Rapid positioning method for power line fault

Abstract

The invention relates to the technical field of power system operation and maintenance, in particular to a power line fault rapid positioning method. The method comprises the steps of data acquisition and preprocessing, feature image construction, convolutional neural network model construction and training, fault identification and positioning, and result verification and optimization. The method can improve the positioning precision, achieves the precise recognition and positioning of a power line fault through the powerful feature extraction capability of the convolutional neural network, and reduces the false alarm and missing report. The fault repair time is shortened, the fault position is quickly positioned, accurate navigation is provided for operation and maintenance personnel, the fault search and repair time is remarkably shortened, and the power supply reliability is improved; the intelligent operation and maintenance promotes the development of the operation and maintenance of the power system towards the intelligentization and automation direction, reduces the operation and maintenance cost, and improves the operation and maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and maintenance, and in particular to a method for quickly locating a power line fault. Background Art

[0002] Traditional power line fault location methods rely on manual inspections and simple instrument measurements, which are not only time-consuming and labor-intensive, but also have limited positioning accuracy. With the development of smart grid technology, higher requirements are placed on real-time monitoring and rapid response to power line faults. Although some methods based on signal processing and machine learning have been proposed, there are still problems such as low recognition accuracy and slow positioning speed when processing complex and changeable power line fault data. Summary of the invention

[0003] The technical problem to be solved by the present invention is: to overcome the shortcomings of the prior art and propose a method for quickly locating power line faults. The method utilizes the powerful feature extraction and pattern recognition capabilities of convolutional neural networks to achieve intelligent recognition and precise positioning of power line faults.

[0004] The technical solution of the present invention is:

[0005] A method for quickly locating a power line fault comprises the following steps:

[0006] Data collection and preprocessing: collect the operation data of power lines in real time and preprocess the data to improve the data quality;

[0007] Feature image construction: convert the preprocessed data into a two-dimensional feature image and use convolutional neural network to extract features;

[0008] Convolutional neural network model construction and training: Design a convolutional neural network model and train the model using historical fault data containing power line fault type and location labels;

[0009] Fault identification and location: The feature image to be detected is input into the trained convolutional neural network model, and the model outputs the fault type and fault location information;

[0010] Result verification and optimization: The fault location information output by the model is compared and verified with the actual on-site conditions, and the model is fine-tuned and optimized based on the verification results.

[0011] Preferably, the convolutional neural network model includes at least two convolutional layers, at least one pooling layer, at least one fully connected layer and an activation function.

[0012] Preferably, the data collection and preprocessing step also includes denoising the collected data to eliminate noise interference in the data.

[0013] Preferably, the feature image construction step also includes normalizing the preprocessed data to ensure that the data is at the same scale to facilitate subsequent feature extraction and model training.

[0014] Preferably, in the convolutional neural network model construction and training steps, the specific structure, number of layers, and parameters of the model are adjusted and optimized according to the actual application scenario and data.

[0015] Preferably, in the fault identification and location step, the fault location information output by the model is the fault area displayed in the form of an image or specific coordinate location information.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] Improve positioning accuracy: Utilize the powerful feature extraction capabilities of convolutional neural networks to accurately identify and locate power line faults and reduce false alarms and missed alarms.

[0018] Shorten fault repair time: quickly locate the fault location and provide accurate navigation for operation and maintenance personnel, significantly shorten the fault finding and repair time, and improve power supply reliability.

[0019] Intelligent operation and maintenance: Promote the development of power system operation and maintenance towards intelligence and automation, reduce operation and maintenance costs, and improve operation and maintenance efficiency.

[0020] Strong adaptability: The convolutional neural network model can adaptively process power line fault data of different scales and types, and has strong generalization ability. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention are described clearly and completely. 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 should fall within the scope of protection of the present invention.

[0022] Example

[0023] This embodiment provides a method for quickly locating a power line fault, comprising the following steps:

[0024] Data collection and preprocessing: Install smart sensors on power lines to collect real-time operating data such as current, voltage, and temperature. Preprocess the collected data, including denoising and normalization, to improve data quality and provide a reliable data foundation for subsequent feature image construction and model training.

[0025] Feature image construction: The preprocessed data is converted into a two-dimensional feature image so that feature extraction can be performed using a convolutional neural network. Feature images can include heat maps and spectrum maps of time series data, which can intuitively display the operating status of the power line and provide key information for subsequent fault identification.

[0026] Convolutional neural network model construction and training:

[0027] Design a convolutional neural network model that includes multiple convolutional layers, pooling layers, fully connected layers, and activation functions. The convolutional layer is used to extract local features from the input image; the pooling layer is used to reduce the dimension of the feature map and reduce the amount of calculation; the fully connected layer is used to map the extracted features to the fault type space; the activation function is used to introduce nonlinearity and improve the expressiveness of the model.

[0028] Data enhancement: In order to increase the generalization ability of the model, the training data is enhanced, such as rotation, scaling, flipping, etc., to generate more training samples.

[0029] Loss function selection: According to the requirements of the fault location task, select an appropriate loss function, such as the cross entropy loss function for classification tasks and the mean square error loss function for regression tasks.

[0030] Optimization algorithm: Use optimization algorithms such as stochastic gradient descent and Adam to train the model. By continuously iteratively updating the model parameters, the loss of the model on the training set is gradually reduced.

[0031] Early stopping method: To avoid overfitting of the model, use the early stopping method to monitor the performance of the model on the validation set and stop training when the loss on the validation set no longer decreases.

[0032] Model evaluation: After training, use the test set to evaluate the model and calculate indicators such as accuracy, recall, and F1 score to measure the performance of the model.

[0033] Fault identification and location: The feature image to be detected is input into the trained convolutional neural network model. The model automatically determines the fault type based on the features of the input image and outputs the fault location information. By analyzing the model output, the rapid identification and accurate location of power line faults can be achieved.

[0034] Result verification and optimization: The fault location information output by the model is compared and verified with the actual situation on site to evaluate the positioning accuracy of the model. Based on the verification results, the model is fine-tuned and optimized, such as adjusting the model structure, increasing training data, etc., to improve positioning accuracy.

[0035] Working principle: The working principle of the patent of this invention is to collect real-time operation data by installing intelligent sensors on the power lines, convert these data into two-dimensional feature images and input them into the constructed convolutional neural network model. After training, the model can intelligently identify and quickly locate power line faults, output fault type and location information, thereby realizing efficient and accurate diagnosis of power line faults.

[0036] Although the present invention has been described in detail by way of preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person of ordinary skill in the art may easily think of changes or substitutions within the technical scope disclosed by the present invention, and these shall be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for quickly locating a power line fault, characterized in that: The steps include: Data collection and preprocessing: collect the operation data of power lines in real time and preprocess the data to improve the data quality; Feature image construction: convert the preprocessed data into a two-dimensional feature image and use convolutional neural network to extract features; Convolutional neural network model construction and training: Design a convolutional neural network model and train the model using historical fault data containing power line fault type and location labels; Fault identification and location: The feature image to be detected is input into the trained convolutional neural network model, and the model outputs the fault type and fault location information; Result verification and optimization: The fault location information output by the model is compared and verified with the actual on-site conditions, and the model is fine-tuned and optimized based on the verification results.

2. The method for quickly locating a power line fault according to claim 1, characterized in that: The convolutional neural network model includes at least two convolutional layers, at least one pooling layer, at least one fully connected layer and an activation function.

3. The method for quickly locating a power line fault according to claim 1, characterized in that: The data collection and preprocessing step also includes denoising the collected data to eliminate noise interference in the data.

4. The method for quickly locating a power line fault according to claim 1, characterized in that: The feature image construction step also includes normalizing the preprocessed data to ensure that the data is on the same scale to facilitate subsequent feature extraction and model training.

5. The method for quickly locating a power line fault according to claim 1, characterized in that: In the convolutional neural network model construction and training steps, the specific structure, number of layers, and parameters of the model are adjusted and optimized according to the actual application scenario and data.

6. The method for quickly locating a power line fault according to claim 1, characterized in that: In the fault identification and location step, the fault location information output by the model is the fault area displayed in the form of an image or specific coordinate location information.