, please refer to the above method embodiment, which will not be repeated here.

[0153] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0154] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0155] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. An integrated line intelligent monitoring method, characterized in that: include: Step S1: collecting current and temperature waveform data and multi-dimensional environmental parameters of the connection point of the transmission line, and performing pre-processing to obtain multi-dimensional data of electric and temperature waveforms; Step S2: extracting features from the multi-dimensional data of the electric temperature waveform and constructing a dynamic fault waveform feature library; Step S3: Use the AI ​​enhanced fault analysis algorithm to perform preliminary automatic analysis on the input multi-dimensional data of the electrical temperature waveform, and determine the fault type by comparing the dynamic fault waveform feature library; Step S4: Based on different types of fault waveform data, a temperature prediction model is constructed using time series analysis and machine learning methods to predict the line temperature change trend. Combined with the preset temperature threshold, an early warning mechanism for temperature anomalies is established. When the predicted temperature exceeds the temperature threshold, a warning signal is automatically triggered. Step S5: After the temperature prediction model triggers the warning signal, the positioning module of the Beidou satellite navigation system is used to locate the fault and determine the location of the fault point. The Beidou telemetry short message function is used in combination with the adaptive retransmission mechanism to transmit the current and temperature waveform data, fault type, and fault location back to the control center; Step S6: Based on the Beidou positioning function, the monitoring terminal is continuously positioned and monitored. The monitoring data is integrated with the geographic space information in combination with the map service, a spatial distribution data model of the monitoring line is constructed, and the monitoring data is visualized. The specific steps of constructing the dynamic fault waveform feature library in step S2 include: S2.1: Acquire the multi-dimensional data of the electric temperature waveform, perform denoising and smoothing, and use an adaptive algorithm to dynamically divide the time series of the processed multi-dimensional data of the electric temperature waveform into N intervals according to the local changes and trends in the multi-dimensional data of the electric temperature waveform; S2.2: Use the neural network model to perform convolution operations on the multi-dimensional data of the electric temperature waveform to automatically learn the spatial and temporal features. At the same time, use wavelet transform to extract multi-scale frequency features and form feature vectors through feature fusion methods; S2.3: Set three fault types: porcelain bottle grounding, tree barrier grounding, and transformer damage. According to the mapping relationship between the feature vector and the fault type, mark the feature vector of each interval and mark the corresponding fault type; S2.4: storing the marked feature vectors in a database to form a dynamic fault waveform feature library; The specific steps of step S4 include: S4.1: Formulate an integration strategy to integrate the prediction results of the time series analysis model and the machine learning model to form a comprehensive temperature prediction model. The formula of the integration strategy is: X toal =α×X time +(1-a)×X mot ; Among them, X toal represents the prediction result of the comprehensive temperature prediction model, X time represents the prediction result of the time series analysis model, X mot represents the prediction result of the machine learning model, and α represents the weight of the time series analysis model; S4.2: Input different types of fault waveform data into the comprehensive temperature prediction model to obtain the predicted temperature X = {x1,...,x M }, where x M represents the predicted temperature value of the Mth fault waveform data, where M represents the number of fault waveform data; S4.3: Set the temperature threshold h1, if x M ≥h1, the warning signal will be automatically triggered.

2. The integrated line intelligent monitoring method according to claim 1, characterized in that: The specific steps of the AI ​​enhanced fault analysis algorithm in step S3 include: S3.1: Select the trained bidirectional deep network as the basic framework, use the dynamic fault waveform feature library to train the bidirectional deep network, and obtain the AI-based fault analysis model; S3.2: Use a lightweight deployment strategy to integrate the AI-based fault analysis model into the monitoring system; S3.3: When new multi-dimensional data of electrical temperature waveforms are input into the monitoring system, the fault diagnosis algorithm automatically extracts features and, combined with the attention mechanism, automatically performs fault diagnosis based on the AI-based fault diagnosis model; S3.4: Compare the analysis results with the dynamic fault waveform feature library, and determine the most matching fault type based on the comparison results. For the fault type that cannot be determined by the model, introduce an expert system for secondary confirmation; S3.5: Output the research and judgment results to the user interface and collect feedback from actual use; S3.6: Evaluate the actual effect of the fault analysis algorithm, and iteratively optimize the fault analysis algorithm based on feedback and new fault waveform feature data.

3. The integrated line intelligent monitoring method according to claim 2, characterized in that: The specific steps of constructing the monitoring line spatial distribution data model in step S6 include: S6.1: Use the Beidou positioning system to continuously collect positioning data from the monitoring terminal and perform preprocessing; S6.2: Map, match and integrate the preprocessed positioning data with the geographic spatial information to form a spatial data set; S6.3: Based on the spatial distribution characteristics of the monitoring points, a line model is constructed to connect the continuous monitoring points into lines to form the spatial path of the monitoring line; S6.4: Perform buffer analysis and overlay analysis on the constructed spatial distribution data model to obtain the spatial relationship between the monitoring line and the surrounding environment; S6.5: Use geographic information system visualization tools to display the monitoring line spatial distribution data model and its analysis results in the form of graphics combined with visual elements.

4. The integrated line intelligent monitoring method according to claim 3, characterized in that: The multi-dimensional data features of the electric temperature waveform in S2.2 include: the peak value, valley value, average value, standard deviation, energy in a specific frequency band, and waveform duration of the waveform.

5. An integrated line intelligent monitoring system, which is used to implement the integrated line intelligent monitoring method according to any one of claims 1 to 4, characterized in that: include: Data processing module, fault analysis module, temperature prediction module, fault location module, and space model building module; The data processing module is used to collect current and temperature waveform data of the connection points of the transmission line and perform feature extraction; The fault analysis module is used to identify and classify the multi-dimensional data of the electric temperature waveform using an AI model based on a deep learning method; The temperature prediction module is used to construct a temperature prediction model using time series analysis and machine learning methods, predict the line temperature change trend, and establish an early warning mechanism for temperature anomalies; The fault location module is used to locate the location of the fault using the Beidou satellite navigation system and transmit the fault information; The spatial model building module is used to build a monitoring line spatial distribution data model and perform visual display.

6. The integrated line intelligent monitoring system according to claim 5, characterized in that: The data processing module includes: a data acquisition unit and a feature library construction unit; The data acquisition unit is used to collect current and temperature waveform data from the connection point of the transmission line; The feature library construction unit is used to extract features from the multi-dimensional data of the electric temperature waveform, and to organize and store the extracted fault features in a dynamic fault waveform feature library.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the integrated line intelligent monitoring method described in any one of claims 1 to 4 are implemented.

Images